Aviation Instruction

and
Training
Aviation Instruction
and
Training

Edited by
ROSS A. TELFER

Professor and Head, Department o f Aviation

University o f Newcastle
Australia
First published 1993 by Ashgate Publishing

Reissued 2018 by Routledge

2 Park Square, Milton Park, Abingdon, Oxon, OXl 4 4RN
711 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

Copyright © R. A. Telfer 1993

All rights reserved. No part of this book may be reprinted or reproduced or

utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.

Notice:
Product or corporate names may be trademarks or registered trademarks,
and are used only for identification and explanation without intent to
infringe.

Publisher's Note
The publisher has gone to great lengths to ensure the quality of this reprint
but points out that some imperfections in the original copies may be
apparent.

Disclaimer
The publisher has made every effort to trace copyright holders and
welcomes correspondence from those they have been unable to contact.

A Library of Congress record exists under LC control number: 93004434

ISBN 13: 978-1-138-61075-0 (hbk)

ISBN 13: 978-1-138-61077-4 (pbk)
ISBN 13: 978-0-429-46327-3 (ebk)
Contents

List of tables viii

List of figures X

List of contributors xiii

Acknowledgements XX

Introduction 1
R.A. Telfer

Section 1: Aptitude Testing, Selection and Licensing 9

1. Introduction to section 1: pilot aptitude testing, selection and

licensing 11
R.A. Telfer
2. Aptitude assessment in pilot selection 15
K. Alexander and G. Stead
3. Aptitude testing and selection in aviation 34
D. Bartram
4. The selection and training of air traffic controllers in Australia 52
G. Hannan andR. O'Hara
5. Computer-based testing in flight crew licensing 69
GJ.F. Hunt and L.M. Hunt
Section 2: Approaches to Pilot Training 89

6. Introduction to section 2: approaches to pilot training 91

R.A. Telfer
7. Human factors training in aviation 95
D.E. Maurino
8. Human factors in training 114
J. Bent
9. Pilots' approaches to learning 121
P. J. Moore and R.A. Telfer
10. Transfer for flight simulation 130
G. Lintern and G. McMillan
11. The technology of part-task training 163
D J. Allerton
12. The evaluation of training 188
K. C. Given

Section 3: Ab Initio Pilot Training and Instruction 203

13. Introduction to section 3: ab initio pilot training and instruction 205

R A . Telfer
14. What makes aviation instruction different? 209
R A . Telfer
15. Effective aviation instruction 219
R A . Telfer
16. Academic training for the ab initio pilots 237
B. Blyth
17. Preparing better flight instructors 253
C. Elshaw
18. Instructional design and curriculum development in aviation 271
H.H. Lehrer
19. Aviation medicine and flight instruction 291
L. Lambeth
20. Educating the aviation professional 309
W.E. Pippin
21. Non-technical training and education of ab initio student pilots
for KLM Royal Dutch Airlines 323
A. Droog

vi
Section 4: In-Service Pilot Training 335

22. Introduction to section 4: in-service pilot training 337

R.A. Telfer
23. Aviation training and instruction 343
R.E. Somerville
24. Conversion courses and students’ learning needs: a new
approach for Air New Zealand 356
B. Sinclair
25. Design, development and implementation of a CRM program 368
A. Mellor
26. Crew management in supplementary airlines 385
T. Wilson
27. Establishing a cockpit resource management training program 393
R.L. Wilkinson
28. Using a generic non-normal checklist to train crew
co-ordination skills 411
A.N. Johnston
29. A skills-based approach to training aeronautical decision-making 430
M. Wiggins and D. O'Hare

Author Index 477

Subject Index 482

vii
List of tables

2.1 Examples of aptitude and abilities items 22

2.2 Example of a behavioural profile 27
4.1 Course structure of the Diploma of Applied Science (ATS) 61
7.1 ATPL training course in Human Factors 101
10.1 Experimental design for flight transfer experiments 133
10.2 Flight transfer tests of motion in training 139
14.1 Commercial aviation instruction and public education 212
15.1 Effectiveness in the pre-instructional phase 221
15.2 Effectiveness in the instructional phase 224
15.3 The post instruction phase 226
15.4 Conditions for learning a skill 228
15.5 Helping students retain knowledge and skills 232
17.1 Most frequently quoted reasons for students failing basic
flying training 256
18.1 Taxonomy of domains and categories of learning 277
18.2 Contrasting test types and test characteristics 283
19.1 Effective performance time at altitude 297
21.1 The flight training syllabus: number of flying hours 325
21.2 Groundschool - subjects and hours 327
29.1 Factors identified as contributing to weather-related
crashes in Australia 434
29.2 Hazardous thought patterns and their respective antidotes 442
29.3 The DECIDE model of aeronautical decision-making 443
29.4 The four dimensions and their respective levels which pilots
used to rate 16 airports from the most preferable to the least
preferable during a deviation from planned flight 449
29.5 Weather seeking strategy errors observed during simulated
VFR cross-country flights 450
29.6 A summary of the differences which have been established
between experienced and inexperienced pilot performance 451
29.7 Summary of the behavioural characteristics compiled
from the three stages which comprise Anderson's (1983)
ACT* model of skill acquisition and Rasmussen's (1983)
S-R-K model of human behaviour 461

ix
List of figures

3.1 Prediction of operational flying training outcome for

88 Royal Navy pilots 45
5.1 Accomplished pilot abilities 71
5.2 Process-content interactions in test validity 73
5.3 Assumed versus actual content-processing levels in
commercial pilot licence examinations 76
5.4 A competency specification (performance + abilities
+ task context) 77
5.5 The generation of test items from competency specifications 78
5.6 The CALES system defined 80
5.7 Status page in CALES 84
5.8 Test certificate 84
5.9 Item report 85
7.1 Disciplines frequently involved in Human Factors activities 98
7.2 Domains for CRM training 104
7.3 LOFT simulation scenario 105
7.4 Human Factors: the training task and the organizational
context 108
10.1 The essentials of a transfer experiment to examine relative
effectiveness of different simulator configurations 131
10.2 Trials to criterion in transfer between instrument and contact
conditions 134
10.3 Transfer from no-motion, asynchronous-motion, and
delayed-motion conditions to an asynchronous motion
condition 141
10.4 Transfer from a high and a low to a medium frequency of
phugoid response 144
10.5 Transfer from a high and a low to a medium damping ratio 144
10.6 Transfer between variation in frequency of phugoid response
and damping ratio at a low gain 145
10.7 Transfer between variations in frequency of phugoid
response and damping ratio at a high gain 145
10.8 Transfer from high and low to low and high settings of
control gain 147
10.9 Transfer between variations in frequency of phugoid
response and damping ratio at a low gain with responsiveness
of the system equalized 147
10.10 Transfer from 1:3 or a 1:27 control-display ratio 149
10.11 Transfer from short to long and from long to short
control-display lags 150
10.12 Transfer between systems of different order 152
10.13 Percent transfer to a control with 6 lbs of frictional force
from higher and lower values of frictional force 154
12.1 The training cycle 190
12.2 Training versus job performance 193
12.3 Internal and external evaluation 194
14.1 Aviation instruction theory 210
16.1 Navigation 242
16.2 Example from the Ground Training syllabus 243
16.3 Schematic form of the syllabus 244
16.4 Airline course structure 246
16.5 Lecture brief 247
21.1 Flight and crew management are team leadership 329
21.2 Modules of non-technical training 330
23.1 The pilot as an ’Energy Manager' 346
23.2 B747-400 glass cockpit 350
25.1 Suggested camera positions 376
25.2 Suggested seating layout at a venue 380
26.1 Accidents versus experience 386
26.2 The PILOT problem solving model 389
27.1 DARTS diagram 400
27.2 Attitude diagram 401
27.3 Leadership grid diagram 402

xi
27.4 Leadership circles diagram 402
27.5 Assertiveness diagram 403
28.1 Sample engine fire drill 414
28.2 Sample air conditioning overheat drills 415
28.3 Aer Lingus generic non-normal checklist 418
28.4 Generic Engine Failure and Shutdown check 419
28.5 Generic Systems Malfunction check 420
28.6 Generic Hydraulic System Failure check 421
28.7 Generic Emergency Briefing: Cabin check 423
29.1 The inter-relationship between the structural components
which comprise the ACT* model of skill acquisition 453
29.2 The processes involved in the transition from the
interpretive stage to the tuning stage of skill acquistion 457

xii
List of contributors

Ken Alexander and Greg Stead are Aviation Psychologists employed by

SHL Asia Pacific, a specialist firm which designs and develops assessment
materials for the world of work. Within its specialist aviation division,
SHL Asia Pacific has devised selection tests which are widely used
internationally, for the selection of ab initio and experienced pilots for
general aviation, for domestic and international commercial operators and
for training institutions. They have between them over forty years of
experience in aviation assessment and selection.

David Allerton is the Professor of Avionics at the College of Aeronautics,

Cranfield Institute of Technology. His research interests cover flight
simulation, part-task training systems, displays, avionics, navigation
systems and flight modelling. From 1977-81 he was a Principal Engineer
with GEC Marconi.

Dave B artram is Professor and Head of the Psychology Department at Hull

University and a director of Newland Park Associated Limited (Chartered
Occupational Psychologists) and Bartdale Ltd (Human Factors specialists).
His work over the past ten years has focussed on the development of
computer-based assessment methods especially in the area of pilot selection.
He has published extensively in this area and has produced a range of
software including the MICROPAT tests and the NFER-NELSON
SCREENTEST system.

John Bent is a senior training captain and Flying Training Manager

(Policy) for Cathay Pacific Airways in Hong Kong. He is the co-designer
and manager of a new workshop for aircrew Air Force and in three airlines.
His training experience has included military flying instruction,
management of a flying school in Germany, and development of a new
multicultural pre-school and primary school in Hong Kong. The last half of
his sixteen years with Cathay Pacific has been in the check and training
department. He is now engaged in the development of future flight training
programs for the airline.

Bryan Blyth is the Chief Ground Instructor at the British Aerospace

Flying College in Scotland and has an interest in the development and use
of CBT in ab initio training. Following a long career with the UK CAA, he
joined British Aerospace 1987 as Chief Ground Instructor where he worked
as a member of a management team creating a new concept in ab initio
training. The team had responsibility for setting up the Ground Training
course and its infrastructure.

Ir Andr6 Droog from The Netherlands first graduated as an Aeronautical

Engineer and later also became a Psychologist, specializing in the
Psychology of Personality and Organizational Psychology. From 1973 until
1988 he was a psychologist-consultant in the field of pilot selection. In
1988 he took his current position of Psychologist with the KLM Flight
Academy. He trains and counsels ab initio trainees and instructors in the
field of Crew Cooperation, Cockpit Resource Management and Human
Factors.

Chris Elshaw is an Applied Psychologist working for the UK Ministry of

Defence. Before taking up his current post he worked for the Royal Air
Force where he was head of a research group conducting applied research
into aircrew training. He has lectured on training methods and been
involved in international collaborative research. In 1992 he joined the Army
Personnel Research Establishment at Famborough and heads a group
undertaking research to support selection and retention of personnel for the
British Army.

Ken Given is a Wing Commander and the Commanding Officer of the

Royal Australian Air Force's School of Management and Training
Technology at RAAF Base, Wagga, New South Wales, Australia. He has
been directly involved with a range of training jobs for this past seventeen
years which have included instmctional posts at the School of Radio and the
Officers Training School, training design posts at the RAAFs School of
Technical Training and the RAAFs major fighter base at Williamtown in

xiv
NSW, and occupational analysis positions with the USAF at Brooks AFB in
San Antonio, Texas and in Melbourne, Australia. He has also been a staff
officer, with specific responsibilities for training, in Headquarters Australian
Defence Force in Canberra.

Greg Hannan teaches psychology to teacher education and social work

students at the University of Tasmania, Launceston. Prior to this post, he
worked as a mathematics teacher and then school counsellor for the New
South Wales Department of Education. In recent years, he has combined his
interest in cognition and personnel selection to the selection of Air Traffic
Control students at the University's Centre for Air Traffic services. His
recent involvement in aviation training has involved teaching Human Factors
and training and learning concepts to Air Traffic Control students.

G raham Hunt is the founding Professor and Head of Massey University’s

School of Aviation. With a background in aviation and instructional
psychology, his research focus has been on the analysis and development of
human competency. As the Director of the New Zealand Civil Aviation
Authority - Massey University Human Resource Development in Aviation
Program (HURDA), he developed a new approach to specifying competency
requirements for flight crew licences.

Lynn Hunt is the Assistant Director for Academic and Instructional

Services at Massey University's School of Aviation. She is an Instructional
Psychologist specializing in the research and development of methods for
improving learning, especially in ab initio pilot training. She was a principal
researcher in the New Zealand Civil Aviation Authority - Massey University
Human Resource Development in Aviation Program (HURDA) where she
developed the CALES program for computer-based testing.

A. Neil Johnston joined the ab initio pilot training program of Aer Lingus,
the national airline of Ireland, directly from school. He is now a Boeing 737
Captain with Aer Lingus. He was the founding chairman of the Human
Performance Committee in IFALPA (the International Federation of Airline
Pilots' Associations) and is currently chairman of the Human Factors
Working Group in LATA (the International Air Transport Association). He
represents LATA on the Human Factors and Flight Safety Study Group at
ICAO (the International Civil Aviation Organization).

xv
Leonard Lambeth is a Squadron Leader in the RAAF Specialist Reserve.
His medical practice is in Psychiatry and Aviation Medicine and he is
Lecturer in Aviation Medicine at the University of Newcastle. Dr Lambeth
also lectures in a Human Factors course run by the University of Newcastle,
and is a qualified pilot. Dr Lambeth's major interests are in the application
of psychiatry to aviation, and the Human Factors role in aviation accidents.
His other major interest is in the role of aviation medical specialists in
general aviation, particularly their role in flying training.

Henry (Hank) Lehrer is Professor of Aeronautical Science at Embry

Riddle Aeronautical University (ERAU) in Daytona Beach, Florida where
he teaches both undergraduate and graduate courses. Dr Lehrer, an Airline
Transport Pilot and Certified Flight Instructor (Instrument and Multi-engine),
has over 30 years of teaching experience at the elementary, secondary, and
higher education level. Lehrer, a successful grant applicant and researcher,
has an extensive publication and presentation background and is the
founding editor of the Journal of Aviation/Aerospace Education and
Research (JAAER).

Gavan Lintern worked in Human Factors research at the Aeronautical

Research Laboratories (Australia) from 1971-74 and with US Navy Flight
Simulation Research in Florida, 1978-85. He returned to the University of
Illinois in 1985 and is now Associate Professor in the Institute of Aviation.
Lintem’s research on the design and use of simulators for flight instruction is
based on his theoretical interest in learning, transfer and retention of
complex perception-action skills. He emphasizes the interaction of
individuals with their environment.

Daniel M aurino is the Secretary of the ICAO Flight Safety and Human
Factors Group. After obtaining a degree in education, he joined Aerolineas
Argentinas, where he held several management positions, including that of
Training Manager for the airline. In 1988 he joined CAE Electronics in
Montreal, Canada, to participate in a flight simulator training research
program. In May 1989, he joined ICAO with the responsibility of
developing and implementing the organization's Human Factors program.

G rant McMillan is an Engineering Research Psychologist with the Human

Engineering Division of the Armstrong Laboratory at the Wright-Patterson
Airforce Base, Ohio. His 19 years of professional experience have
emphasized the analysis and modelling of human performance in man-

xvi
machine systems. During the past 13 years he has managed a research
program investigating visual and motion cueing and time delay issues in
flight simulator design. His research has used both ground-based and in­
flight simulators to evaluate simulator variables on pilot performance and
transfer of training.

Allan Mellor taught High School for two years before becoming a
professional pilot. He joined Trans Australian Airlines (now Australian
Airlines) in 1977 and left to join Cathay Pacific in 1988. He designed and
co-authored Cathay Pacific’s CRM and has advised a number of major
airlines on setting up their own CRM programs. He has flown F27, B727,
A300, B747-200, -300, -400 aircraft and currently an L1011 Captain.

Phillip Moore is Associate Professor and Head of the Department of

Education at the University of Newcastle, Australia. He is an educational
psychologist interested in the psychology of learning, studying and
individual differences.

Robert O 'H ara is a Senior Lecturer at the Centre for Air Traffic Services at
the University of Tasmania. He took up his appointment four years ago.
Prior to that time, he had 16 years experience in Air Traffic Services with
the Civil Aviation Authority. His special interest is in the education of Air
Traffic personnel.

David O 'H are is a Senior Lecturer in Psychology at the University of

Otago, Dunedin, New Zealand. He was appointed Lecturer in Psychology at
the University of Lancaster, UK in 1977. Since 1982, he has pursued
research interests in pilot decision-making and aircraft accident analysis. He
is the co-author (with Stan Roscoe) of Flightdeck Performance: The Human
Factor, published in 1990 by The Iowa State University Press. He holds
British and New Zealand PPL’s, and the FAI 'Silver C' certificate for gliding.

William Pippin is Professor and Chairman of the Department of Aviation at

The Ohio State University in Columbus, Ohio. He is a retired Navy Captain
and carrier aviator with 45 years of experience in aviation. He has held
executive and command positions in several areas of aviation including
aircraft maintenance, administration, operations, safety, and training.
Captain Pippin first joined the faculty of The Ohio State University in 1976
as Professor and Chairman of Naval Science. Professor Pippin accepted the
Chair of Aviation in July, 1990.

xvii
Bob Sinclair is a retired Air New Zealand Captain who has been involved
in various aspects of military, general aviation and airline pilot training for
thirty-four years. In the last five years of flying, he has been a senior check
and training captain and performance instructor. Dining this time, he has
studied towards a university Business Studies Diploma in Instructional
Systems and now works in this field for Air New Zealand’s Flight
Operations.

Roy Somerville started his career in aviation as an apprentice with the De

Havilland aircraft company. He joined the RAF in 1962 and flew mostly
single seat fighters. He graduated from the Central Flying School (RAF) in
1970 as a flying instructor. His last four years with the RAF were with the
Red Arrows. He joined civil aviation in 1975, the last twelve years being on
the B747. He has been involved with the 400 model from the beginning and
is currently employed as a Senior Training Captain with Cathay Pacific
Airways.

Ross Telfer is the foundation Professor and Head of the Department of

Aviation at the University of Newcastle, Australia, where Australia's first
University degree in aviation is offered. He is the author/co-author of seven
books, including The Psychology o f Flight Training (Iowa State University,
1988). He was principal researcher of the Australian study of pilot
judgement training, has worked with airlines on training methods, and
currently is examining how pilots leam and optimal ways of instructing
them.

M ark Wiggins is a postgraduate student in the Department of Psychology

at the University of Otago, Dunedin, New Zealand. He received his
BA(Hons) in Psychology from the University of New England, Australia in
1990. He has also been a Junior Research Fellow at the University of New
England, and has an Australian private pilot's licence. His research interests
include pilot training and aeronautical decision-making.

Dick Wilkinson flew with the Royal Australian Navy prior to joining
QANTAS where he was Manager, Base Training until mid-1992. He is now
enjoying line flying. Apart from his extensive aviation experience, Captain
Wilkinson is qualified in law and has a keen interest in pilot training. He is
the Chief Executive Officer of the Australian Civil Air Academy at
Cessnock Airport, which trains airline, commercial and University of
Newcastle student pilots.

xviii
Tony Wilson is an aviation safety consultant. Originally a teacher, he
learned to fly in 1951, and joined Trans-Australia Airlines (later Australian
Airlines) as a pilot in 1957. On his retirement from airline flying in 1989, he
was Administrator of Australian Airlines' innovative Aircrew Team
Management Program. As principal of Crew Dynamics, he currently
teaches, advises and writes on Human Factors for various airlines and other
organizations.

xix
Acknowledgements

The University of Newcastle provided study leave to enable this book to be

researched and written.
Key staff in a number of organizations provided an opportunity for
discussion and advice. These included: Captain Brian Wyness, Manager,
Flight Operations, Captain Bob Sinclair, Instructional Systems Co-ordinator
and F/O Jan Peterson, Aircrew Team Training, Air New Zealand·, Captain
John Bartlett, Manager, Flight Operations, Ansett New Zealand; Professor
Graham Hunt, Lyn Hunt and Roger Crossthwaite at Massey University’s
School o f Aviation, New Zealand; Captain John Bent, Manager, Training
(Policy), Cathay Pacific Airways; Dr Andre Droog, K L .M Training School,
Haren, Netherlands; Karsten Severin, Psychologist, and Wolfgang Siebinger,
Manager of Theoretical Instruction, Lufthansa German Airlines, Bremen; Dr
Beat Schar, Head, Flight Crew Recruiting and Training, Swissair; Professor
David Allerton, Head, Department of Avionics, and Dr Helen Muir, Head,
Department of Psychology, Cranfield Institute o f Technology; Associate
Professor Darryl Fisher, Head, and Greg Hannan, Psychologist, Department
of Education, and Bob O'Hara, Air Traffic Control Centre, University o f
Tasmania, Launceston; Dr Stan Roscoe, Illiana Aviation Sciences Ltd,
Areata, California; Dana D. Siewart, Director, Flight Operations, Timothy
Burke, Director of Communications, and George Hammond, Chairman,
Department of Aviation, Centre fo r Aerospace Sciences, University o f North
Dakota; William E. Pippin, Chairman and Dr Richard Jensen, Department o f
Aviation, The Ohio State University, Columbus, Ohio; Dr Henry Lehrer,
Aeronautical Science Department, Embry-Riddle Aeronautical University,
Daytona Beach, Florida.
 Sharon Field undertook the formidable task of co-ordinating contributions,
removing errors, and applying her ingenuity to the challenges of computer
technology.
Photographs were supplied by Cranfield Institute of Technology, KLM,
QANTAS, Royal Australian Airforce, SHL Asia Pacific, Swissair and the
University of Tasmania. These are acknowledged on the pages where they
appear.
At Ashgate Publishing, John Hindley and Suzanne Evins were instrumental
in enabling an idea to become a book.
Grant, Jacob, Karina, Martin and Sam are thanked for their very special
contributions.
Finally, thanks are due to the twenty eight authors who found the time
needed to produce their contributions. The deadline usually required the
juggling of priorities and always tested commitment. The results speak for
themselves.

Ross Telfer
Newcastle
Australia
March, 1993

xxi
Introduction
R.A. Telfer

In general and commercial aviation there is an emergent awareness of the

importance of human factors in general, and of instruction in training in
particular (see, for example ICAO, November, 1991). The converging
demands of technological change, growing needs for airline pilots at a time
when the pool of experienced applicants is diminishing, and rising
recognition of the importance of human factors to aviation safety are
straining the capacity of traditional training. This book is both a response,
and a further stimulus, to the new consciousness of the fundamental role of
instruction and training in aviation.
This book shows how professionals in the aviation industry and academic
researchers complement each other in their pursuit of more effective and
efficient training. Theory and practice each have a contribution to make:
the first providing the conceptualization and scientific testing of possible
solutions, and the latter contributing the depth of experience and expertise in
the workplace. Each needs the other.
Contributions to this book thus come from regulatory authorities, airlines,
universities, colleges, flying schools, the armed services, and private
practice or consultancies. They have their origins in Europe, Australia, New
Zealand, Asia, USA and Canada. Such a mix brings differences in approach
and argument. Rather than changing both by the substitution of a synthetic
style, these differences have been retained for they emphasize the common
ground. It is the song, not the singer, that is important.
Further, the book demonstrates that there is an emerging profession of
flight instruction. In that sense 'instruction' is a generic term encompassing
education, instruction and training. More accurately, 'education' denotes a
broad preparation, including non-vocational or liberal studies, usually
provided by universities. Instruction and training are linked specifically to
aviation. The first is usually provided in ground school or by textbooks as
Aviation Instruction and Training

preparation for licence tests in so-called 'theory' subjects, and the last is used
in reference to the development of practical flying skills and procedures.
A recognizable theoretical base for aviation instruction is becoming
clearer, courses of instructor preparation are in place, and are being
developed further as both initial and in-service courses. Management is
perceiving the contribution of ground and airborne instruction to the efficient
operation of aviation in a variety of contexts. However, this understanding
is slow in arriving, because of tensions which fog accurate perceptions of
progress.
Fortunately, the tensions are insufficient to overcome the extent of the
achievement, in international terms, of contemporary aviation training and
instruction. They are, however, retardants which the profession will need to
address in order to progress further.

The retardants

First, there is the difference between the approaches of the airlines and
armed services on the one extreme, and that of the aero clubs and flying
schools on the other. This is not simply a function of the fact that one deals
with professional airline transport or air force pilots and the other with
recreational and general aviation pilots. Nor is it a result of the different
demands made of air force pilots. It is more a matter of availability, demand,
and economic survival.
This difference is most apparent in the attrition rates at both extremes, if
attrition is seen as the difference between those who start and those who are
ultimately successful in training. The services and the airlines draw on an
undiminishing pool of those eager to gain optimized flight training virtually
without personal cost, and to receive in addition the bonus of a career with
the apparent potential of high social and economic status. Airlines can
afford to be uncompromising in their pursuit of high standards, usually at the
expense of a high attrition rate.
Because such losses are costly in budgetary and human terms, airlines and
the services have turned to specialists in selection to devise methods and
materials to aid the process of prediction of success. In Section 1 (Aptitude
Testing, Selection and Licensing) successful methods are discussed.
Alexander and Stead describe the approach of SHL Asia Pacific to its
international work in pilot selection. Professor Bartram then provides
insight into the application of computer-based assessment methods through
his work at Hull University and in private practice. The parallel situation in

2
 Introduction to Aviation Instruction and Training

the selection of air traffic controllers is then discussed by Hannan and

O'Hara from the University of Tasmania. Finally, the Hunts (Massey
University, New Zealand) show how to apply computers to the testing
involved in flight-crew licensing.
In contrast with airlines and services, flying schools are continually
scouting for trainees. To have a job, junior flight instructors may be
expected to find their own trainees for without student pilots, the flying
school is insolvent. Few applicants, if any, are considered unsuitable. The
manager and the instructor are both aware of the need to get the student up
to licence standards. Failure affects all three.
Second, there is the tension between standardization and
individualization. Since Bell and Lancaster derived the monitorial system
by applying the principles of the Industrial Revolution to the classroom,
instructional theory and practice has tended towards greater recognition of
individual differences in the learning process. Indeed, education is now
perceived as a process rather than as a simple transfer of subject matter
(Telfer, 1989). The emphasis has shifted from the instructor to the trainee.
Objectives are considered in terms of what the student is to achieve, rather
than what the instructor is to do. There are compelling arguments for
instruction to be trainee-centred.
Third, there is the tension between tradition and innovation. The basic
pattern of ab initio flight instruction is one which has been established over
the years. The patter, the sequence and the individual exercises are
essentially unchanged. Despite the introduction of sophisticated simulators,
computer assisted instruction and technological changes in training aircraft,
there have been few major changes in the essential approach to teaching
people to fly. That may well be because traditional methods worked in the
past and still work today.
On the other hand, where innovatory methods have been introduced by
using computer and audio-visual technology, there appears to have been
little thought about the role of the expert instructor. Largely, this is because
of the absence of evaluatory studies to point to means of refining the
process.
In Section 2 (Approaches to Pilot Training) some of die innovatory forces
are revealed. Maurino (ICAO) describes the influence of human factors on
training; Bent (Cathay) points to a new way of looking at flight training;
Moore and Telfer (University of Newcastle) describe the implications of
their research into how pilots learn. How can a flight simulator enhance skill
transfer? This is the question addressed by Lintem and McMillen
(University of Illinois); Allerton (Cranfield Institute of Technology) looks

3
Aviation Instruction and Training

at the role of part-task simulation, and Given (RAAF) points to the

importance of appropriate evaluation of training and instruction.
Fourthly, there is the tension between the pursuit of quality and the
restraint of budget. To an airline accountant, time off-line for training is
non-revenue producing. If not to be avoided, it is certainly to be constrained
as much as possible. Aviation trainers who have experienced the ways of
the bean-counters will find familiarity in a writer's analogy of the ongoing
search for funds. Cash, he said ,'... had to be rooted out by trained flunkies,
the way French farmers root out truffles with trained pigs' (Kurstler,1982).
Given the scale of expenditures in aviation, it is ironic to observe
resistance to relatively minor training costs. Resource management course
organizers have had to argue vigorously for residential programs to be sited
away from the company offices, yet the pay-off in operating hours and lack
of distraction is obvious. The duration of training workshops is a similar
matter of contention.
By saving relatively small amounts of money, the accountants put the
training itself at risk. The little extra time, especially when it is not
scheduled for a structured activity, may provide the vital opportunity for
reflection, consolidation and personal awareness.
Airline training courses in the human factors area are usually related,
basically, to changing attitudes. The company is hoping that the experience
of the course or workshop will induce voluntary changes in the ways
participants perform their jobs. Most advertising agencies would wilt at the
prospect of achieving that in a concentrated episode of a day or two. Pre-
reading, follow-up, management support, peer advocacy ... each addition
enhances the likelihood of success. Each paring away at time and resources
reduces the probability.
In fact, it can be demonstrated that training contributes to companies'
strategic goals. This contribution becomes more demonstrable when jobs
are reduced to performance skills which training can improve and extend.
This in turn affects productivity and profitability (McDonald, 1987). The
key activity for the trainer is to present this argument in terms of
performance, productivity and enhancement of the company’s operating
ethos.
How do airlines develop and introduce training programs for their air
crew? Based on insight derived from extensive experience, models and
hints for the design and development of airline resource management
programs are provided by Mellor (Cathay), Wilkinson (QANTAS) and
Wilson (Crew Dynamics) in Section 4 (In-Service Pilot Training). Each has
an unusual personal background (education/psychology or law) in addition

4
 Introduction to Aviation Instruction and Training

to airline command. In the same section, Somerville provides the viewpoint

of a highly experienced training captain in Cathay Pacific Airways, while
Wiggins and O'Hare (Otago University, New Zealand) provide a valuable
position paper on training for decision-making. Bridging the academic and
professional perspectives are Sinclair (Air New Zealand) and Johnston (Aer
Lingus), both captains linking extensive experience with academic
backgrounds.
Fifthly, there is tension between the role and responsibility of aviation
instructors and the recognition accorded them. In terms of parity of
esteem and salary the instructor is well down the aviation scale. The use of
flight instruction as a transient position to accumulate the hours needed to
progress to an airline position is a hackneyed strategy and an unfortunate
syndrome for the industry. The services’ tour-of-duty concept, with its
rationale that any skilled pilot is a good instructor, is another problem area
(Telfer and Biggs, 1988). Knowledge of content is, in itself, insufficient
(Boiko and Livingston, 1989).
As stated above, the knowledge-base for instruction is becoming much
clearer. Research on effective instruction has identified a '... substantial
body of reasonably consistent, hard-won knowledge about the associations
of teaching processes and student learning' (Waxman and Walberg, 1991).
There are also clear differences between highly experienced and novice
instructors. One of the main differences, directly affecting teaching
effectiveness, is the ability to use experience to unite previously unrelated
facts. If we want quality instruction, we need to attract highly-experienced
pilots.
In fields such as radiography where experts have examined between ten
and twenty thousand X-rays, they are able to process larger amounts of
problem data than a novice even though their memory capacity is identical.
Prior experience is the difference which will provide more competent flight
training in terms of the ability to have situational awareness, control of the
aircraft, and the ability to provide individualized instruction at the same time
(Badre, 1982).
Another factor is that of occupational socialization: the way trainees
develop values, attitudes and airmanship to enable them to maintain the
standards of the profession. Because of role modelling, the choice of
instructor is vital. Exposure to the standards of others is an influential factor
in instruction (Lintem and Roscoe, 1986), and one which the aviation
industry cannot afford to leave to chance. The potential influence of an
instructor is so great that it merits a career path and status to attract the
experienced pilot.

5
Aviation Instruction and Training

Finally, there is the tension between theory and practice. In an industry

in which theoretical developments underpin the aeronautical engineering,
navigation and meteorology of day to day operations, there appears to be far
more limited faith in the place of theory in training and instruction. This
tension can be reduced if the true complementarity of the professional and
the researcher is emphasised. Each needs the other, and the industry needs
both.
In Section 3 (Ab Initio Pilot Training and Instruction) this situation is
explored in depth by Blyth (British Aerospace), Droog (KLM), Elshaw (UK
Ministry of Defence), Lehrer (Embry-Riddle Aeronautical University),
Pippin (Ohio State University) and Telfer (University of Newcastle). The
special place of medicine in aviation instruction is taken up by psychiatrist
and private pilot, Lambeth (University of Newcastle).
The content of the book follows a chronology of the pilot's career: from
selection, through initial qualification then further training of different
purposes. The significant difference of this book is that it is also based on
the recognition that a pilot's training does not finish there. The cycle
continues for as long as the professional (as many others, apart from pilots,
are involved) is employed. In the case of the pilot, however, the input of
aviation instruction is for as long as a pilot flies. The simplicity of that
statement tends to conceal the gravity of its significance. The quality of
aviation instruction and training affects all associated with the industry for
the length of their active participation.
It is a specialist field, conducted in a unique context, by professionals.
This book is intended to demonstrate and support that fact.

References

Badre, A.N. (1982), 'Selecting and representing information structures for

visual presentation’, IEEE Transactions on Systems, Man and
Cybernetics, SMC-12(4), pp. 494-504.

Borko, H. and Livingston, C. (1989), 'Cognition and improvisation:

Differences in mathematics instruction by expert and novice teachers',
American Educational Research Journal, vol. 26, pp. 473-498.

International Civil Aviation Organization (1991), Report o f the Flight Safety

and Human Factors Regional Seminar, Bangkok, 25-28 November.

6
 Introduction to Aviation Instruction and Training

Kurstler, J.H. (1982), A Clown in the Moonlight, Sphere Books, New York.

Lintem, G. and Roscoe, S.N. (1986), 'Adaptive perceptualmotor training', in

Aviation Psychology, Roscoe, S.N. (ed.), Iowa State University Press,
Ames.

McDonald, F.J. (1987), 'Strategic evaluation of training', in May, L.S.,

Moore, C.A. and Zammit, S.J. (eds.), Evaluating Business and Industry
Training, Kluwer, Boston.

Telfer, R. (1989), I f it ain't broke, why fix it? Instruction and preventive
maintenance’, in Jensen, R. (ed.), Proceedings o f the 5th International
Symposium on Aviation Psychology, The Ohio State University,
Colombus.

Telfer, R. and Biggs, J. (1988), Psychology and Flight Training, Iowa State
University Press, Ames.

Waxman, H.C. and Walberg, H.J. (1991), Effective Teaching: Current

Research, McCutcheon, Berkley.

7
Section 1

Aptitude Testing,
Selection and Licensing

(Photo: Qantas)
1 Introduction to section
1: pilot aptitude
testing, selection and
licensing
R.A. Telfer

Because a failed pilot is expensive in terms of both equipment and people,

endeavours to identify the underlying abilities which make a successful pilot
go back almost half a century to post-World War Π USA (Adams, 1987).
There remain those in the industry who employ a type of phrenology, but
instead of bumps on the cranium they are guided by other externals such as
clothing, appearance, posture, or accent, Ί knew the minute he came in the
door
Interviews, opportunities for observation in different situations, pencil and
paper tests, tests of judgement, tests of skill, test flights in a simulator or
aircraft: all are methods of selection used in the aviation industry. Some
types of skills tests which have been identified as being of value include:

- Coordination test of joystick and footpedals;

- Rudder control test;
- Pursuit of a rotating target with a hand control;
- Two hand coordination; and,
- Direction control (Schmidt, 1988).

The essential quality of an aptitude test is one of predictability. It has to

be able to identify the applicants who are likely to fail their training, or pose
a danger as a practising professional. If a test can do so, it has predictive
validity. Any pilot aptitude test used in selection is valid only to the extent
that it is able to predict the individual's performance as a pilot. The
Aviation Instruction and Training

predictive validity is demonstrated by analysing the correlation of the test

with criteria such as hours to solo; hours to command; or freedom from
incident or accident. The higher the correlation, to a maximum of 1.0, the
better. Because of the time and effort involved in gathering such data by
tracking a large number of subjects through their aviation career, full details
of long-term validity are rarely available.
However, because of the complexities of individuals, the job of flying and
the ways in which we judge people's success in the job, the actual
correlation figure of a useful test may not be very high (Ashman and Telfer,
1983). Where a perfect correlation is 1.0, a test achieving a correlation of
0.4 may still be able to differentiate those who will do better than others. It
has been estimated that by itself, a predictor test rarely correlates above
0.45.
Approximately twenty per cent (0.45 squared multiplied by 100 = 20.25
per cent) of the abilities identified in the selection test will be common to the
criterion. The prediction correlation can be improved to about 0.70 by
administering a battery of tests which are intended to represent the abilities a
pilot needs. One test could be based on numerical skills; one on spatial
skills; one on diagrammatic skills and another on literacy skills. This 0.7
correlation represents almost half of the relevant pilot abilities (0.7 squared
multiplied by 100 = 49 per cent) (Schmidt, 1988.)
The development of computerized aptitude testing has enabled batteries
of tests to be combined in a more sophisticated way. Such tests can not only
cover a wide range of abilities, but can facilitate scoring and maintenance of
records of accuracy and speed.
A major difficulty is deciding the criterion against which to calculate the
correlation. In the case of entry to flying school, one could choose a
composite score which incorporated flying (judgements by instructors) and
ground school (marks on tests). To-day, airlines and the services seeking
early indicators of command potential may be also check leadership, social
skills, airmanship, group dynamics and decision making.
Questionnaires or other instruments used for testing also need to have
content validity. If spelling and mathematics are included in the test, then it
is assumed that one can generalize from these items to the abilities that are
vital for pilots. In other words, those who do best in spelling and
mathematics are the best pilots. As you will have concluded from this
example, content validity can be quite contentious. The content of the test
has to be related to the purpose of the test.
There is also a construct validity to consider. This refers not to how the
questionnaire or test is constructed, but to the variables (termed 'constructs')

12
 Introduction to Aptitude Testing, Selection and Licensing

it takes into account. For pilots, such constructs may include risk-taking,
speed of response, and accuracy. They are derived from a job analysis and
psychologists' views of what makes a successful pilot, but can also be
derived by statistical methods which calculate what most affects results.
As an introduction to this Section on Aptitude Testing, Selection and
Licensing, Alexander and Stead provide the background to aptitude
assessment and testing. Both highly experienced aviation psychologists
specializing in pilot selection with SHL Asia Pacific, they are able to cite
examples of airline requirements and how they were met in the design of
aptitude tests and selection techniques. They compare the task of selecting
individuals for ab initio training, and weigh interviews against aptitude
assessment. A typical selection process is presented with examples of the
types of test items which would be employed. In an interesting conclusion,
they forecast likely trends in testing and selection, one of which is a greater
use of computer based systems.
Micropat, the computer based system devised by Professor David
Bartram and others at Hull University exemplifies that advance. In the
second chapter of the section, Bartram points to the advantages for
employers of good selection procedures, and provides further background on
the aspects of validity, base rate and selection ratios. A new generation of
computer-based pilot aptitude tests is introduced, with specific detail of the
Micropat and its validation data. Bartram also looks to the future,
hypothesizing that the roles of systems manager and trouble shooter may be
those fly-by-wire aircraft demand of their pilots. This in turn will demand
changes in selection methods for which Bartram offers some consumer
advice in his conclusion.
The third chapter of this section is provided by Hannan and O'Hara of the
University of Tasmania at Launceston. It is based on a case study of the
training and selection of Air Traffic Controllers in Australia. After a sketch
of the antecedents to the development of the University of Tasmania/Civil
Aviation Authority program for the Diploma of Applied Science (Air Traffic
Services), they point out that there were no aptitude tests used in Australian
selection until 1958. The existing method of recruitment and training is
evaluated, with consideration being given to the implications of new
administrative goals of efficiency through rationalisation, cost savings and
staff reductions.
Professor Graham and Lynn Hunt of Massey University's School of
Aviation collaborated with the New Zealand Civil Aviation Authority to
develop a computer-based testing system for pilot licensing, based on
specified competencies. In their chapter, they introduce the basic problems

13
Aviation Instruction and Training

in contemporaiy pilot licensing methods and describe the ways in which the
Computer-Based Aviation Licensing and Examination System (CALES) was
developed. Examples of test items and the method of generating them are
presented before the CALES model is explained. This model directly links
competence and testing, an important part of the learning cycle (Boekaerts,
1991).
Computer technology seems poised to exert an even greater influence in
aviation. Already it is dominant in sales, reservations, training, aircraft
systems and operations. Pilot selection, testing and licensing will follow.
There would be considerable support for the next task for computers to be
implementing a basic principle of job design for the long-haul pilot:
reduction of boredom and fatigue (Grandjean, 1988).

References

Adams, J.A. (1987), 'Historical review and appraisal of research on the

learning, transfer and retention of human motor skills', Psychological
Bulletin, vol. 101, pp. 41-74.

Ashman, A. and Telfer, R. (1983), 'Personality profiles of pilots', Aviation,

Space and Environmental Medicine, vol. 54, no. 10, pp. 440-443.

Boekaerts, M. (1991), 'Subjective competence, appraisals and self-

assessment', Learning and Instruction, vol. I , no. 1. pp. 1-17.

Grandjean, A. (1988), Fitting the Task to the Man - A Textbook o f

Occupational Ergonomics, Taylor and Francis, London.

Schmidt, R.A. (1988), Motor Control and Learning: A Behavioral

Emphasis, (2nd ed.), Kinetics Publishers, Champaign.

Telfer, R.A. (1985), 'Micro-computer based psychological testing and record

keeping', Defence Force Journal, vol. 54, October, pp. 57-61.

14
2 Aptitude assessment in
pilot selection
K. Alexander and G. Stead

It is generally accepted that aptitude, ability and personality testing had its
first major impact in the selection procedures for military personnel in the
first World War. With the demand for pilots being high, especially in the
latter stages of the war, came an equally high drop-out rate during training.
In addition, the high proportion of flying accidents were attributed back to
the failure of many pilots to operate their aircraft as they had been trained to
do. What grew out of these problems were selection techniques that
involved the use of aptitude tests aimed at predicting flying training success
and reducing flying accidents.
The selection of pilots became a little more sophisticated during World
War Π. Aptitude tests included the assessment of abilities, occasionally
personality, and some form of rudimentary psychomotor skills. These tests,
unsophisticated by today's standards, showed that those pilot applicants who
tended to perform better in their entry tests tended to perform better in their
flight training programs.
After World War Π, many of the established and emerging airlines
throughout the world recruited ex-military pilots, who had, in the main,
undertaken aptitude screening before being accepted into the military.
However, airline operators began to realize that different types of flying
required different aptitudes, skills and styles and the commercial operators
began to use their own methods for identifying those with the potential and
the aptitude to operate as safe and efficient pilots.
Currently, there are few avenues into flying which do not include some
form of aptitude assessment and the majority of the flying training programs
within the military, and with many of the larger commercial operators
throughout the world, use fairly sophisticated procedures to ensure that they
Aviation Instruction and Training

have selected the most motivated and skilled personnel. With this in mind, it
is appropriate to review the rationale and logic behind aptitude assessment
of pilot selection, to review the applications of such assessment and to look
closely at the types of assessment and screening processes which are most
widely used. It is also appropriate to outline the general principles which
underpin pilot selection. This chapter closes with a review of the likely
developments and trends in aviation assessment over the next ten years.

Rationale behind aptitude assessment

Historical precedent

Obviously, the experience gained by the military in using aptitude tests in

pilot selection has spread to many of the world's airlines. However, the
catalyst that prompted many of the airlines to introduce aptitude screening
occurred with the advent of the jet era in the late 1950s, although a small
number of airlines such as Scandinavian Airlines System (SAS) began
aptitude testing in the early 1950s.
Increasing aircraft fleet sizes created a demand for more pilots which
could not be taken up solely with recruits from the military. Less
experienced general aviation pilots were recruited in increasing numbers and
the early 1960s also saw the creation of a number of airline ab initio cadet
pilot training schemes.
It is not surprising, then, that a historical precedent existed which tended
to be emulated, partly because there was no better method or technique
which had presented itself as an alternative. There was, of course, anecdotal
evidence from early selection procedures highlighting failure rates,
expensive accidents (both in terms of aircraft and personnel) and the likely
ease of transition into flying from previous occupations and activities.

Rapidly changing aviation environment

There are few fields of human endeavour where we have moved more
quickly than we have in the area of aviation. From the Wright Brothers first
manned flight to today's highly sophisticated military and commercial
aircraft, flying and aircraft operations have changed dramatically.
Accordingly, in selecting for a career in aviation, some effort needs to be
made to ensure that the individual has the capacity to make the transition to
new and improved aircraft, new aviation operating techniques and different

16
 Aptitude Assessment in Pilot Selection

local, national and international requirements.

Factors such as an individual's adaptability to change, intellectual capacity
to absorb new information, and the qualities for becoming an effective team
member and potential leader in complex operating situations, all became
vital issues in the selection of future pilots. Once again, it became apparent
that it was possible to assess a number of these attributes during the
selection process.

Accident prevention

It is not unnatural for aviation organizations, both military and commercial

alike to place emphasis on safety and accident prevention. Aviation is a
field where there has always been a close investigation of the cause of
accidents and there tends to be an accepted international code which leads
to detailed investigation of accidents and the publication of findings. Many
of these investigations highlight human error as being a prime or contributing
cause to accidents and attempts are obviously made to ensure that such
accidents are prevented in the future.
Some of the human factors' contributing to accidents have been identified
as aptitude for flying, attitude to flying responsibilities and obligations,
communication skills and team interaction issues, all of which lend
themselves in someway to pre-screening, usually prior to employment and/or
prior to the commencement of aviation training.

Costs versus benefit in aviation assessment

Another persuasive argument for the use of sophisticated assessment

procedures is the issue of 'utility', or 'cost versus benefit'. The high costs of
training will come as no surprise to the aviation professional, or the student
contemplating undertaking flying training for whatever purpose.
Naturally, those with the responsibility for operating and maintaining
fleets of sophisticated aircraft are anxious to ensure that time and money are
not wasted nor safety compromised in attempting to train people who may
ultimately not reach the desired standards. It is not difficult to understand
the logic of investing a relatively small amount of time and expense, during
the selection process, in using reliable aptitude testing techniques which may
avoid costly training failures and selecting potentially unsafe pilots.
Not surprisingly, there has been much interest in the evidence which
enables the qualification of the benefits in relation to aptitude assessment
and in this area 'validation' evidence is all important. Validation involves

17
Aviation Instruction and Training

demonstrating that aptitude assessment bears a clear link to likely future

training and operational effectiveness as a pilot. Such research evidence,
more rigorously collected in recent years, shows the obvious connection in
all types of flying and provides the final plank in a platform supporting the
use of aptitude assessment in flying training.

What is aptitude testing?

Aptitude testing assesses individual components of ability, and even

behaviour, related to the knowledge and skills a person now has or has the
ability to acquire in the future. Typically aptitude tests have included verbal
and numerical reasoning, abstract reasoning, mechanical reasoning, spatial
reasoning, clerical speed and accuracy and manual dexterity. It should be
pointed out, that these types of aptitude tests are not measures of intelligence
(IQ), but rather of specific attributes.
During World War Π military psychologists constructed multiple aptitude
test batteries. Air Forces in particular developed specialist aptitude batteries
that were used in the selection of pilots, navigators, wireless operators,
bomb aimers and other specialists. Also included in aptitude testing are
those tests that involve psychomotor skills. Various psychomotor tests
were, again, developed during World War Π and more often than not
involved the assessment of eye, hand and foot co-ordination. In more recent
years aircraft simulators, from the basic link trainers through to complex
multi-million dollar 747 simulators, have been used to assess applicants
flying aptitude and importantly, potential for learning.
Another and perhaps slightly more controversial component of aptitude
testing is personality assessment. Whilst personality cannot be considered
to be in any way similar to ability or skill or to be a measure of the capacity
to learn, many flying organizations, be they military or airlines, have some
form of personality criteria that are often called motivational, attitudinal and
interpersonal characteristics. For example, one airline had the following
behavioural requirements for its pilots, who were required to:

- be team oriented;
- have appropriate maturity, confidence and stress tolerance;
- be adaptable and flexible; and
- be able to exercise sound judgement and decision making.

In discussing aptitude assessment in pilot selection, it is important not to

18
 Aptitude Assessment in Pilot Selection

overlook all the principal components that go into establishing whether an

individual has the necessary skills, abilities, motivation and personal
characteristics required to be a professional, efficient and safe pilot. After
all, these are the main overall concerns for most aviation organizations.

Alternatives to aptitude assessment

There are, of course, some aviation organizations who do not use any form
of aptitude assessment, or, at best, use only a rudimentary simulator check in
their selection of pilots. With beliefs ranging from, Ί can tell a pilot when I
see one', through to, 'If you can fly a 747 for one airline you can fly a 747 for
any airline', selectors use alternatives to aptitude assessment, usually basing
judgements on interview, flying experience and track record with previous
employers. These techniques are, of course, only useful up to a point.
The issues become more complex however, when trying to assess
individuals for ab initio training or for selection into organizations that have
different operational objectives and standards requirements. The weight of
hard evidence suggests that interview predictions are simply not as effective
as those integrating aptitude assessment. The logical decision seems to be to
merge these different forms of assessment into an effective, integrated
overall selection procedure.

Selection

As mentioned earlier, the concern generally is to establish whether an

individual has the necessary aptitudes and abilities to undertake flying
training, to be effective as an operating pilot and whether or not the
individual has the necessary motivations and behavioural characteristics to
ensure that they will operate in a professional way.
Occupationally oriented assessment processes can obviously be very
useful in selection, particularly if they help in more efficiently identifying,
perhaps from a large number of applicants, those who have the potential to
complete training in minimum time and meet operational standards without
difficulty.
Accordingly, many commercial operators, the military and some training
institutions use some form of aptitude assessment to select ab initio or
experienced pilots. However, selection is often not a straight forward issue.
There are, obviously, differences in selecting pilots into say, an airline where
that organization recruits both experienced and ab initio (cadet) pilots.

19
Aviation Instruction and Training

Experienced pilots come with the prerequisite minimum hours, the licences
and the endorsements. Amongst this experienced group, varying operating
standards, degrees of skill, experience and personal qualities are found to the
extent that a thorough aptitude testing process is a routine component of
most selection programmes.
However, when it comes to selecting ab initio trainees then the selection
process has to rely more heavily on aptitude testing. Ab initio candidates do
not have the experience, the track record or the licences on which to be
assessed for their overall suitability. With the high cost of training an ab
initio trainee from nil experience through to checking out as a line pilot, the
selection process must be as rigorous as possible, accurately assessing the
many criteria needed to become an effective pilot. Most airlines who have
ab initio training schemes include ability and reasoning testing, personality
assessment and psychomotor skills evaluation as well as the traditional
interview and a review of educational qualifications in their assessment of
candidates.

A typical selection process

Larger flying organizations such as airlines and military services tend to use
the more sophisticated selection procedures that involve a number of
assessment steps. This is not unusual in that they have the infrastructure and
experience available to utilize in selecting from a large number of
candidates, be they for ab initio training programs or for direct intakes of
experienced pilots.
A typical airline selection procedure will usually involve a multi­
assessment approach in the screening of applicants. This approach may
include for, say, experienced pilot candidates the following stages:

- screening of application forms for relevant experience, age and track

record;
- psychological assessment by way of paper and pencil testing and/or
computer based assessment;
- initial interview which is often designed to assess the candidate's
overall suitability to progress to further stages of the selection process
by way of questioning their general aeronautical knowledge, their
background and track records;
- skills assessment using either or both a simulator and 'co-ordinator'
assessing psychomotor skills; and a
- final interview before a panel of experienced pilots/personnel who

20
 Aptitude Assessment in Pilot Selection

integrate and merge all the available data on a candidate in order to

make a balanced overall decision about the suitability for appointment.

Additional factors such as medical condition, fitness and eyesight are

assessed separately, as there is usually a distinct pass/fail implication
involved.

Types of assessment procedure and psychological screening

Paper and pencil tests

By far the most common type of aptitude assessment for flying training is
pencil and paper psychological testing. It is the most widely used technique
because of cost effectiveness, ease of administration and flexibility in
transportation and use of materials and because large numbers of candidates
can be assessed at one sitting.
Generally, assessments of aptitude and ability are conducted, along with
assessments of personal style and emotional pattern. The aptitude and
ability tests (Table 2.1) normally encompass verbal and numerical reasoning,
in tests of some fifteen to thirty minutes in duration and occasionally
diagrammatic or 'non verbal' tests are also used in order to assess problem
solving capacity which is more symbolic or diagrammatic in type. In such
tests, an indication is being sought of the individual's ability to solve
problems similar to those they might face in either training or operational
work as a pilot. Reading and understanding the written or spoken word,
following the logic in instructions or arguments, communicating effectively
to others, solving mathematical problems, inferring trends in data, reading
from a chart or schedule, solving logical problems from diagrammatic/visual
indications from instruments and read outs, are all involved in a pilot's job.
In a testing situation it is legitimate to assess the level of a person's capacity
to perform such work.
These timed tests are normally administered in one sitting, seldom lasting
for more than two hours in total and are usually preceded by clear and
concise instructions and a practise section where the testee has an
opportunity to complete items similar to those in the actual test. It is difficult
to prepare for such tests and the main value in some practise or prior
exposure is to lower test anxiety and build confidence in the testee.
Accordingly, most organizations conducting such tests will offer some
briefing or practise opportunities before the actual tests commence.

21
Aviation Instruction and Training

Table 2.1
Examples of aptitude and abilities items

Verbal Test 1

In this test you have to select the word which best completes the word
box. Two of the words in the box are related. Your task is to complete
the word box by finding the answer which bears a similar relationship to
the third word in the box.

1. quiet narrow
_________________________ A B C D E
? noisy calm thin rapid loud wide

2. decant pour
_________________________ A B C D E
level ? discount pore even descant drink

Numerical Test 1

In this test, select the number which best completes the series.

3. 1.5 8.5 15.5 ? 29.5 A B C D E

17 23.5 21 22.5 20.5

4. 3 6 ? 24 48 A B C D E
20 12 18 9 21

Verbal Test 2

Here your task is to evaluate each statement in the light of the passage.
Read through the passage and evaluate the statements according to the
rules below, then mark the appropriate box on the answer sheet.

22
 Aptitude Assessment in Pilot Selection

MARK BOX A MARK BOX B MARK BOX C

If the statment is If the statement is If neither the
patently true or patently untme or the statement nor the
follows logically, opposite follows opposite is
given the information logically, given the patently true or
or opinions information or follows logically,
contained in the opinions contained or if you cannot
passage. in the passage. say without
futher information.

In October of this year it was estimated that there were 10,000 word
processor installations in Australia, with a projected increase of 20 to 30
per cent a year. Recent growth has not been matched by that of skilled
operators so that companies must invest not only in the work processors
themselves, but also in courses to train existing staff to operate them.

5. There are insufficient ready trained word processor operators to meet

demand in Australia.

6. Last year, over 2,000 word processors were installed in Australia.

Numerical Test 2

Bennett Ltd.
Factory Production Costs

Answer the questions below using Units Total

the figures presented in the table. produced cost per
For each question, you are given per hour unit
five possible answers to choose 5 $4.65
from. Only one answer is 6 $4.56
correct in each case. 7 $4.52
8 $4.53

23
 Aviation Instruction and Training

7. Units produced are sold for $12 each. What is the maximum gross
profit per unit?

A B C D E
$4.65 $7.48 $4.52 $7.44 $7.35

8. Labour currently costs $15 per hour. At a production rate of 6 per hour,
what will the unit production cost be if labour costs rise to $18 per
hour?

A B C D E
$7.56 $5.02 7.52 $4.65 $5.06

F o llo w in g C o d e d In s tru c tio n s

In the questions below there are figures on the left in

boxes, to which instructions in circles have been
applied. Using the coded instructions in the table,
select the configuration that would result from
applying the instructions, and mark the appropriate
box on the answer sheet.

9.

10.

24
 Aptitude Assessment in Pilot Selection

Diagrammatic Test

In these questions, there are a series of diagrams which follow a logical sequence on the left of the page.
From each row one diagram has been left out. You are to choose the diagram which will complete the
sequence from the five options provided on the right of each row. Mark your answer on the answer sheet.

A B C D E

11.

12
·

Often ability tests are accompanied by questionnaires designed to assess

an individual's preferred work style, personality, attitude and approach to
work situations. Such questionnaires will vary in their length and style of
questioning and take about an hour to complete and generally produce a
profile showing an individual's work attributes and style against a reference
group or norm group. From such a profile it is possible to identify, for
example, whether an individual is appropriately forceful in controlling work
situations, is democratic in outlook to handle a team work setting, is flexible
or traditional in approach to work and how orderly and systematic they are
in executing tasks. In addition, attributes like ambition, competitiveness,
anxiety and sociability can be assessed.
Generally, there is no 'ideal' against which an individual is compared,
although most organizations do have some form of 'profile' (Table 2.2) by
which to assess candidates against. Different attributes are required in
different types of flying such as those attributes which may be required for
optimum handling of an emergency situation, as opposed to the attributes
required for effective operations in routine flying assignments.
With both ability and motivational personality measures there is some
error of measurement inherent in the testing process itself and there is some
potential for the individual to change over time. Accordingly, individuals
are, on occasions, given the opportunity to re-sit tests after an appropriate
interval of not less than twelve months.

25
Aviation Instruction and Training

Computer based assessments

With the advent of efficient personal computers, a number of the previously

mentioned pencil and paper tests can now be administered, scored and
interpreted using the personal computer. In such cases, the questions
previously part of a pencil and paper booklet or questionnaire, are screen
administered and the testee responds on the computer keyboard. Various
levels of sophistication exist, up to and including 'adaptive' tests where the
computer's speed and memory are used to individually adapt the questions
asked within a test to suit the testee's previous answers and the speed with
which the answer was derived. This added sophistication in assessment has,
until now, been offset by the lack of portability of such equipment and the
requirement to have a machine for each individual candidate.

Computer technology in pilot aptitude testing

(Photo: SHL Asia Pacific)

Personal computers have also been exploited to vary the type of

assessment beyond what was possible with normal pencil and paper
testing. For example, the computer can more closely simulate 'real life'

26
 Aptitude Assessment in Pilot Selection

Table 2.2
Example of a behavioural profile

Relationships with people

1 2 3 4 5 6 7 8 9 10 Controlling - Takes charge, directs,

manages, organizes, supervises others.

1 2 3 4 5 6 7 8 9 10 Affiliative - Has many friends, enjoys

being in groups, likes companionship,
shares things with friends.

1 2 3 4 5 6 7 8 9 10 Democratic - Encourages others to

others to contribute, consults, listens and
refers to others.

Thinking Style

1 2 3 4 5 6 7 8 9 10 Data Rational - Good with data, operates

on facts, enjoys assessing and measuring.

1 2 3 4 5 6 7 8 9 10 Forward Planning - Prepares well in

advance, enjoys target setting, forecasts
trends, plans projects.

1 2 3 4 5 6 7 8 9 10 Conscientious - Sticks to deadlines,

completes jobs, perseveres with routine,
likes fixed schedules.

Feelings and Emotions

1 2 3 4 5 6 7 8 9 10 Relaxed - Calm, relaxed, cool under

pressure, free from anxiety, can switch off.

1 2 3 4 5 6 7 8 9 10 Emotional Control - Restrained in showing

.. emotions, keeps feelings back, avoids
outbursts.

1 2 3 4 5 6 7 8 9 10 Decisive - Quick at conclusions, weighs

things up rapidly, may be hasty, tasks risks.

27
Aviation Instruction and Training

situations, so as to test a candidate's ability to read, integrate and understand

cockpit instruments, to interpret diagrams visually and to use eye and hand
co-ordination in fine judgement and decision making tasks, amongst others.
Somewhere between PC simulation and full simulator assessment of an
individual lies 'complex co-ordinator' assessment devices which use pedals
and joy stick arrangements, in conjunction with PC presented stimuli to test
aspects of pilot operating performance. These type of aptitude testing
devices are especially useful in assessing the co-ordination skills of ab initio
candidates. They can also be reliable predictors of training and on-line flying
performance of those pilots who have considerable flying experience gained
before being employed.

Simulator assessment

Tests of the pencil and paper or computer administered type are standardized
samples of work related behaviour from which a prediction of subsequent
work behaviour can be made. As far as possible, the content of the test
simulates the content of the decision making involved in the job. It follows
logically that the closer one comes to simulating the real job which has to be
done, the more effective the prediction of job behaviour is likely to be. With
this in mind, most airlines, for example, use flight simulators for assessment
purposes.
Provided the simulator is sufficiently simple to be usable in assessment
without prior training on its use, then such simulators can represent an ideal
opportunity to expose the candidate to a 'real life' situation and to measure
the effectiveness of the responses which are made, using the related
computer for assessment and diagnosis purposes. Training organizations
have long recognized the potential for simulators in training, because of the
cost effectiveness and safety aspects of simulating situations rather than
using a real aircraft. It is, though, only a short step to modify these
simulators for assessment.
A typical assessment procedure for a pilot candidate with flying
experience could involve, over a one hour session:

- straight and level flying skills;

- climbing and descending turns;
- steep turns;
- ILS interception and approach; and
- ADF tracking, back-tracking, holding and approach.

28
 Aptitude Assessment in Pilot Selection

During these simulator sessions a candidate's basic instrument scan and

co-ordination skills can also be assessed.

Group assessments for style and group behaviour

Another form of simulation exercise, though removed from the aircraft

simulator, is where a group of individuals are observed in a work activity or
decision making situation. Trained observers look for the candidate's
contribution to group effectiveness and the individual's ability to inter-relate
with other members of the group and, in certain instances, lead and direct the
group towards problem identification and solution. Used more extensively
in management assessment and potential identification, this form of
assessment has been used by some airlines and by many military
organizations to assess team/crew behaviour, decision making, leadership,
communication and task awareness.

Multiple assessment strategies

Various methods of assessment have been mentioned and each method has
its advantages and disadvantages. Each method has its own validity, or
capacity to predict subsequent work performance which can be measured in
statistical terms. If the different methods predict an aspect of subsequent
work performance, then it is possible to add these methods together to
enhance the overall prediction. With minor statistical corrections to account
for the degree of similarity in the processes involved, a 'multiple assessment'
can be made.
There is an increasing trend to combine different assessment methods,
weighting the contribution of each according to its predictive capacity. Using
this multiple assessment technique, it is also possible to integrate aptitude
and behavioural assessment measures with other inputs in the selection
process, like performance in interview, previous education and training and
previous work experience.
More modem approaches use 'intelligent recruitment systems' to make
the best possible prediction of likely future success and to provide
information to the selection panel of the strengths and weaknesses of each
candidate against the competencies required for the job. This can lead to
refined assessment interviewing, and to the development of induction and
training guidelines for the personnel who will take selected candidates
through to graduation as a pilot.

29
Aviation Instruction and Training

The general approach to aptitude assessment in selection

Before any assessment can be fully effective, one must have a very clear
understanding of the content of the particular flying job involved and an
awareness of the competencies which are critical for successful performance
in such a job. In this respect, then, it is rare to find any assessment of
aptitude and ability which is not predicated either on previous research or
upon detailed job analysis. Because some flying responsibilities are graded
and vary according to aircraft type and responsibility level, this analysis of
the job and the competencies required must relate to the aircraft in use, or
planned for the future and the operating scenarios of the organization
concerned.
It is not at all unusual to have a different set of competencies and a
different selection system for the selection of ab initio pilots entering a
'cadet' training scheme, compared to those which might be used to consider
experienced pilots from other organizations, or from the military services.
After the job analysis, the competencies identified need to be weighted in
accordance with their contribution to overall success. On the basis of such
weights, a system is devised to ensure that the competencies are assessed as
fairly, objectively and ethically as possible. This underpins the choice of
assessment methods and devices and the development of appropriate
reference norms which can be used to interpret the scores obtained by
candidates once they undertake the assessment program. At this stage,
report formats are devised to ensure that different parts of the selection
process are properly integrated in order to facilitate effective, objective
selection decision making.
Once a full selection system has been installed, the proper procedure is to
conduct ongoing validation and review, evaluating the effectiveness of
various parts of the process and improving or modifying the process as
required. Airlines, training organizations, military services and professional
groups are co-operating in this process more fully than ever before, gradually
enhancing the selection process over time. Of course no selection process is
ever perfect, but the objective should always be to maintain the highest
possible standards and success rates.

Likely future developments and trends

Increased use o f assessment tests and test data

30
 Aptitude Assessment in Pilot Selection

With the efficiency and cost effectiveness of sophisticated selection

processes, it seems inevitable that there will be increased use of fully
integrated aptitude and assessment selection systems throughout the aviation
industry, especially within airlines. This is becoming increasingly evident in
those airlines who select candidates for ab initio flight training schemes.
Refinement of the processes will gradually lower failure rates and
increase the likelihood of selecting pilots who will meet appropriate
standards in the short term and who can undertake upgrade and conversion
training, in due course, to meet the standards and objectives which will exist
in the future.
Information about individual strengths and weaknesses, appropriately fed
back to the individual in an environment of training and development will
enhance the individual's capacity to meet training and operational standards.
Accordingly, more time and effort is likely to be devoted to the discussion of
aptitude and personality/behaviour results in the ongoing performance
assessment and counselling area.
It is also probable that assessment data will be collated for research
purposes and will begin to flow on to other areas outside of selection, such
as aircraft and cockpit design, operational systems review, accident
investigation, team development and so on.

Purpose designed aptitude assessments

It is not unreasonable to suggest that the near future will see the evolution of
purpose designed assessment tests taking into account different types of
aircraft, flying environments, types of aviation operation and for specific
needs i.e. ab initio selection, experienced pilot selection, conversion
training, ratings and endorsements.

Greater use o f computer based and integrated systems

More extensive availability of personal, mainframe and networked computer

systems should lead to increased use of computer based assessment. In
addition, there is already evidence to suggest that computer based selection
models can make better overall predictions of likely future success than
alternative techniques. It is logical to take advantage of the computer’s
capacities in this respect. This particularly relates to multiple assessment
systems where different data must be merged, integrated and weighted to
produce an overall end recommendation. Computers are uniquely suited to
this form of integration.

31
Aviation Instruction and Training

Increased use of flying simulators, with their extensive computer back-up

so as to simulate real life situations, can also be expected.

Expanding research evidence and data

There will be an increasing world wide trend to publish and pool aptitude
and style assessment data. The sharing of this information will enable early
evaluation of the differences both in assessment and in subsequent pilot
performance, according to factors like age, sex, cultural background and
education, which will lead to refinement in the processes of selection and
training. This in turn will facilitate achievement of other important social
and political objectives like fairness, equality and non-discimination.

Summary

Aptitude assessment in pilot selection is widely accepted in commercial

airlines and tends to be increasing, partly because of its demonstrated
success and also because of the increasing demand for pilots, both
experienced and, more so, ah initio trainees.
Effective pilot selection procedures should not be considered static and
fixed. Whilst they should be standardized and as objective as possible, it is
important that the process be dynamic and flexible enough to meet and cope
with changing aircraft technology and operational requirements. Today’s
pilots are quite different from those of a generation ago. With ever changing
advancements in technology and aircraft operating systems, it is almost
certain that the skills, aptitudes and personal qualities required of a pilot,
within another generation will be again quite different from today’s pilot
requirements.
However, the challenge for any selection system is to continue to recruit
efficient and safe pilots. A comprehensive and fully integrated selection
process that focuses, not just on the immediate requirements, but also on the
longer term needs of the aviation organization, will go a long way in
ensuring that the most suitable candidates are selected.

References

Anastasi, A. (1982), Psychological Testing, (5th ed.), MacMillan, New

York.

32
 Aptitude Assessment in Pilot Selection

Bond, N.A, Bryan, G.L, Rigney, J.W. and Warren, N.D. (1986) Aviation
Psychology, University of Southern California, Los Angeles.

Stead, G. (1987), 'Pilot selection - the QANTAS airways experience', Paper

presented to the Western European Association fo r Aviation Psychology
Conference, Baden Bei Wien, Austria, September.

Stead, G. (1991), Ά validation study of the QANTAS pilot selection

process', in Farmer, E. (ed.), Human Resource Management in Aviation,
Proceedings of the X V I11 WEAAP Conference (1989), Avery Technical
Academic Publishing Group, Brighton, pp. 3-18.

33
3 Aptitude testing and
selection in aviation
D. Bartram

There is probably no situation which benefits more from the technology of

aptitude testing than that of pilot selection. Training courses for military fast-
jet pilots are probably the most expensive of all occupational training courses
(with costs in excess of £2 million per person often quoted). The machines
that pilots have to master are becoming increasingly complex and more and
more expensive to replace: with modem fighters costing £15 to £20 million a
piece and airliners considerably more.
Given these costs, and also the potential consequences of accidents,
picking the right people for training as pilots is very important. It represents a
major investment decision and carries with it substantial risk factors. A well-
designed selection assessment procedure enables us to quantify the 'risk
factors' and hence make safer investment decisions.
In this chapter I .will try to show how good selection procedures can help
to reduce training costs and the risks of subsequent operational failures.
Through the use of well-designed measures of ability and aptitude, good
selection can:

- reduce training wastage;

- reduce the risk of training accidents; and
- increase the operational effectiveness of the successful trainees.

In a chapter of this length, it is not possible to give a systematic in-depth

review of all the work on pilot selection and the developments of selection
systems, so I will only attempt to sketch out the broader picture. However, I
will try to illustrate in more detail the role of psychological testing in
selection by focussing on the woik on pilot selection in which my colleagues
 Aviation Instruction and Training

and I have been involved over the past years.

To set such work in its context, we need first to consider the background
to selection for flying training and some general principles of personnel
selection.

Assessing the relationship between aptitude and training performance

To appreciate the importance of good selection it is necessary to see why

poor selection is such a bad idea - because in practice good selection will
never be perfect: just better than poor selection!
The worst sort of selection procedure is no selection at all. If you selected
people at random for a training course, and then gave them as much training
as they needed to reach the criterion of 'successful' performance, you would
find that people varied in the time they took. Some would take so long that
old-age would overtake them before they reached a sufficient level of
competence. Others would give the impression that they had been 'borne to
fly' and would be going solo in a few hours. A large amount of this variation
can be accounted for by differences between people in those abilities or
aptitudes which underlie the acquisition of the skills or knowledge being
trained. The more relevant aptitude they have, the less training required for
them to develop their new skills.
If you put these same randomly selected people onto a course of fixed
length, then as you decrease the length of the course so increasingly more
people will fail to reach the 'criterion' level of performance by the end of the
course. Those failing to reach the criterion will tend to have lower levels of
aptitude than those who succeed.
Turning this argument round, if we are able to assess the relevant aspects
of ability, we can then predict who is likely to pass and who is likely to fail
training on a fixed length course, or obtain a reasonable estimate of time-to-
criterion for those operating open-ended training (as much PPL training is).
The qualities required for success in training can be divided roughly into
three categories:

l. Qualities which are absolutely necessary. There are certain levels of

physical and/or mental disabilities or shortcomings which would make it
impossible for someone to leam to fly - however good the training and
however much time they had. These are probably fairly rare in practice.
They would include: gross problems with motor-coordination, very
poor spatial orientation ability, below average general ability,

35
Aviation Instruction and Training

uncorrectable visual or auditory defects and other forms of physical

handicap. These sorts of problems tend to be pretty obvious - to
applicants and selectors. In practice, there is a very strong relationship
between self-selection and relevant aptitude: people who are clearly
unsuited to flying are generally aware of their shortcomings and do not
apply.
2. Qualities which are essential, given the demands o f the training course.
These qualities can be considered under the general heading of
'trainability factors'. Military and Civil commercial flying training is
carried out with groups of trainees all progressing through a fairly
demanding syllabus and time schedule. Because of the enormous costs,
it is deemed uneconomic to tailor training to people individually - with
'back coursing' often being the only option. Apart from the training cost
implications, there are other good reasons for preferring candidates with
high ability. There is evidence to show that people with low levels of
aptitude not only take longer to train but also have lower final levels of
operational effectiveness. For example, people with the longest HGV
training times have been shown to have the highest subsequent accident
rates.
3. Qualities which are essential to meet the organizational requirements.
These would include any factors which are not likely to influence the
candidates chances of passing or failing a particular training course itself
but which are deemed important for their subsequent operational
effectiveness - or for some subsequent stage of training. Flying training
courses often divide into three main stages: Basic, Advanced and
Operational. Clearly, selection has to be concerned with assessing the
qualities required to pass all three. Each is likely to call on different
aspects of ability and to build on skills which need to have been
consolidated in the earlier stages.

So, the critical issue is: Can we identify and measure the relevant
aptitudes before people are chosen for training? If so, we can then select
only those with a high likelihood of passing through all the stages of training?

Measuring validity

The relationship between measures of potential assessed at selection (e.g.

aptitude test scores) and training performance can be represented using
correlation coefficients. A correlation coefficient can vary from zero
(indicating no relationship) to one (indicating a perfect relationship). A test

36
 Aptitude Testing and Selection in Aviation

with zero validity is no more use than random selection. A test with a
validity of 1.0, on the other hand, would be able to predict with perfect
accuracy each and every training success and failure.
Our measure of training performance would depend on the nature of the
course. If we are looking at fixed length courses, then the appropriate
measure would be training outcome (success or failure). For open-ended
courses, we would look at the prediction of the time taken to reach the
training criterion. The correlation obtained can be used as a measure of the
validity of the selection measures.

Validity, base rate and selection ratios

Validity depends on a whole host of factors, but the most important - also the
most obvious - is the relevance of the qualities assessment by the measure to
successful training performance. Other things being equal, the higher the
relevance, the greater the correlation will be.
The practical value of any given level of validity depends on two other
factors. First the proportion of applicants who would be successful if we
selected at random from the pool of applicants (the base rate). If 100 per
cent would be successful or if none would be, then we would gain nothing by
using aptitude tests to select people. When the base rate is 50 per cent, the
information provided by a relevant test will have its greatest impact.
However, the impact of the test is not symmetrical around the 50 per cent
point. As the base rate drops towards 0 per cent, the need for improvements
over random selection become more and more urgent. As it rises from 50 per
cent to 100 per cent random selection becomes more and more effective and
the impact of any test on final pass rates drops. In practice, the base-rate for
applicants to flying training courses is more likely to be above than below 50
per cent - though it will vaiy with the nature of the course and the procedures
used to attract applicants.
The second factor is the proportion of candidates actually selected (the
selection ratio). The smaller this proportion, the more impact a given level
of validity will have on the training success rate. To maintain a given pass
rate in training when applicant numbers are falling (assuming the base rate
stays constant) we have to increase our selection ratio - as we need more
people in - and increase validity to stop the proportion of successful ones
decreasing.
Base rate, selection ratio and validity all interact with each other in a quite
complex manner. However, it is possible to identify some rales of thumb for
optimizing the overall selection strategy:

37
Aviation Instruction and Training

1. Increase the absolute size of the applicant pool as much as possible

without lowering the base rate. The costs here are those involved in
advertising and other means of reaching the right potential applicants
and the cost of then sifting the applications.
2. Use selection measures which have high validity, as changes in validity
(for given base rates and selection ratios) are directly related to the
utility (cost-effectiveness) of the selection procedure.
3. Keep the selection ratio small - this of course depends on having a good
number of the right sort of applicants in the first instance.

To get a grasp on the value of aptitude testing in the selection process we

need to have a 'feel' for what a 'good' level of validity would be. Where we
are predicting a pass/fail criterion from applicants who are already highly
self-selected (and hence there is a reasonable base rate) for courses which are
quite long - nine months to two or three years - we cannot expect to obtain
validity coefficients much greater than 0.60 for short courses, 0.50 for longer
more complex ones. Unless our selection ratio is very small, validities of less
than 0.20 are unlikely to be of much practical benefit. So the range of useful
values we are looking for is from 0.20 up to 0.50 or 0.60.

Selection for flying training

To get an idea of the sort of abilities required for pilot training we need to
look at how the pilot's task has changed over the past few decades.
The first pilot selection took place in the middle of the last century. In
1849 Sir George Cayley persuaded a boy to skim down a hillside on his
Yorkshire estate in the glider he had constmcted. Four years later he flew his
coachman across a valley on the same estate. The coachman apparently
resigned on the spot after the flight - the first recorded example of poor
selection leading to training drop-out!
The first powered, controlled flight was the Wright Brother's at Kitty Hawk
in 1903. The pilot flew by feel using direct control of wing warping. We can
track the demands made on the pilot if we look at the subsequent
development of the cockpit - its controls and instrumentation. For a long
time, the controls were little more than stick, rudder and throttle, with one or
two instruments giving engine revs, oil pressure etc. Early selection was
based wholly on self-selection. With the advent of the World War One, we
see the start of systematic pilot selection - initially concentrating on physical
screening and only later on psychological testing of aptitude.

38
 Aptitude Testing and Selection in Aviation

As we moved into the seventies and eighties we saw the emergence of the
modem fighter and commercial aircraft. A mushrooming of displays of
various systems and more and more system controls. This reached its peak in
the seventies in planes such as Concorde. In addition, pilots have become
less and less directly 'connected' to the flying control surfaces. Their controls
now actuate servo-mechanisms or - with fly-by-wire - send demands to a
computer.
What are the implications of these changes for pilot selection? Hilton and
Dolgin (1991), in their excellent history of the trends in military pilot
selection over the past 80 years, highlight the fact that despite all the changes
and developments in flying, three dominant themes run throughout the history
of pilot selection: the importance of general ability ('intelligence'),
psychomotor coordination and personality:

1. General ability is an important factor in 'trainability'. The ease and speed

with which people can absorb new information is well-predicted by
measures of general ability. Increasingly, more specific abilities related
to information management skills are being seen as of importance.
2. Personality is more controversial. There is wide acceptance of the
importance of temperament and disposition in operational roles: there is
very little agreement on how we should measure it.
3. As good hand-eye coordination and hand-foot-eye coordination are
essential requirements for controlling 'traditional' aircraft, tests of
psychomotor coordination tended to form the basis of all early pilot
selection programs.

Given the changes in the nature of the pilot's task, if we look at the validity
of motor coordination tests as predictors of flying training, they have tended
to decline from the 1950s through to the 1980s, while evidence for the
importance of personality has tended to remain level (see Hunter, 1989). In
the early days of flying, physical control over the aircraft was the main
purpose of training and hence psychomotor coordination figured large in the
validity stakes. Nowadays, the process of control is far easier and a far
smaller part of the pilots task: he or she is now a systems-manager who has
to deal effectively with inter-personal relationships on and off the flight deck
and with a complex communications network. Coordination tests are still
good predictors of basic flying training but do not predict which of those who
succeed at basic flying will also succeed later.
The problem we currently face in pilot selection is not how to increase
levels of validity but how to try to maintain them at the level (0.40 to 0.50)

39
Aviation Instruction and Training

they were 30 or 40 years ago when life was much simpler! How do we halt
this falling level of prediction? How do we get it back up to the level it was
in the fifties - when flying was simpler and less expensive. Can we simply
improve the traditional tests, or do we need to look at new areas of human
ability?

The new generation of computer-based pilot aptitude tests

Most of the major new developments in pilot aptitude testing have taken
place in the military setting and have involved computer-based assessment.
In the USA the Basic Attributes Tests (BAT: see Carretta, 1987) were
developed from a set of recommendations made by Imhoff and Levine
(1981). In the UK work commenced at about the same time, adopting a
rather different approach, on the development and validation of MICROPAT
(Bartram, Dale and Smith, 1982; Bartram and Dale, 1983; Bartram 1987).
A few years later, the RAF implemented computerised versions of their
battery (including the coordination measures SMA and CVT) and began to
research the potential of a set of tests similar in concept to the US BAT.
Initial MICROPAT validation was carried out against Army and Navy flying
criteria. A tri-service evaluation is now being carried out which includes the
tests developed by the RAF and the MICROPAT tests.
While the above batteries were developed for military use, the overlap
between military and commercial flying - where aptitude is concerned - is
very great. MICROPAT is now being used by Cathay Pacific Airways and
VARIG and is being considered by a number of other airlines.
Within Europe a number of other systems have been developed. In most
cases these have been designed for general personnel selection use - rather
than pilot selection in particular. However, systems like the Dutch
TASKOMAT have been producing very promising results in pilot selection
validation studies.

The MICROPAT tests

Work on MICROPAT began after an examination of the validity of the

existing Army Air Corps selection aptitude tests as predictors of training
outcome. These tests, the SMA, CVT and INS-B, are used by all three
services in the UK (the RAF, the Royal Navy and the Army Air Corps).
While they performed moderately well in predicting basic flying training, they
were performing very poorly by the time pilots reached advanced training

40
 Aptitude Testing and Selection in Aviation

(with the correlation with training outcome being well below 0.10).
Subsequent evidence from both the RN and RAF confirmed these tests are
moderate predictors of early stages of training but poor predictors of later
stages. As we saw earlier, the validity of this type of test has declined over
the past few decades.
The tests produced for MICROPAT included measures of motor
coordination - as this is still important, especially for the prediction of basic
flying training - but focussed on more complex information management tasks
designed to assess those aspects of aptitude related to the later stages of
training. It is not possible, in the space of this chapter, to describe the
various tests in detail. However, the following brief descriptions may give the
'flavour' of the system:

- ADTRACK is a set of coordination tasks which adapt their difficulty level

to the ability of the candidate.
- LANDING is a complex coordination and scheduling test presented as a
simple landing approach simulator.
- PLANE tests speed and accuracy of visual discrimination.
- SCHEDULE is an information management test designed to see how well
people can prioritise and keep track of changing patterns of information
over time.
- NAVOR is a 'sense of direction' test - designed primarily for navigators.
- RISK assesses people’s approach to 'risky' tasks and how well they can
adapt their behaviour to changing odds of success.

At the time they were developed, these tests were quite novel in a number
of respects:

- They were adaptive - in that they changed in response to the

performance of the test taker.
- They used item generation - that is, the computer 'made up’ the test as it
went along on the basis of pre-specified rules. There were no sets of
questions or items built in.
- They were dynamic - they were 'real-time' continuous tasks rather than
lists of questions.

In approach, they fell somewhere between traditional item-based paper-

and-pencil tests on the one hand and use of simulators as task-based test
devices on the other. While some people have advocated the use of
simulators as ab initio selection devices, it is generally felt that the costs

41
Aviation Instruction and Training

associated with this would far outweigh the benefits in terms of validity.
Certainly, simulators and light-plane tests ('grading') have an important role to
play but they are only cost-effective when used for the final short-listed
candidates. To be effective, aptitude tests need to be used early in the
selection process when numbers are large and per capita costs need to be
kept low. The mle in optimizing selection cost-efficiency is to use high-
validity low-cost procedures as early as possible, and high-cost ones as late
as possible.

Developing and validating an aptitude test battery

The development and validation program carried out with MICROPAT can
be used to illustrate a number of general principles about selection validation
in general. In outline, the development and validation procedure involved
eight steps:

1. Analyse the training demands and search sources of evidence about the
abilities required.
2. Design and develop the tests.
3. Obtain information about the psychometric properties of the test
measures and the distributions of scores for unselected samples of
people.
4. Administer the tests to a sample of people prior to the start of training
and keep their results secret.
5. Wait until all the sample have completed training.
6. Compute the relationship between the test scores and their success or
failure.
7. Allow for range restriction on scores in the selected sample relative to
those unselected (i.e. applicant) sample.
8. Cross validate - get another sample and see how well the prediction
developed from the earlier sample actually works on this new sample.

The key problem in this procedure is: How do we combine the various
test scores (e.g. measures of coordination, general ability, spatial orientation,
and so on) which we obtain from each test taker into a single score such that
the single composite score will provide the best possible prediction of
training outcome in future samples of people. There are three main ways
this is done:

1. Unit weighting. Simply add all the scores together (having converted

42
 Aptitude Testing and Selection in Aviation

them to equivalent scales), thus giving each one an equal weighting.

2. Rational weighting. Multiply each score by a 'weight' which reflects our
judgement about its relative importance. Then add together the weighted
scores. This judgement is based on our knowledge of the properties of
the test and the analysis of the training requirements, together with the
information we have from the wider research literature.
3. Empirical weighting. Use an actuarial, data-driven procedure (multiple
regression or discriminant function analysis) to calculate weights for
each test score. These weights will be optimal for the specific sample on
which they are based.

The last of these works very well if we have a very large amount of data to
work on. Typically, though, we do not, and when we don't there is a serious
drawback: it can produce highly inflated correlations if we have a lot of
different test measures and not many people in our sample. We can correct
for this 'inflation' by a process called 'cross-validation' - seeing how well the
equation developed from one sample of people works when used to predict
outcome for a different sample of people. We will always find that the
validity 'shrinks' when applied to a new sample. The amount of 'shrinkage'
can also be estimated statistically. Recent research has shown that these
statistical estimates are very reliable (Paechter and Bartram, 1990).

The problem o f spurious estimates o f validity

Notions of 'shrinkage' and 'population correlations’ may seem very esoteric

but they have very real implications in cash terms! Assumptions about range
restriction, 'shrinkage' on cross-validation and other technical issues do have
substantial impacts on the correlations people obtain and which we use to
represent the 'validities' of their tests. We frequently see quite large
correlations reported in the literature which are based on data from rather
small samples of people who had been given very large numbers of selection
measures. These correlations should not be taken directly as estimates of
predictive validity.
This misleading use of multiple correlations can be illustrated by two
recent examples in the literature. Eissfeldt (1991) reports a study on the
selection of ATC applicants. He quotes a correlation of 0.60 (an unshrunken
multiple correlation) based on 20 measures and 162 people. The estimated
population correlation for this, however, is much lower: 0.30. Stead (1991)
reports an analysis of QANTAS data using 46 predictor measures with a
sample of 186 people. He quotes validities of around 0.55 for 'best'

43
Aviation Instruction and Training

combinations of these predictors. However, estimates of the population

correlation for these are below 0.10 - well below the point which would
normally be regarded as being 'useful'.
The above examples are not isolated ones: Blinkhom and Johnson (1990)
in a controversial paper expressed similar concerns over the way in which
evidence is obtained and presented for the validity of personality tests in
selection in general. The reporting of uncorrected multiple correlations as
predictive validities is unfortunately very common - taken at face value they
could result in very expensive miscalculations being made.
What you want to know in selection is how well a test will predict the
training outcome for your current and future applicants - not how well
you can use test scores to explain what happened to applicants in the past.

Results o f the MICROPAT validation studies

The full results of the validations studies are summarised elsewhere (Bartram
1991). For the present we will look in detail at just one part of the data. The
results for Royal Navy Operational Flying Training (OFT) outcome for a
sample of 88 men. Figure 3.1 shows the correlations both 'uncorrected' and
after adjustment for effects of selection and shrinkage.

1. The Pscore, which was the measure used in the selection of these people
and is based on the CVT, SMA and INS-B, even after adjustment has a
correlation with training outcome of less than 0.05.
2. The Unit-weighted MICROPAT composite score has a correlation of
about 0.15 while the Rational-weighted one is twice as high at around
0.30. Very similar results were obtained with the Army Air Corps data.
3. Weights which were empirically derived from analysis of the earlier
Army Air Corps sample cross-validate well onto this sample. As we
would expect, though, their ability to predict in the new sample is lower
(0.20 to 0.25) than it was in their original sample (around 0.40).
4. Weights derived from the Royal Navy sample itself, provide the highest
values. After adjustments which allow for cross-validation, a validity of
0.35 is obtained: slightly better than the results for the Rational-
weighted score.

Taking these results and those from the earlier Army Air Corps validation
together, there are some important general conclusions to be drawn:

1. There is no single measure of validity. Depending on how you assess it, it

44
 Aptitude Testing and Selection in Aviation

can be anywhere between about 0.15 and 0.50.

2. The Rational-weighted score is a good general predictor of success in
flying training across training courses (the validity was around 0.30 for
both Army Air Corps and Navy samples, despite considerable differences
in training courses and equipment).
3. With empirical weighting and cross-validation, it may be possible to
produce slightly better course-specific equations (0.35-0.40). However,
these will be less 'robust' in the long run - as they are likely to be more
sensitive to changes in course syllabus, training resources etc.

Figure 3.1
Prediction of operational flying training outcome for 88
Royal Navy pilots

The unadjusted values are 'raw' correlations while the others are
adjusted for effects of range restriction and 'shrinkage'. The AAC
empirical weights show how well weights derived from the Army
Air Corps sample data predict outcome for the Royal Navy
sample.

45
Aviation Instruction and Training

Figure 3.1 is intentionally rather complex. It illustrates how there is no

single value for the validity of a battery of aptitude tests. It also shows how
some estimates of validity will shrink more than others when applied in
selecting people in from future samples: note how the correlations in Figure
3.1 which were based on empirically-derived weights shrink when they are
adjusted for cross-validation.

The value o f validity

What are the implications of a predictive validity of around 0.30? What does
it mean for success rates in training? What is the balance of costs and
benefits? For any base rate and selection ratio, it is possible to calcuate the
change in success rate in training with changes in validity (the following
examples illustrate the principle, but are much simplified in detail).
Assume we have a training course which has a pass rate of 70 per cent and
that the validity of the current selection procedure is 0.10 and the selection
ratio is 10 per cent. With an increase in validity to 0.30 and with no change in
base rate or selection ratio, the pass rate would increase to 81 per cent.
So, if there was an annual training intake of 100 people we would expect
an increase in the number passing from 70 to 81 (on average). This
represents a 37 per cent reduction in the wastage rate (from 30 to 19). There
are various ways to attach values to such improvements. The net value of a
successful trainee can be equated with the per capita training cost, while the
net loss associated with each failure can be estimated to be half the training
cost (this assumes failures occur on average half way through training and
that training costs are uniformly distributed).
Thus, if the full training cost per person were one million pounds, the
utility when validity is 0.10 would be: £70 million - £15 million = £55
million. For a validity of 0.30 it would be: £81 million - £9.5 million = £71.5
million. This is a net increase in utility of £16.5 million per year.
This sort of modelling makes no assumptions about capital investment
costs; savings in plant and personnel required if output is held constant and
the selection ratio decreased; or situations where the net value of the
successful trainee is greater than the training costs. However, while different
assumptions and different cost-accounting procedures produce different
values, the saving in training costs associated with increases in validity of
selection procedures are always orders of magnitude greater than the cost
associated with them.
In the above case, suppose our improved selection procedure costs an
extra £500 per candidate (which is a very high cost for a selection

46
 Aptitude Testing and Selection in Aviation

procedure). If we apply it to half (i.e. 500) of the applicants (which would be

a very large proportion to use such an expensive procedure on), the total
increase in selection cost is still only £250,000 - leaving a net benefit of
£16,250,000.
This sort of example should make clear why it is important not to use over­
inflated estimates of validity when modelling the impact of a new selection
procedure on training. If we had taken the validity of the new procedure to be
0.60 instead of 0.30, the estimated success rate on the course would have
been 94 per cent, and hence the utility would be: £94 million - £3 million -
£0.25 million = £90.75 million. This is an increase in net utility of £35.75
million. More than double the previous estimate. If planning and investment
was based on this figure, you would be in for a serious disappointment!

To maintain validity, increasingly sophisticated methods of pilot

selection will be needed (Photo: Swissair)

Future developments

Of course, we cannot stand still. Aircraft continue to develop and the pilot's
role continues to change. The new generation of fly-by-wire cockpits (like

47
Aviation Instruction and Training

the A320, or the SAAB Gripen) look cleaner and simpler - more like the old
days - but they conceal a mass of complex systems which are beyond the
direct control of humans. Increasingly pilots will need to be selected and
trained as sophisticated systems managers and trouble shooters. Less and
less will they need the 'traditional' flying skills.
As these changes occur, so we will need to develop more and more
sophisticated methods of selection if we are to maintain a reasonable level of
validity. If training costs and aircraft costs continue to increase as they have
in the past, the need for good selection will become even greater than it is
now.
We will also continue to see changes in the types of tests needed and I
believe there will be an increasing need to develop objective measures of
relevant personality characteristics. At present the formal assessment of
personality in pilot selection tends to be based either on self-report
inventories (such as the OPQ or 16PF) or on projective methods. There is
some evidence to show that self-report inventories can provide valid
predictive measures (e.g. Bartram and Dale, 1982; Dale and Bartram, 1985),
but the evidence is very equivocal and self-report measures suffer from
openness to distortion when used as selection tools. The evidence for
projective tests - while they are less 'fakeable' - is even weaker. Harsveld
(1991), for example, has demonstrated that the data produced by the Defence
Mechanisms Test does not form a reliable basis for the prediction of flying
training results. The most consistent positive evidence is for biographical
data inventories and for evidence of interest and motivation relevant to a
career in flying (see Hunter, 1989, for a review).
The most significant advances in this area are likely to come in the
development of task-based performance measures of personality and
motivation. Some of the MICROPAT tests already embody this notion: the
measures of 'error tolerance' obtained from tracking tasks and performance on
the test RISK are all measuring individual differences in style rather than
level of performance. However, we have a long way to go in designing new
tests to cover this important domain.

Conclusions

I hope that this Chapter has provided some indication of the potential benefits
of objective assessment for selection into flying training. There are some
important messages here for anyone who might be considering changes in
their selection procedures:

48
 Aptitude Testing and Selection in Aviation

1. Do not be taken in by test procedures which look good, which impress

pilots by their apparent realism, or which are sold on the basis that other
people are using them - so they must be good. They will only improve
training performance if they have been demonstrated to have predictive
validity across training courses which have relevant features in common
with your own.
2. Look very carefully at claims about validity: Have they been cross-
validated? Are they based on robust estimation procedures? Remember
that a whole range of different validity coefficients can be produced from
the same set of data using different sets of assumptions.
3. Consider the full cost-benefit analysis - do not treat selection and training
as independent cost centres. In general, investment in improving selection
has a better overall return than the same investment in improving training.

While aptitude is essential, it is still only part of the whole picture.

Effective selection procedures need to integrate the assessment of ability and
aptitude with other assessments: physical and medical, temperament, social
skills assessment, team work and so on.

References

Bartram, D. (1987), 'The development of an automated pilot testing system

for pilot selection: the MICROPAT project', Applied Psychology: An
International Review, vol. 36, pp. 279-298.

Bartram, D (1991), 'The development and validation of MICROPAT for

Royal Navy aircrew selection: 1985-199Γ, Ministry o f Defence SP(N)
Report R 163.

Bartram, D. Dale, H.C.A. and Smith, P. (1982), Leconfield Trials o f the

MICROPAT System (Report ERG/Y6536/82/5), Ergonomics Research
Group, University of Hull.

Bartram, D. and Dale, H.C.A. (1982), 'The EPI as a selection test for military
pilots’, Journal o f Occupational Psychology, vol. 55, pp. 287-296.

Bartram, D. and Dale, H.C.A. (1983), MICROPAT Version 3: A

Description of the Fully Automated Personnel Selection System being

49
Aviation Instruction and Training

Developed fo r the Army Air Corps (Report ERG/Y6536/83/7),

Ergonomics Research Group, University of Hull.

Blinkhom, S. and Johnson, C. (1990), 'The insignificance of personality

testing', Nature, vol. 348, pp. 671-672.

Carretta, T.R. (1987), Basic Attributes Test (Bat) System: Development o f an

Automated Test Battery fo r Pilot Selection (AFHRL-TR-87-9),
Manpower and Personnel Division, Air Force Human Resources
Laboratory, Brooks.

Dale, H.C.A. and Bartram, D. (1985), 'Personality and the selection of

military pilots', inSorsa, M. Report o f the XVI Conference o f the Western
European Association fo r Aviation Psychology, Helsinki, pp. 81-89.

Eissfeldt, H. (1991), 'DLR selection of air traffic control applicants:

Predictive validity', in Farmer, E. (ed.), Human Resource Management in
Aviation, Avebury Technical Publishing Group, Aldershot, pp. 37-44.

Harsveld, M. (1991), 'The Defence Mechanisms Test and success in flying

training', in Farmer, E. (ed.), Human Resource Management in Aviation,
Avebury Technical Publishing Group, Aldershot, pp. 51-65.

Hilton, T.F. and Dolgin, D.L. (1991), 'Pilot selection in the military of the
Free World', in Gal, R. and Mangelsdorff, A.D. Handbook o f Military
Psychology, John Wiley, New York, pp. 81-101.

Hunter, D. (1989), 'Aviator selection', in Wiskoff, M.F. and Rampton, G.M.

(eds.), Military Personnel Measurement: Testing, Assignment, Evaluation,
Praeger, New York, pp. 129-167.

Imhoff, D.L. and Levine, J.M. (1981), Perceptual-Motor and Cognitive

Performance Task Battery fo r Pilot Selection (AFHRL-TR-80-27, AD-
A094 317), Manpower and Personnel Division, Air Force Human
Resources Laboratory, Brooks.

Paechter, M. and Bartram, D. (1990), 'Validity and cross-validation in small

samples', Ministry o f Defence SP(N) Report TR 240.

Stead, G. (1991), Ά validation study of the QANTAS pilot selection process’,

50
 Aptitude Testing and Selection in Aviation

in Fanner, E. (ed.), Human Resource Management in Aviation, Avebury

Technical Publishing Group, Aldershot, pp. 3-18.

51
4 The selection and
training of air traffic
controllers in Australia
G. Hannan and R. O ’Hara

This chapter outlines the development of Air Traffic Services in Australia

and then focusses on the training and selection of Air Traffic Controllers.
Issues and developments related to training and selecting practice are
explored and the chapter concludes with a commentary on the future policy
options in this area.

Before air traffic controllers: 'Aeradio' operators

Civil aviation in Australia took a big leap forward after the great air race
from London to Melbourne to celebrate Melbourne’s centenary in 1934. Prior
to this time there was little need for air traffic control (ATC) as there were
few aircraft, vast open spaces, relatively good weather and the tallest
obstacle, Mt Kosciusco, being only 7,311 ft (2,228 mt). However by 1938
there was a significant number of larger aircraft carrying passengers between
captial city centres.
Communications between ground stations and aircraft began in Australia
around 1933. On behalf of the Air Council, Amalgated Wireless of
Australasia, Pty Ltd (AWA), commenced a morse code high-frequency
'Aeradio' service to the small number of aircraft who required that service. In
1934 the first of three ground communication stations were set up in
Melbourne, Launceston and Darwin. Aeradio was the predecessor to today’s
Flight Service. AWA commenced operations with six Aeradio stations. By
the time the Department of Civil Aviation was created in 1938, 12 stations
were available. This number grew to 17 by mid 1939 (Meyer, 1985).
 The Selection and Training o f Air Traffic Controllers

Using the medium of morse code an operator not only had to be proficient
but also repair his HF equipment. For that reason staff were required to hold
a First Class Certificate of Proficiency issued by the Postmaster General's
Department. Meyer (1985) indicates that the functions of the Aeradio
operator were to:

- transmit and receive, by wireless telegraph or telephone, messages to

or from aircraft and other ground stations;
- to operate HF equipment;
- to have a knowledge of the radio and power equipment at his station;
- to effect running repairs and make adjustments as required; and,
- at outstations, where no meteorological officer was stationed, to
prepare and broadcast meteorological reports at regular intervals.

The first air traffic controllers

Like most significant changes in services provided by Government, a major

disaster heralded the introduction of an ATC service in Australia. In 1938 the
crash of the 'Kyeema', a DC-2, on approach to Melbourne from Adelaide in
overcast conditions, resulted in a board of inquiry' (Charlwood, 1981). The
board recommended Government act quickly on improving radio services
and regulating the use of radio. In addition, the board recommended the
establishment of Flight Checking Officers whose functions were to:

- check flight plans prior to lodging;

- check position reports during flights;
- close an airport if conditions warranted; and,
- advise pilots of changing weather conditions.

The Flight Checking Officers were mainly ex-pilots with considerable

flying experience and, in addition to the functions listed above, had the
authority to refuse flight plans if they did not conform to the standards
established by the Department of Civil Aviation (DCA).
By 1945, civilian ATC was using facilities left behind by US Armed
Forces. VHF was available but only in a rudimentary form. ATC initially
used HF communications transmitting on the NDB frequency and receiving
on another. By this time three classes of officer were responsible for ATC
in Australia. These were Flight Checking Officers, Flight Control Officers
(whose responsibility was to log position reports of flights) and Aerodrome

53
Aviation Instruction and Training

Controllers who were located in control towers that Charlwood (1981)

suggested were little better than 'elevated country conveniences'. Area
control was conducted by the Flight Checking Officer. Aircraft enroute to
the major aerodromes passed their positions through Aeradio, the
predecessor to today's Flight Service. Approach Control was not introduced
until 1948.

Controlled air space and flight service

In 1947 Australia introduced a new system of ATC which saw the

introduction of control areas and corridors of air space in which
non-controlled aircraft could not enter. In order to fly in the control areas
aircraft had to be fitted with two way radio.
By the late 1940s Aeradio’s name was changed to Communications
Section, but the functions essentially remained to provide communications
for, and on behalf of, ATC. A restructure of the section in 1966 created the
Flight Service and Aeronautical Telecommunications Section. Flight Service
essentially became responsible for the air-ground communications, the
provision of operational information and search and rescue alerting. The
fixed service (or point-to-point) resided with the Communications Officers
(later to be called Aeronautical Telecommunications Officers (ATOs).
Flight Service Officers are the other major employee group who together
with Air Traffic Controllers (ATCs) form the hub of the Civil Aviation
Authority's air traffic services network. Flight Service provides a broad range
of services to aircraft outside controlled airspace and to those aircraft in
controlled airspace but beyond ATC radio coverage (e.g. international
aircraft crossing the Pacific Ocean). Within Australia, aircraft flying in
controlled airspace are provided with positive separation by ATC, while
within non-controlled airspace, FSOs provides pilots with information with
which they can effect their own separation, as well make other operational
decisions.
Some of the functions conducted by FSO to pilots include provision of:

- an air/ground communication service;

- a briefing and flight plan acceptance service;
- a traffic information service to alert pilots to other air traffic which may
be in close proximity to their aircraft or intended route;
- a search and rescue alerting service; and
- a co-ordination role with ATC and other Flight Service units, regarding

54
 The Selection and Training o f Air Traffic Controllers

the provision of pre-flight, in-flight and post-flight services.

The functions of ATC and Flight Service evolve from a similar base. Both
are information users depending upon information supplied by the pilot
(usually from a flight plan) to effect their duties. While ATCs use radio,
radar and visual sighting and give instructions to pilots with which they must
comply, FSOs provide pilots with information and advice so that the pilot can
make an objective assessment necessary to avoid other aircraft and decide
whether or not the flight should continue.

The recruitment and training of ATC personnel

Recruitment and training

Prior to 1945 the only training available to staff was a four week RAAF ATC
course conducted at Laverton Air Base in Victoria. Between 1945 and 1947
on the job training was utilized. However it was not until Donald George
Anderson (later Sir Donald Anderson, Director General) became
Superintendant of Air Traffic Control that standards were set and staff had to
pass a training course. Because of the inconsistency of standards across the
country, all existing staff were interviewed and a culling of unsatisfactory
operatives took place. In 1948 the International Civil Aviation Organization
(ICAO) was formed, and Australia was expected to conform to international
standards.
Recruitment was not a problem after World War Π. There was a large
pool of ex-pilots and aircrew ready and willing to continue careers in
aviation. In fact the key recruitment criterion in the earlier stages was
aircrew experience. Most recruits were pilots and navigators, although later
in the mid 1950s some master mariners and Aeradio staff were included.
The first civilian ATC school was conducted in Melbourne in August,
1947. The course lasted 3 months and included subjects such as
meteorology, navigation, radio, regulations and aerodrome/en-route control.
Over the next 44 years various locations around Melbourne would be used as
the School for Air Traffic Controller Training. The first course included
existing controllers, the second the hierarchy of ATC and the third senior
people at existing aerodromes.
By the late 1950s the aircrew source for ATC in Australia was
diminishing. England had already introduced a cadetship scheme but this
was delayed in Australia for some four years. Instead, recruitment drives

55
Aviation Instruction and Training

turned to the United Kingdom and the United States. An agreement between
England, Australia and New Zealand required no poaching of each other's
controllers so selection teams targetted aircrew and ex-aircrew in these
countries. Altogether about 250 men and one woman arrived in Australia to
train on a 'short course' and were then assigned to staff various aerodromes
around the country.
Finally, in 1964, Australia introduced its first cadetship scheme, a two year
course which involved a substantial airline attachment to leam aircraft
performance, radio communication and other aircrew functions followed by
theory and practice in ATC. Cadetships provided an opportunity for the first
time for non-aircrew to pursue a career in ATC. Following a rigorous series
of aptitude tests and a medical examination, the trainee could anticipate three
years before the issue of a licence endorsed with a sector rating. Importantly
it provided a school leaver with the chance of direct training into his/her
career choice immediately after matriculation.
However the recruiting of personnel with aviation experience into short
courses continued, as did overseas recruiting. For applicants with
considerable experience as civil pilots, military pilots/navigators, FSOs or
RAAF ATCs, a short-term course in ATC was also available. As with the
'long-term' course, training and entry standards remained high. Aptitude
testing, a medical examination and an interview were required before final
acceptance. Trainees could then expect nine months of high pressure theory
and practice.
As with the long-term course, successful completion of the short-courses
was followed by placement at a major ATC centre for rating and licensing.
However neither short-term nor long-term courses carried any formal
qualification, and on leaving the service, an operator had to start from scratch
to attain a tertiary qualification and retrain for another career.

Selection

With the establishment of ICAO and a growing awareness of the need for
high standards from 1948 onwards, the Department of Civil Aviation made
moves to develop more formal selection procedures. Charlwood (1992)
reported his first involvement in interviewing candidates in 1955. Apart from
a sound knowledge of aviation, candidates were expected to show sharpness
and confidence in order to considered good prospects in ATC. Charlwood
used the following description:

that he should have a very real interest in aviation, usually have a

56
 The Selection and Training o f Air Traffic Controllers

knowledge of basic aeronautical principles, and, perhaps above all ...

demonstrate that he was of a type who leaned towards an operational life
... a 'doer' rather than a dreamer, (p. 90)

However it was not until the late 1950s that aptitude tests were developed
to assist in the selection of ATCs. The responsibility for the development
and implementation of aptitude testing was given to Paul Jones, a psychology
graduate with some years experience in vocational guidance. He was
appointed in 1957 to standardize the pilot licencing examinations, an
appointment which is another example of services being implemented only
after disasters occur. Jones' appointment to this position followed the crash
of a light aircraft outside Melbourne which resulted from the pilot's serious
miscalculation of fuel requirements for the flight. In itself this event may not
seem so momentous except that the pilot had, four days previously, passed
his aircraft fueling examination with 100 per cent.
In 1958, Jones was asked to develop and implement aptitude tests for the
selection of ATCs. After conducting a thorough job analysis, he trialled
many tests including spatial manipulation, clerical speed and accuracy,
mathematical ability, short-term memory and map reading. In addition, tests
of intelligence and personality were trialled in the developmental stage. The
rationale for the selection of tests was based on the descriptions of the tasks
performed by ATCs and the supposed pre-requisite cognitive and affective
abilities. As well, attention was paid to the selection of RAAF pilots and
ATCs in work done by George Naylor (1954), an academic from the
University of Queensland.
In 1959 Jones commenced a full scale validation of the trial battery (Jones,
1962). A system of rating existing controllers (by five supervisors) allowed
the selection of two criterion groups: those controllers rated above average
and those rated below average by all five raters. These two groups were then
used to determine the discrimination ability of the tests. At the end of the the
validation study the following tests were determined to have maximum
potential for selecting suitable candidates to train as ATCs:

- an eight minute speeded test of numerical ability;

- a speed and accuracy test;
- a general intelligence test; and
- a modified 16PF test (42 items that discriminated between the two
criterion groups). (Jones, 1962)

Further work in subsequent years saw modification of this battery. A

57
Aviation Instruction and Training

significant addition was a test of short-term memory which was administered

simultaneously with a map reading test and therefore not only testing
short-term memory but resistance to distraction. The speed and accuracy test
was dropped. This battery was used up until 1984, the year of the last major
recruiting drive conducted by the Civil Aviation Authority (CAA) as it
became known. These selection programs for ATCs (and indeed many other
professional officers employed by the CAA) were conducted by a group
known as the Psychological Services Section, of whom Paul Jones was the
founding officer.
Little research work was conducted on the battery after the initial
validation. In fact the authors have found no published studies prior to 1983.
Two studies since then (Lee and Leary, 1983; Hobbs and Lake, 1985) were
initiated by the CAA as a result of high attrition rates in the ATC training
programs. Hobbs and Lake's study report poor predicability of simulator
performance and only 'acceptable' predicability for ATC theory. They
recommended action be taken to redress this problem, but beyond raising the
selection cut-off points and modifications to the battery, did not offer any
wide ranging strategies.
Two factors could account for the apparent decline in the effectiveness of
the tests to predict performance as reported by Hobbs and Lake. Firstly
Jones' study did not consider simulator performance as a performance
criterion and secondly, the nature of the job would have changed significantly
since the early 1960s when that initial research and validation was
conducted.

Recent developments and their impact on training and selection

From 1984 the CAA (and its predecessors, the Department of Transport and
Communications and the Department of Aviation) pursued the goal of an
integrated Air Traffic Services (ATS) system which would see the
amalgamation of ATC and Flight Service operations. Integration would offer
a staff capable of conducting the full range of functions in a future ATS
system. It was argued that such a structure could have a number of
significant benefits such as :

- a more varied and comprehensive career path could be provided for

ATS staff;
- services could be better tailored to meet the changing requirements of a
burgeoning the air transport industry;

58
 The Selection and Training o f Air Traffic Controllers

- technological change could be better accommodated; and,

- staff could be distributed between ATC and Flight Service to optimize
system operation.

The achievement of an integrated ATC and Flight Service structure did

occur at the management level but not at the operational level. However the
concept had a significant impact on the future direction of training. This will
be discussed shortly.

Recruitment, training and a formal qualification

A number of reviews (Department of Aviation, 1984; Ratner et al., 1987;

Clark, 1987) were undertaken into ATS prior to 1988 which made
recommendations for significant changes to training of ATS officers. In
addition to these reviews, during this period, there was mounting concern
among ATCs and their union body. There was a perception that they were
losing professional standing because of the lack of a formal tertiary
qualification.
Given these two factors the Federal Government called for expressions of
interest from tertiary institutions to establish an ATS training program, in a
'co-operative educational venture' with the CAA, in a tertiary environment. A
Federal Cabinet decision in 1987 gave approval for the establishment of The
Centre for Air Traffic Services at the University of Tasmania at Launceston
(formerly the Tasmanian State Institute of Technology). An AUD $4.5
million building program was commenced in early 1988 and was completed
in time to accommodate the first students in February 1989. In response to
the push for the integration of ATC Flight Service operations, the University
incorporated ATC and Flight Service syllabi with academic strands allowing
students to qualify with a Diploma of Applied Science (Air Traffic Services).
An original premise of the Diploma program was that a broad-based
education would become increasingly important as individuals rose into
supervisoiy and managerial ranks. It was expected that the knowledge,
skills and abilities obtained through completion of tertiary studies would be
valuble to individuals who would be expected to perform in these roles. The
program reflected the CAA's interest in meeting the changing technological
and managerial demands within the organization, with exposure to subjects
ranging from aviation, computing technology, management and education.
This approach was consistent with a view expressed by Tallman who
argued:

59
Aviation Instruction and Training

... problem-solving, the decision-making, the judgement, the

communication skills associated with a university educational programme
ought to significantly benefit the individual as well as companies and the
government. (Tallman, 1990, p. 41)

The framework of the program is shown in Table 4.1 on the following

page. In order to be considered for entry into the course an applicant must
have completed Year 12 schooling and be eligible for normal admission to a
science-based degree course at a tertiary institution in his/her own state.
Selection is competitive and is based upon the results of aptitude tests and
interviews. A medical examination to the standard required for ATS must be
passed before an applicant will be selected for entry.
Students enrolled in the Diploma of Applied Science (Air Traffic Services)
generally have no aviation experience and are paid a CAA indexed
scholarship of approximately AUD $9,000 per year. This is a significant
change to the courses run by the CAA up until this time, in that selected
trainees were considered employees and were paid a base salary far in excess
of the scholarship allowance mentioned above. This development has
resulted in a change in the number and characteristics of the applicant pool,
the most significant differences being a lower number of applicants and a
greater percentage of young school leavers.
To date, 244 students (79 per cent male, 21 per cent female) have
commenced the two year program. At the time of writing three courses (a
total of 95 students) have graduated. Two courses are still in progress.
Ninety-one students have failed or withdrawn resulting in an attrition rate of
37.29 per cent. Most withdrawals occur during or at the end of the first year,
but beyond first year the number of attrites is small. Successful completion
of the University program does not qualify the student to operate as an ATC:
there is still approximately six months of training (with CAA) in a major area
control centre before a licencing can be achieved. However on completion of
the two years, graduates must be able to demonstrate traffic flow
management at a high standard before a Diploma is awarded.
To be able to demonstrate proficiency in traffic flow management students
develop and practice their skills on two ATS simulators. Simulators are used
to model a complete airways system and the aircraft participating in this
system. The first of these facilities, 'basic simulator’, is available to students
commencing the second semester of the program. It provides standard
procedural consoles and the lessons are delivered via a prepared script (of
increasing complexity and difficulty) with fellow students providing input as
'pilots' or adjacent 'sector controllers'. In this role they are known as

60
 The Selection and Training o f Air Traffic Controllers

■blip-drivers'. The students and instructors interact on a one-to-one basis to

ensure that errors are corrected immediately.

Table 4.1
Course structure of the Diploma of Applied Science (ATS)

Year 1:
Semester 1 Semester 2

Fundamental of Aviation Science Radio, Radar and Navaids

Aeronautical Principles Air Navigation 1
Meteorology 1 Meteorology 2
Computing Systems Computing Concepts
ATS1 - Intro, to Air Traffic Services ATS3 - Flight Information Service
ATS2 - Area Control ATS4 - Area Control 2
Field Training 2 - Industry Introduction to Management
Familiarization in Melbourne Simulation 1

Summer School
Field Training 3 - Airline and ATC Centre Attachment - Procedural Control,
Training and Learning

Year 2:
Semester 1 Semester 2

Air Navigation 2 Communication Electronics

Aviation Administration Systems Analysis and Design
(computing)
Programming and Problem ATS7 - Operations
Solving (computing) ATS8 - Radar Control
ATS5 - Aerodrome Approach Simulation 3
ATS6 - Search and Rescue
Simulation 2

Field Training 4 - ATC Centre Attachment - Radar Control

(Source: Tasmanian Institute of Technology submission, 1990)

61
Aviation Instruction and Training

The use of a simulator for training of students in the Diploma of Applied

Science (Air Traffic Services), University of Tasmania at Launceston
(Photos: University of Tasmania).

62
 The Selection and Training o f Air Traffic Controllers

The second facility, a Ferranti computer-driven simulator, provides an

environment which is a reproduction of an area approach control centre
similar to that which may be found at Melbourne, Adelaide or Perth airports.
It incorporates both radar and procedural consoles. Integrated or stand-alone
exercises can be generated and input is provided from 'dedicated blip-drivers'
or from co-ordination received from students operating in adjacent sectors.
Three units of simulation are provided in the program using generic
airspace. Each unit becomes longer and progressively more demanding for
the student. During the second semester, students are exposed to 37.5 hours,
third semester 75 hours and in their final semester, 90 hours of simulation.
Each student has an ATS instructor available to facilitate learning and to
provide verbal and written feed-back. It is during the simulation exercises
and subsequent assessment that most failures in ATS course occur. Students
are permitted one supplementary examination in an ATS unit. Failure to pass
this examination results in the CAA withdrawing a students scholarship
which is a pre-requisite for continuation in the program.
All students operate in parallel (group training), proceeding at the same
rate, but not necessarily learning at the same rate. Students control aircraft
generated by a computer as part of a series of pre-determined exercises. The
computer is not programmed to respond to the skills of each individual.
Consequently it may not be possible for an instructor to gain insight into the
individual's capabilities, and therefore weaker students may be left behind.
As a result of this problem, remedial programs have been developed and are
offered outside the normal scheduled hours of simulation.
Recently the CAA has requested the University to remove the Flight
Service component from the program to reflect the direction CAA is taking in
modification of its airspace and service delivery.

Selection

Selection of applicants for the Diploma course has been based on academic
achievement, performance on the battery of aptitude tests descibed earlier
and interview performance. However the attrition rate remains high.
The need for updating of aptitude tests has been highlighted by Hobbs and
Lake (1985) and more recently by Hannan (1991). The latter explores the
premise on which aptitude testing has been based up until the present. The
articulation of a set of abilities thought to be necessary to perform the ATC
job followed by the development of a set of discrete pencil and paper tests in
a static environment, measuring skills independently of one another, is seen
as a less than adequate process to assess the capability of an individual (who

63
Aviation Instruction and Training

knows nothing about ATC) to perform such a complex task sometime in the
future. It is argued that the availability of micro-computer technology and the
advances in cognitive science should make it possible to create sophisticated
selection tasks incorporating moving environments with both audio and
visual inputs. This type of task should enable far more accurate prediction of
success of trainees than the discrete pencil and paper tests batteries seen to
date. Recent advances in the USA by Embry-Riddle Aeronautical University
(Gibb et al., 1991) and FAA (Aerospace Sciences Inc, 1991) are testimony
to this claim. Research in Australia in this area is now being planned.

The future

Major developments

It could be argued that ATS has evolved gradually in the period from the
1930s. Changes in technology and gradual increases in the demands of
aviation up until the mid 1980s has seen a system shape itself, at a relatively
comfortable pace, to meet the demands placed upon it. However since the
the mid 1980s several developments have provided the impetus for dramatic
change. The deregulation of the financial system and the Government's
agenda of financial self-sufficiency of organizations such as the CAA,
followed by the deregulation of the aviation industry in 1990 has placed
incredible demands on the organization to change rapidly. Coupled with
these factors, the availability of satellite and computer technology has
provided the potential for a radical modernization and rationalization
program that will impact not only on the nature of the ATC job but
necessarily on selection and training programs.
In a sense a new phase in the evolution of ATS in Australia has begun.
The CAA has recently announced its intention to establish a new progressive
ATS system to be achieved by 1995. The major goals are rationalization,
cost savings and staff reductions through efficiency and technological
advances. The major strategy to achieve this system is the development of a
2-Centre concept (Melbourne and Brisbane), from which most ATS functions
will be be performed. This will result in the number of country and remote
Flight Service Units (and possibly Control Towers) being reduced
progressively and will eventually see the phasing out of the FSO role and the
Airways Systems Data Officers (ADSOs). This has been made possible
through technological developments such as the aviation satellite system and
the introduction of major projects such as the National Aeronautical

64
 The Selection and Training of Air Traffic Controllers

Information Processing System, Avfax, Flight Watch, TAAATS and AMATS

procedures.

Implications for training and selection

As part of its centralization program the CAA has agreed to retrain suitable
FSOs and ADSOs as Air Traffic Controllers. This has meant that the intake
of new students for the Diploma of Applied Science program has been
suspended for approximately three years. The job of selection of FSOs and
ADSOs has been contracted to Saville and Holdsworth, a occupational
psychology company, who have based their selection program on a test
battery developed for selection of ATCs in the United Kingdom.
The first FSO conversion course commenced at the University of Tasmania
in January 1992. The 28 week course (preceded by a five weeks attachment
to the CAA) will be available to approximately 260 FSOs. The course is
designed to bring students to a level suitable for licensing as an ATC, subject
to successful completion of a sector rating at a major area control centre.
The conversion course offers six theoretical and practical units, five of which
are based on existing units offered in the Diploma program. Selection testing
for ADSOs is progressing and a conversion course commenced in October
1992.
The educational needs of ATS personnel are probably greater today than
at any other time in Australian aviation history. This is not only because of
the modernization of the Australian airways system but also the Federal
Government’s policy on industry restructuring, career path development and
multi-skilling of the workforce. Modem ATCs will be be required to do a lot
more than perform a aircraft separation function at a console. They will be
expected to be able to provide a wider range of ATS functions as well as
develop skills that will facilitate quality training of new officers and allow
effective movement into management roles.
A profile of the future ATS graduate seems to be a young person capable
of independent thinking and self-sufficiency. This person will require a
sound understanding of technology and its place in ATS as well as an ability
to work co-operatively with others in a team approach. The graduate will
need to be a good communicator, have well developed inter-personal skills,
be assertive, willing to share ideas and above all capable of adapting to and
initiating change. There is wide recognition that the provision of a broad
general education is a better basis for this type of development than
vocationally specific approaches used in the past (Flach, 1991; Tallman,
1990).

65
Aviation Instruction and Training

This expansion of the role and the expectations of any professional will
require a reassessment of selection criteria and training emphases. Until
now, the selection criteria for students have focused primarily on specific
aptitudes. This rather narrow focus ignores the influence of many other
dimensions of an individual's makeup that will influence the capacity to fulfil
the roles alluded to above. Factors such as personality attributes,
communication style and motivational emphasis will need to figure more
prominently in selection criteria if the organization is to meet its goals for the
21st century. The challenge for the CAA and other potential service
providers is to how best to implement such educational programs to achieve
the desired workforce.
To date the training models tried in Australia have been:

- on-the-job training;
- in-house vocational specific training courses; and,
- a broad-based tertiary education course followed by on-the-job training.

Recent argument suggests the latter is inefficient in that the student is

distracted from the learning of ATC operations by apparently irrelevant
course work for which no immediate application can be seen. The trap with
the vocationally specific training courses is that students learn only the
specific job functions and are are not exposed to other sources of knowledge
enhancement and attitude change, necessary for the development of a well
rounded professional. In addition, the students are immediately exposed to
the work culture and attitudes of the prevailing generation of staff, are
significantly influenced by them, therefore lessening the likelihood of new
staff being change agents.
These issues related to course delivery, along with the need to prepare the
individual for a rapidly changing work environment, will have to be aired and
discussed in order to develop training program that will deliver the type of
professional operator and the work culture required in the future.

References

Aeropace Sciences Inc. (1991), Air Traffic Control Specialist Pre-Training

Screen Preliminary Validation, Report prepared for the FAA, Virginia.

Charlwood, D. (1981), Take-Off To Touchdown, AGPS, Sydney.

66
 The Selection and Training o f Air Traffic Controllers

Charlwood, D. (1992), personal communication.

Clark, N. et al. (1987), Review o f Air Traffic Services Training Program,

Department of Transport and Communication, Canberra.

Department of Aviation (1984), A Review o f the Future Role of the Flight

Service Officer in the Provision o f Air Traffic Services, Department of
Aviation, Canberra.

Gibb, G. et al. (1991), The Development of an Experimental Test Battery for

Air Traffic Specialists, Embry-Riddle University, Florida.

Flach, J. (1991), 'The human factor: a positive factor for aviation safety',
Journal ofATC, vol. 33, no. 2.

Hamilton, R.M. (1991), 'The use of simulation in the education of air traffic
service officers at the University of Tasmania', Paper presented to the
ASCILITE Conference, University of Tasmania, Launceston.

Hannan, G J. (1991), 'Selection of air traffic controllers in Australia',

Unpublished Ph.D. degree proposal, University of Newcastle, Australia.

Hobbs, A. and Lake, R. (1985), 'Validation of air traffic control selection

tests', Unpublished paper prepared for the Psychological Services Section,
Department of Aviation, Canberra.

Lee, P. and Leary, K. (1983), 'Analysis of trainees passing and failing

courses at the Department of Aviation Central Training College,
Unpublished paper prepared for the Psychological Services Section,
Department of Aviation, Canberra.

Meyer, R.C. (1985), Aeradio in Australia, AGPS, Canberra.

O'Hara, R.S. (1992), 'Air traffic services attrition', Unpublished MEd.

dissertation, University of Tasmania at Launceston.

Ratner, R. et al. (1987), Review o f the Air Traffic Services System of

Australia, Ratner and Associates, California.

Tallman, K.L. (1990), 'The role of colleges and universities in meeting

67
Aviation Instruction and Training

current and future ATC challenges', Journal ofATC, vol. 32, no. 4.

Tasmanian State Institute of Technology (1990), Submission fo r

Accreditation o f the Diploma o f Applied Science in Air Traffic Services,
TSIT, Launceston.

68
5 Computer-based
testing in flight crew
licensing
G.J.F. Hunt and L.M. Hunt

Developing and maintaining banks of examinations in the personnel

licensing departments of aviation regulatory authorities, or in the training
departments of airline flight operations, is one of the many headaches that
human resource developers have to deal with. The problem is generated
from at least four sources:

1. rapidly changing equipment and operating procedures;

2. the difficulty of attracting well qualified flight crew personnel to
positions in flight examinations;
3. the fact that when such people are recruited, despite their obvious
operational ability, they do not have sophisticated skills in competency
identification, tests and measurement;
4. the trend that the longer they remain as examiners, the better they are at
developing tests, but the more out of touch they become in aviation
systems and procedures. In the end, their knowledge is often quite out
of date.

In today's economic climate, the process of evaluating flight crew

performance is likely to attract fewer, rather than more resources. Many
licensing authorities are recognizing that the task of constructing tests, as
assumed under their traditional examining role, might more appropriately be
delegated to industry. This changed organizational role can lead authorities
to focus more critically on the monitoring and management of the
examination systems for which they are legally responsible. The process of
constructing tests can be devolved to the 'master performers' in the flight
Aviation Instruction and Training

and technical operations of airline and aviation organizations. These

aviation experts can provide the licensing authority with test items and
examinations which are valid representations of contemporary aviation
knowledge and practice. With industry as an integral partner in the
development of tests, examinations can be continuously up-dated as and
when new changes in systems and procedures occur.
For this approach to flight crew testing two issues had to be addressed:
first, the objective and detailed explication of the knowledges and skills to
be assessed; and secondly, a technology for transferring the test items and
examinations from industry experts to a licensing authority. In New
Zealand, the Computer-Based Aviation Licensing and Examination System
(CALES) project, jointly developed by Massey University and the New
Zealand Civil Aviation Authority (Crook and Hunt, 1988) was a means by
which this could be done. The objectives for the development of this system
were as follows, to:

1. determine the validity of the items to be developed;

2. devolve the construction of the individual test items to industry experts;
3. increase the reliability and validity of all items and examinations;
4. provide the end user with access to qualification assessment at the
moment in competency development when he or she is ready;
5. have the examination system fully automated and continuously
accessible.

This chapter reviews some of the considerations which were made in

developing the CALES project.

A basis for defining the flight crew competencies to be measured

Validity

Test validity should be a principal axiom for all examiners. Whenever the
axiom is recognized, it is expressed as some measure of agreement that a
test measures what it purports to measure - usually a prediction for some
identifiable performance. In elaborating this further, the American
Psychological Association (1974) declared that if one was 'to demonstrate
the content validity of a set of test scores, one must show that the behaviours
demonstrated in testing constitute a representative sample of the behaviours
to be exhibited in a desired performance domain'. In applying this statement

70
 Computer-Based Testing in Flight Crew Licensing

to aviation one could assert that in order for licensing examinations to be

valid they would need to be able to demonstrate the relationship between the
items for assessment and the abilities being measured in licence, rating, or
other competency requirements.
In a previous study (Hunt, 1990), six dimensions or accomplishments of
pilot competency were identified and analysed for their construct validity
(Figure 5.1). An accomplishment was defined as the broad functional
capability which contributed to personal or group expertise. Each
accomplishment was seen as a constellation of abilities required by job
incumbents to attain their mission's objective. Using a top down hierarchical
analysis, each accomplishment was analysed in terms of its underlying
performances. In this context, a performance was defined as a statement of
procedural knowledge ('intellectual skill', 'knowing how') that was required
for an accomplishment to be achieved. Each of these performances were in
turn examined to see what specific abilities would need to be inculcated for
the performance to be mastered. The resultant 'map' of pilot abilities
provided a basis for developing construct validity measures.

Figure 5.1
Accomplished pilot abilities

71
Aviation Instruction and Training

Knowledge-process requirements fo r test validity

Tests require some specification of the knowledge assumed to be

prerequisite to the declared capability being measured. In earlier studies of
pilot competency (Hunt, 1990) five functional classes of capability were
identified as being requisite to competency. These were, declarative
knowledge (facts and locating type recognition); procedural knowledge
(knowing how and why); idiosyncratic cognitive strategies; motor skills
and attitudes. In tests of aviation theory declarative and procedural
knowledge (Newall and Simon, 1972; Anderson, 1976) form the mainstay
of test item types. The difference between declarative and procedural
knowledge is important in competency measurement. Knowing rules
implies human performance that is highly precise and predictable. Assessing
declarative knowledge in contrast, restricts the measure to a production of
information which may represent the gist of some idea, the regurgitation of a
fact or the identification of an object or procedure based upon little or no
understanding. Being able to test that a student can locate the air speed
indicator (ASI) on an instrument panel, is no guarantee that he or she can
discriminate between true air speed (TAS) and indicated air speed (IAS),
nor understand under what circumstances each measure should be used.
This deeper ability would require the student to apply procedural
knowledge.
In addition to a specification of knowledge, test developers also need to
know the level of complexity at which the knowledge or skill is to be
processed. Gagne and Glaser (1987) have assumed that different learning
outcomes require different conditions for competency acquisition. Similarly,
if the content of learning (especially declarative and procedural knowledge)
can be specified in terms of facts, concepts, rules and procedures, so can the
thinking processes which are required to transform the information into
expressed outcomes. The degree to which an item matches the expected
knowledge, or content type with the required level of information processing
may be just as important to the efficacy of the test as its purported
congruence with performance constructs. This approach to validity may be
described as the interaction of subject-matter knowledge and task-related
information processing. Such a model, described by Merrill and others
(Merrill, 1983) as a performance-content matrix, has been further modified
by the authors and described as process-content interactions. The model is
illustrated in Figure 5.2.

72
 Computer-Based Testing in Flight Crew Licensing

Figure 5.2
Process-content interactions in test validity

As an example of its application to the analysis of a test item, the process-

content intercept requiring a student to remember a rule (Figure 5.2) is
illustrated in the following item:

Example: Frequent inspections should be made of aircraft exhaust

manifold-type heating systems to minimize the
possibility of:

a. exhaust gases leaking into the cockpit;

b. a power loss due to back pressure in the exhaust system;
c. a cold running engine due to heat withdrawn by the
heater.

Content

The content of a test item is made up of one of four different levels of

information. Facts are the most basic bits of information. These are verbal
or written statements to which the examiner will simply require the student
to name, state, or locate (point to) the object or event:

Example: Which flight instrument in a Piper Warrior (PA28) aircraft,

would provide information about the aircraft's altitude?

73
Aviation Instruction and Training

a. Air Speed Indicator.

b. Vertical Speed Indicator.
c. Altimeter.
d. Tachometer.

Secondly, concepts are groups of objects, events or symbols that all share
some common characteristic. In measuring a concept, an examiner will
require the student to demonstrate that he or she understands the meaning of
the objects, events or symbols.

Example: What is the meaning of the term 'MVFR', as used in the

categorical outlook portion of Terminal and Areas Forecasts?

a. A ceiling less than 1,000 feet, and/or visibility less than 3

miles.
b. A ceiling of 1,000 to 3,000 feet, and/or visibility of 3 to 5
miles.
c. A ceiling of 3,000 to 5,000 feet, and visibility of 5 to 7
miles.

Thirdly, rules are statements of relationships among classes of objects

and events. Rules are used to provide explanations and predictions about
classes of objects and events, and to allow for the classification of examples
of these classes.

Example: A pilot flying a single engine aeroplane observes a multi-

engine aeroplane approaching on a collision course from the
left. Which pilot should give way and why?

a. Each pilot should alter course to the right.

b. The pilot of the single engined aeroplane should give way;
the other plane is to the left.
c. The pilot of the multi-engined aeroplane should give way;
the single engine aeroplane is to its right.

Finally, a cognitive procedure is an ordered sequence of mental steps

that must be worked through in order to achieve an outcome. The behaviour
will often involve the student making some sort of mental calculation or
'thinking through' a series of pre-requisite requirements as part of executing
complex manipulative skills.

74
 Computer-Based Testing in Flight Crew Licensing

Example: What is the approximate base of the cumulus clouds if at

2,000 feet AMSL, the temperature is 20° C and the dewpoint
is 8° C?

a. 3,000 feet AMSL.

b. 4,000 feet AMSL.
c. 5,000 feet AMSL.
d. 6,000 feet AMSL.

Information-processing levels

The thinking process, or information-processing level that a student is

required to access in order to answer a question in a test may be at one of
three levels; remember, use, or hypothesize. Remember requires the
student to recall, or recognize from memory some piece of information. Use
requires the student to apply some previously learned information to a
particular task or context. Hypothesize requires the student to identify
solution options and create new information as a result of the process.

Content-process validity applied to licensing examinations

In a study designed to examine the match between the regulator's

perceptions of the levels of subject-matter content and information
processing that theory examinations required of candidates, and the actual
content and levels of information processing required by the examination, a
random sample of 150 test items was selected from British (CAA),
American (FAA) and New Zealand (NZCAA) Commercial Pilot Licence
item banks (Hunt, 1991). Examiners were asked to identify the percentage
of items in a Commercial Pilot Licence examination which would be found
at each content-process intercept. To do this, they were asked to locate the
percentage of items in an examination which would require a candidate to
remember factual information, use rules, create new rule-based
solutions from given information, and so on. Independently, a group of
senior flight instructors were asked to place the sample items on the
appropriate content-process intercept. The results of this study are
illustrated in Figure 5.3. Regulatory authorities assumed that for
Commercial Pilot Licence examinations, test items would cluster in modal
peaks emphasizing the candidate's need to process subject-matter
information in terms of 'remembering cognitive procedures', 'using rules’,
and 'using cognitive procedures'. In fact, the results proved the items were

75
Aviation Instruction and Training

requiring much lower levels of intellectual effort and were quite different
from what they had expected (c2 = 120.75, r < .001). Instead of requiring
candidates to apply a lot of their knowledge in terms of rules and
procedures, more than half of the items simply required pilots to remember
facts, concepts and rules. This is not what the examiners had expected, nor
what they wanted. The modal peaks occurred at the 'surface' end of the
information-processing continuum, rather than at 'deeper' processing levels
(Biggs, 1991) Test items mainly asked candidates to 'remember facts',
'remember concepts' and 'use procedures'. The examiners' objectives were
valid, but their execution was not. They needed a methodology to translate
'what they knew they wanted to measure' into tools which would do it.

Figure 5.3
Assumed versus actual content-processing levels in
commercial pilot licence examinations

Basic principles in CALES test item construction

The methodology for determining content-process validity is provided within

the Item Creation module of CALES (Figure 5.6). Within this module five
general principles of test construction are explicated:

1. Licensing examination items are generated from a competency

specification. The focus of any instructional testing should be on the

76
 Computer-Based Testing in Flight Crew Licensing

competency specification which defines an accomplishment's

performances and resultant abilities in relation to the context in which
they will be applied (Figure 5.4). Each competency specification is a
description of the interaction between the performance to be taught
(made up of 3 or 4 abilities) and the context to which the performance
will be applied. In a competency specification, the performance is
defined through the specific knowledge structures, and the application
through the task context. The task context provides a description of
the standards and level of information processing that should be applied
to the items assessing the competency.

Figure 5.4
A competency specification
(performance + abilities + task context)

2. Licensing test items should reflect the process-content level o f the

behaviours to be measured. In competency development there are four
content types and three process levels which provide the focus for
licensing examinations. The test items used to measure the
effectiveness of student achievement in that assessment must also

77
Aviation Instruction and Training

reflect the same content types and required levels of information

processing if test validity is to be maintained.
3. Licensing tests should be criterion-referenced measures of
competency. Each performance's ability defines a set of test conditions
in much the same way as they are specified in traditional instructional
objectives. A number of homogeneous test items can be generated
from each ability (Figure 5.5). Reliability indices may be obtained by
comparing the performance of the items which relate to each ability.

Figure 5.5
The generation of test items from competency specifications

4. Instructional tests should be made as reliable as possible. In the Item

Creation module, subject-matter experts are provided with the rules of
procedures for creating reliable test items. In addition, test
administrators have the means for extending a given test to any number

78
 Computer-Based Testing in Flight Crew Licensing

of items (limited only by the size of the item bank, and the amount of
time allocated to the test's administration).
5. Instructional tests should include the types o f test items that are most
appropriate fo r measuring each ability within a competency
specification. Given that any particular performance is made up of two
or three abilities, it follows that each of these abilities will need to be
examined in terms of the most appropriate item type. In the CALES
program the following item types are available to the test developer.

a. Supply-type - student supplies the answer.

Short Answer (word or phrase)
Completion Answer (fill in the blanks)
b. Selection-type (Objective-type) - student selects the answer.
True-False (or alternate response)
Matching Items (up to three options)
Multiple-Choice Items

CALES: a model for computer-based testing

The CALES system is a suite of computer programs designed to enable the

creation, administration and evaluation of a bank of valid and reliable test
items. It has three major modules (Figure 5.6): a series of programs for
teaching, developing and banking test items; programs for generating and
administering tests to students; and programs for reporting on student and
item performance.
The decision to use computer technology in this project was based on
financial, instructional and security considerations. If the new examination
system was to be truly effective, a large part of the aviation industry had to
be involved and therefore trained in measurement procedures. Further, it
would be an on-going exercise, educating new subject-matter specialists and
up-dating the examination item bank. The group of candidates to whom
these new measurement skills were to be taught was large, varied, widely
scattered and growing. The content to be taught was stable and relatively
small. The cost of providing a computer-based lesson which could meet
individual learning requirements, be delivered to a large number of people
over a wide location, within a short time period, and be made available to
industry new comers at almost no additional cost, was minimal when
compared to the cost of providing live instructors to deliver the same
teaching.

79
Aviation Instruction and Training

Figure 5.6
The CALES system defined

ITEM CREATION
How to Write Test Items
Writing and Banking Test Items

TESTING
Generating Tests
Test Practice
Administering Tests

PERFORMANCE REPORTS
Student Performance Report
Item Performance Report

In order to maintain the integrity of the examination item banks, security

of the test items received a high priority. Test items written on paper in
various locations around the country and mailed to the Civil Aviation
Authority for item banking, or tests sent from the central examination centre
to distant locations, were highly susceptible to interception at several stages
in the process. Computer files sent either by modem or floppy disk offered a
greater degree of security, particularly if the files were encrypted.

Principles

Three principles shaped the development of the technology. The first of

these was that all users of the system should have their task made easier, not
more difficult. The final system needed to be as user friendly as possible,
recognizing that many users might not have very developed keyboard skills.
This issue was addressed by minimizing the amount of typing required. Input
to CALES is primarily menu and prompt. For example, test developers
identify the examination they want to write their items for by selecting from
a list. When the system needs information, the users is prompted by a
question. For example: 'How many alternatives in this multiple choice
question?' Test candidates manage the test with single, intrinsically logical
key strokes. 'For example, forward and backward arrows for moving

80
 Computer-Based Testing in Flight Crew Licensing

between test items. 'Changing answers does not require deleting the old
answer, this is done automatically when typing in the new response.

Computer based testing offers a high degree of security

(Photo: Swissair)

The second principle was the use of safety nets as described by Alessi
and Trollip (1991). A safety net is a double-check procedure that prevents
the user from inadvertently doing something irreversible, such as deleting
files. Item developers are repeatedly given the opportunity to check the
correctness of their input and change their responses. Even when the item is
completed, the program gives one final check before the test item is stored
in the appropriate directory, 'Do you want to make this item available for
testing?' The most important safety net for test candidates was ensuring they
didn't exit the test before they had completed. The ESC key is pressed to
indicate that the candidate is ready to exit. They must then press Ύ' or ΤΊ’
to verify their intention. These keys are much more difficult to strike
unthinkingly than, for instance the ENTER key.
The third principle related to the amount of control the user had over the
system. In the case of the item developer this was limited. A set procedure

81
Aviation Instruction and Training

needed to be followed, and if consistency between items was to be

maintained it was important not to deviate from this. However, should the
item developer make a mistake, decide to change the item type or
parameters, they are able to cycle back, exit, or alter an input at any stage.
User control for test-takers was considered even more critical. They
needed to be able to move freely around the test, answer or change answers
whenever they chose, and be able to quickly access specific questions.
CALES uses a number of special function keys to enable the test-taker to
control the program, and a 'status page' to provide test management
information for making control decisions.

Test item creation

Learning to write reliable test items

Before subject-matter experts could start developing test items they needed
to improve their own skills in measurement principles. A computer-based
lesson, lasting about one to two hours was designed to provide a clear
understanding of the basic principles of item creation. It was not intended to
make these people test and measurement experts. However, by following
the rules in the program, they would be able to develop good quality,
reliable test items which did not violate important test principles.

Item writing

The second program allows content experts to write their own test items.
The procedure for writing an item is very straight-forward. First, the
proposed examination, then specific topic within the examination is selected.
The item is given a unique name to avoid later confusion with items written
by other master performers in the same content areas. The program then
prompts the user for the additional information it requires. The number of
answer options in a multiple choice question; whether or not a graphic will
be used; and the correct answer. The item itself is formatted on the screen
exactly as the item developer wants it to be presented to the test candidate.
When the developer accepts the item as complete and correct, it is placed as
a file in the appropriate directory. These files may be transmitted, by modem
or floppy disk, to the civil aviation authority where another program loads
them into the corresponding directories in the central item bank.

82
 Computer-Based Testing in Flight Crew Licensing

Testing

Test generation. Whenever a student is ready for testing, a simple command

can instantly generate the appropriate examination. The examination person
generating the test can specify the desired parameters, namely, the number
of items in the test, the pass mark and the time allowed. The whole
procedure takes about 10 seconds.

Test practice. A demonstration test with explanations is available to

students before the actual test. This explains the test program, procedures
and keys to be used. Candidates can practice test procedures and questions
until they are ready to do the actual test.

Test administration. A major concern for students sitting computer-based

tests was that they should not be in any way disadvantaged, particularly in
comparison with paper-based tests. In fact, the prevailing belief was that
candidates should be able to use computer technology to enhance their test
performance by giving them access to information that would enable them to
better manage their time.
The CALES test provides the same facilities as a paper-based test.
Candidates can scroll through questions, answer or not answer at any time,
change their answers, and tag items (bookmarking) for later attention. In
addition, they can call up a status page (see Figure 5.7) that provides on a
single screen an overview of the test structure and their current responses. It
lists the names/numbers of the items (not the items themselves), indicates
whether or not the student has answered the question (in the case of multiple
choice questions the student's answer is also shown), questions which have
been bookmarked and the remaining time to complete the examination. At a
single key stroke, the student can alter the display to show: unanswered
questions, bookmarked questions, or all questions. Any question can be
accessed directly from this page.

Performance report

Student report. Once the student exits the examination, the results are
available for immediate printed out. The result sheet is in the form of a
certificate that contains all the pertinent information as shown in Figure 5.8.

83
Aviation Instruction and Training

Figure 5.7
Status page in CALES

Figure 5.8
Test certificate

Candidate N a m e :.............................John Smith

Candidate ID: .................................. 159433456778899
Date: ....................................14 August 1992 10.45
Licence Type: ...................................Private Pilot
Examination: ....................................Aircraft/Technical/Knowledge:
....................................Aeroplane
Required Pass M ark: .......................70%
Actual M ark: .....................................75%
Maximum Duration of Test (min.): 180
Actual Duration of Test (m in .):....... 170
Pass/Fail Status: ................................Pass

This information is stored in a different directory to the one containing the

actual test, and is not compiled until the examination supervisor types in a
particular command. This minimizes the possibility of interference test
results.

84
 Computer-Based Testing in Flight Crew Licensing

Item report. One of the advantages of computer-based testing is the

detailed information that can be collected about each item's performance.
Such information provides the key to determining the reliability of the item
and the effectiveness or completeness of the student's training. Data
collected by CALES includes the candidates final response, time spent
responding, and the nature of response changes as shown in Figure 5.9.

Figure 5.9
Item report

Item: ..................................LH120991a
Candidate's last answer:..............C
Time spent on item (sec):............25
Number of times:
viewed:................................ 3
answered right:.................1
answered wrong:...............1
not answered:.................... 1
changed wrong-right: .......1
changed right-wrong: .......0
changed wrong-wrong:.... 0
bookmarked:...................... 1

If the majority of candidates spend a disproportionate length of time

answering a particular item, or consistently getting it wrong, it suggests the
item needs reviewing or discarding (this is especially true if candidates are
drawn from different training programs). An incorrect response to an item
that is normally answered correctly, indicates a weakness in the learning of
that particular candidate.

Conclusion

The CALES program has demonstrated that computer-based testing coupled

with a proper analysis of the required competencies produces both efficient

85
Aviation Instruction and Training

and effective evaluation for pilot licensing. Concern that candidates might
be disadvantaged by the technology has not been proven in this experience.
Further, there is no reason to believe that the technology or knowledge bases
used in the programs could not be applied internationally. CALES may
prove to be a prototype for enhancing the quality and delivery of tests while
reducing the human resource infrastructure for national licensing systems.

References

Alessi, S.M. and Trollip, S.R. (1991), Computer-based Instruction (2nd

ed.), Prentice Hall, Englewood Cliffs.

American Psychological Association (1974), Standards o f Educational and

Psychological Tests, APA, Washington.

Anderson, J.R. (1976), Language, Memory and Thought, Lawrence

Erlbaum Associates, Hillsdale.

Biggs, J.B. (1991), 'Student learning in the context of schools', in Biggs,

J.B. (ed.), Teaching fo r Learning, Australian Council for Educational
Research, Hawthorne.

Crook, C. and Hunt, G.J.F. (1988), Competent Flight Crew Licensing II:
the Terms o f Reference fo r the Hurda Programme, Instructional Systems
Programme, Massey University, Palmerston North.

Gagne, R.M., and Glaser, R. (1987), 'Foundations in learning research', in

Gagne, R.M. (ed.), Instructional Technology: Foundations, Lawrence
Erlbaum Associates, Hillsdale.

Hunt, G.J.F. (1990), 'An abilities based approach to pilot competency and
decision making', in ICAO Human Factors Digest, No. 4, ICAO,
Montreal, pp. A302-A309.

Hunt, G.J.F. (1991), 'Getting test items to measure knowledge at the level of
complexity which licensing authorities desire: another dimension to test
validity', in Jensen, R.S. (ed.), Proceedings o f the Sixth International
Symposium on Aviation Psychology, Vol II, Columbus, Ohio, pp. 1169-
1177.

86
 Computer-Based Testing in Flight Crew Licensing

Merrill, M.D. (1983), 'Component display theory', in Reigeluth, C.M. (ed.),

Instructional Design Theories and Models: An Overview o f their
Current Status, Lawrence Erlbaum Associates, Hillsdale.

Newell, A.A. and Simon, H.A. (1972), Human Problem Solving, Prentice
Hall, Englewood Cliffs.

87
Section 2

Approaches to Pilot
Training

(Photo: Qantas)
6 Introduction to section
2: approaches to pilot
training
[Link]

The six chapters in this section incorporate rationales for training; an

emphasis on human factors; research on the ways pilots leam; transfer of
training from simulators to aircraft; the technology of part-task training;
and the evaluation of training in aviation.
It is appropriate that the chapter by Captain Dan Maurino, Technical
Officer in the Personnel Licensing and Training Division at ICAO
Headquarters, Montreal, introduces the section. As Secretary of the ICAO
Flight Safety and Human Factors Group, he has a comprehensive overview
of international developments. Regular symposia conducted by ICAO have
kept him in touch with approaches to human factors training in general
aviation and the airlines, and at ab initio as well as advanced pilot training.
Since 1989, Maurino has been working on the introduction of the ICAO
Human Factors program.
If the first chapter by Maurino presents a macro-view of pilot training,
Captain John Bent, Manager of Training Policy with Cathay Pacific
Airways, provides the micro-view. Until the end of 1991 Bent was the
project manager for what was termed the Line Instructors Workshop. This
innovatory three day program, an example of what has been termed The
Human Resources Revolution' (Kravetz, 1988), provides all training captains
with instructional education. No recipes are provided, but instructors are
asked to consider their personal instructional style in terms of the criteria
implicit in videos, micro-teaching sessions, practical sessions, analytic
discussions, reading, and a reference manual. The course content covers
verbal and non-verbal communication, objectives and evaluation, briefing
and debriefing, stress and arousal, the psychology of learning, teaching and
learning a skill, fault analysis and feedback, instructional skills, command
Aviation Instruction and Training

training, the second officer, and attitudes and airmanship. Given the success
of Bent's instructor workshop, it is valuable to read about his personal
philosophy of pilot training.
How do pilots learn? The initial response would be 'The same as anyone
else'. That, however, is only true up to a point. For a start, the average age
of pilots puts them in the category of adult education for which androgogy
rather than pedagogy provides guiding principles of learning. Androgogy is
based upon the notions that adult learning is characterized by different self-
concepts; a reservoir of experience as a learning resource; a readiness to
leam; and an internal or intrinsic source of motivation (Jarvis, 1987). This
means that there is a difference in the way the instructor interacts with adult
learners, and the ways in which adult learners perceive the relationship
(Yonge, 1985).
Additionally, the working conditions of pilots impose unusual constraints,
especially mid-career when there may be management and family
responsibilities as well. Professors Phil Moore and Ross Telfer of the
University of Newcastle, Australia, have been researching aviation
instruction and learning for over fifteen years. Their chapter discusses the
development of the Pilot Learning Process Questionnaire (PLPQ). Refined
with the assistance of pilot respondents from Air New Zealand, Air Niugini,
Ansett New Zealand, Cathay Pacific, KLM, Lufthansa, Philippine Airways,
and Swissair, the PLPQ enables pilots to identify their characteristic
approach to learning. Given an explanation of the implications, they are
then able to adapt their approach to suit the nature of the learning task
(King, 1992). Moore and Telfer are currently researching the optimal match
of approach to topic, subject or exercise in pilot training.
'The concept that realistic appearance is necessary to assure that training
received in a simulator will transfer to operational equipment underlies the
design and use of training simulators ...' (Caro, 1988). How necessary is the
emphasis on realism in simulation ? This is one of the questions addressed
by Lintem and McMillan in the fourth chapter of the section. Dr Gavan
Lintem's interest in this area can be traced to his earlier work with Dr
Stanley Roscoe examining visual cues and adaptive training in simulators
(Roscoe, 1986). The work of Lintem and McMillan at the Aviation
Research Laboratory in the Institute of Aviation at the University of Illinois
and the extent of their literature review have enabled them to identify the
gaps in attempts to provide answers to questions about simulator design for
training effectiveness. It has also enabled them to reach an unequivocal
conclusion, supported in practice by training operations with fixed-base
simulators at Swissair, Zurich, th a t'... uncritical adherence to principles of

92
 Introduction to Section 2: Approaches to Pilot Training

fidelity and similarity will not necessarily maximize transfer. Noticeable

departures from physical fidelity or similarity ... do not always lead to a
decrement in transfer'. An interesting scenario for the researcher is the
notion that it may be possible to obtain more effective transfer with a
manipulation involving a training task being less similar than a transfer task
(Lintem, 1991).
This theme is continued in the next chapter by Professor David Allerton,
Head of the Department of Avionics, College of Aeronautics, Cranfield
Institute of Technology. He examines development in computer technology
which apply to the design and use of part-task trainers in aviation.
Supported by other experts (see Trollip and Jensen, 1991), he sees an
increasing use of a variety of low-cost part-task trainers which will take over
some of the training currently undertaken in high-cost simulators.
Regulation will be needed to ensure appropriate integration of these aids into
the training syllabus. The key issue, concludes Allerton, is matching the
technology to the training requirement.
In resolving this issue, the partnership previously emphasized in this book
becomes vital. Professionals and academics need to work closely together.
'... To make the most of the opportunity, a partnership is required not only
between people and machines but among people of different expertise ... as
the tale unfolds in time, we forecast a somewhat tortuous path to outcomes
both salubrious and surprising' (Solomon, Perkins and Globerson, 1991).
Matching and effectiveness are the major issues. Consider the estimate
that Computer Based Training (CBT) could save US$180,000 a year for the
average military educational program, reducing training time by up to thirty
per cent. However, when CBT is combined with programmed textbooks
and conventional instruction, the time saving climbs to around fifty per cent
(Walker, 1990).
The final chapter in the section deals with the vital topic of evaluation of
training. Wing Commander Ken Given, CO of the RAAF School of
Management and Training Technology discusses the Systems Approach to
Training (SAT) used by most air forces. Competency based, SAT depends
upon an evaluation to validate the match of training and job performance.
Evaluation is shown to be a link in the training cycle, or, as he puts i t , '... the
cement that holds the elements of the training cycle together'. After dealing
with the theory and practice of evaluation, he concludes that although the
aviation industry will continue to focus on operations, the use of a
systematic evaluation process is an important investment.
In this section, the six chapters on approaches to training illustrate the
multi-faceted development of aviation instruction. They show, too, the

93
Aviation Instruction and Training

importance of specialist advice when training moves into the areas of CBT,
simulation and part-task training.

References

Caro, P. (1988), 'Flight training and simulation', in Human Factors in

Aviation, Weiner, E.L. and Nagel, D.C. (eds.), Academic Press, San
Diego.

Jarvis, P. (1987), Adult Learning in the Social Context, Croom Helm,

London.

King, A. (1992), 'Comparison of self-questioning, summarizing, and

notetaking-review as strategies for learning from lectures', American
Educational Research Journal, vol. 29, no. 2, pp. 303-323.

Kravetz, D.J. (1988), The Human Resources Revolution - Implementing

Progressive Management Practices fo r Bottom-Line Success, Jossey
Bass, San Francisco.

Lintem, G. (1991), 'Instructional strategies', in Morrison, J.E. (ed.), Training

for Performance .Principles o f Applied Human Learning, Wiley, New
York.

Roscoe, S.N. (1986), Aviation Psychology, Iowa State University Press,

Ames.

Solomon, G., Perkins, D.N. and Globerson, T. (1991), 'Partners in cognition:

extending human intelligence with intelligent technologies', Educational
Researcher, vol. 20, no. 3. p. 8.

Trollip, S.R. and Jensen, R.S. (1991), Human Factors in General Aviation,
Jeppeson Sanderson, Englewood.

Walker, J.Y. (1990), Computer Based Training- An Evaluation of the FA-

18 Hornet Computer Based Training System, Unpublished MEd. Thesis,
Faculty of Education, Monash University.

Yonge, G.D. (1985), 'Androgogy and pedagogy: two ways of

accompaniment.', Adult Education Quarterly, vol. 35, no. 2.

94
7 Human factors
training in aviation
D.E. Maurino

On 3 May 1989, a Piper PAS-23 Aztec had just become airborne from
Runway 16 at Nanaimo, British Columbia, Canada, when the aircraft's nose
baggage door opened due to a failure in the lock. The pilot radioed that he
was returning for an immediate landing. At one and a half miles from the
runway and one minute and thirty seconds into the flight, the aircraft nosed
down and crashed. All five occupants died in the crash, and the aircraft was
destroyed. The Transportation Safety Board of Canada indicated in the
safety action of its accident report:

Although the open baggage door should not have caused insurmountable
control difficulties, the distraction that it created may have diverted the
pilot's attention to a point where he failed to recognise the onset of an
unsafe condition.

The report further elaborates:

... it is now generally agreed that proper training and awareness programs
dealing with the effects of stress and other human factors in performance
has the potential to minimise their consequences and to prevent similar
accidents from occurring.

On 19 July 1989, while cruising at 37,000 feet, united Airlines Flight 232,
a McDonnell Douglas DC-10, suffered a catastrophic engine failure, which
disabled all three of the aircraft's redundant hydraulic flight control systems
and made the aircraft almost uncontrollable. In one of the most outstanding
feats of airmanship and teamwork on record, the crew were able to navigate
Aviation Instruction and Training

the crippled airliner to a crash-landing at Sioux City, Iowa, USA, 45 minutes

after the hydraulic failure. In his own account of the near-miracle, the
captain of Flight 232 stated:

... I am firmly convinced that the best preparation we had is a program

that United Airlines started in 1980 called Command Leadership
Resource management (CLRM) training. It is now referred to as Cockpit
Resource Management (CRM ... I am firmly convinced that CRM played
a very important part in our landing at Sioux City with any chance of
survival.

These accidents may well represent extremes of the aviation system in

terms of the complexity of the respective aircraft and operation environments
involved. They also present contrasting elements in the seriousness of the
triggering emergencies and - most notably - their outcomes. Yet they
present two common aspects: firstly, the importance of Human Factors
training (or its absence thereof) in aviation operations; and secondly, both
accidents were generated by people removed in time and space from the
events. In the first case, by deficient maintenance and managerial practices;
in the second, by deficient certification and quality control procedures. Both
events were bred by existing latent system deficiencies: the nose baggage
door lock and the engine/hydraulic failures merely triggered them.
The implications of the foregoing are two-fold: aviation Human Factors
training has a niche in operational personnel training curricula, with a clear
impact in aviation safety and effectiveness. Moreover, aviation Human
Factors training should not be limited to the operational personnel (pilots,
controllers, dispatchers, mechanics, etc.) directly involved in the front line of
day-to-day operations. If aviation safety is to be advanced, Human Factors
training must broaden its audience to include training for managers and
decision-makers in the aviation system.
In this chapter I consider Human Factors knowledge and skill training for
operational personnel and anticipate the future integration of technical and
Human factors training, as well as a growing role for management training.

Background

Human error has long been the predominant factor in aviation accidents and
incidents. For as long as human beings continue to be a central part of
aviation, human capabilities and limitations will impact upon safety. It has

96
 Human Factors Training in Aviation

been unclear in the past to what extent detrimental aspects of human

performance could be addressed by training courses and activities. On die
other hand, it has been equally clear that Human Factors education and
training within aviation has been deficient.
In respect of technical training for pilots, there is agreement on the
training required and appropriate syllabus content. A similar consensus is
only starting to emerge in regards to Human Factors training. This reflects a
number of important influences, including the lack of a widely accepted
definition of Human Factors, different beliefs as to the proper means of
controlling human error in aviation, as well as contrasting views as to the
value of training in Human Factors.
Aviation Human Factors covers a wide field of endeavour, with
overlapping disciplines and subject matter (see Figure 7.1). There are also
intellectual, educational, cultural, organizational and individual biases
associated with each different approach to the subject. For instance, while
some specialists think that basic education is the key to progress, others
emphasize the critical role of equipment design, and yet others centre
activities in developing Human Factors skills. The latter group, for example,
would challenge the usefulness of training directed to the mere acquisition of
knowledge.
This translates into different approaches to the degree of application of
various disciplines, namely medicine, ergonomics, and the behavioural
sciences in seeking solutions to Human Factors issues.
Given this lack of an international consensus it is not surprising that, over
the years, there have been contrasting beliefs throughout the industry as to
the practicality and usefulness of Human Factors training for aviation
personnel. This has perhaps contributed to a state of affairs in which, for
want of an agreed way forward, little concerted action was taken.

Human Factors training in aviation: historical perspective

Aviation Human Factors training is a recent endeavour, and barely ten years
ago many involved in aviation would argue that there was no room for such
training in the industry. Although isolated efforts can be traced back to the
late sixties, it was only in the eighties that such training started to gain broad
acceptance across the international community. The partnership of Captain
Frank Hawkins and Professor Edwards led to the pioneering development by
KLM of its Human Factors Awareness course (KHUFAC) in the early
seventies. They were complemented by the research and

97
 Discipline Definition Specific area of interest Typical area of application

Psychology The science of mind and behaviour. Sensory characteristics, perceptual laws, Display requirements and design, control
learning principles, information systems design, allocation of function,
processing, motivation, emotion, research training system requirements and
methods, psychomotor skills, human methods, selection methods, effects of
errors. emotional and environmental stress on
performance, simulation requirements.
Engineering Applying the properties of matter and the Hydraulics, mechanical, structural, Design of displays, design of controls,
sources of energy in nature to the uses of electrical, electronic, and aerodynamics design of control systems, design of
man. design, systems analysis, simulation, complex systems, design of optical
optics. systems, simulator design.
Human physiology Deals with the processes, activities and Cell structure and chemistry, organ Environmental systems, diet and nutrition,
phenomena characteristic of living matter, structure and chemistry, interaction ofthe effects of environmental factors (heat,
particularly appropriate to healthy or various body constituents to promote cold, hypoxia), establishment of
normal functioning. health and function, functions and environmental requirements.
requirements of body systems.
98

Medicine The science and art of preventing, Effects of various forces, radiation, Toxicology of smoke, chemicals, impact
alleviating or curing disease and injuries. chemical and disease agents; appropriate protection, maintenance of health.
preventive methods of protecting health
and well-being.
Sociology The study of the development, structure Small and large groups or “ teams” , crew Crew selection, passenger safety,
and function of human groups. composition, behaviour of passengers in
emergency situations.
Anthropometry Study of human body sizes and muscle Anatomy, biodynamics, kinesiology. Ground support equipment, access door
strength. size for maintenance, work station layout
tre an h ra n n e n f a d ju s tm e n t n f s e a ts e tn i

Figure 7.1
Disciplines frequently involved in Human Factors activities

(Reprinted from ICAO Human Factors Digest No. 1: 'Fundamental Human Factors
Concepts', (Circular 216-AN/131))
 Human Factors Training in Aviation

development undertaken by Ruffel-Smith, Billings, Helmreich, Lauber,

Foushee and others at NASA in the USA in the late seventies.
Although the philosophical and cultural approaches differed, both efforts
were seminal in recognizing that there is much more than pilot error to
aviation safety, and each represented a much-needed step forward.
Gradually, the notion that the development of Human Factors skills was as
important as the development of technical skills gained acceptance within
the operational and training community. This led, starting in the early
eighties, to a gradual proliferation of Human Factors training programs,
primarily directed towards airline operations. A controversy which extends
into the present was generated in the early days: there is a difference
between knowing how to do things and actually doing them. For some this
translated into a dichotomy: should Aviation Human Factors training address
the development of knowledge or skills? This chapter considers the
provision of both Human Factors knowledge and skills.

The role of ICAO

The International Civil Aviation Organization (ICAO) establishes Standards

and Recommended Practices (SARPs) which govern international civil
aviation. This is accomplished through 18 Annexes to the ICAO Convention
of 1944, covering the different fields of civil air navigation: communications,
meteorology, air traffic services, operations, security, etc. ICAO Annex I
outlines the licensing requirements for flight crews and other operational
personnel. In 1988, ICAO Annex I was amended to reflect ( among other
issues) the need for Human Factors training for pilots.
The 8th Edition of Annex I now requires each applicant for a flight
crew licence to demonstrate appropriate knowledge regarding human
performance and limitations relevant to ... (the licence being issued).' Annex
I also includes a requirement for the demonstration of certain skills, in a
manner which dictates increased attention to particular aspects of human
performance. Thus, for example, the holder of an Airplane Transport Pilot
Licence shall demonstrate the ability '... to ... exercise good judgement and
airmanship; understand and apply crew co-ordination and incapacitation
procedures; and communicate effectively with other flight crew members.'
Similar requirements will be established in the near future for operational
personnel other than flight crew. The Annex I provisions create a
requirement for training in both Human Factors knowledge and skills. It
may be naive to believe that regulations can resolve the controversy, but at

99
Aviation Instruction and Training

least it provides the required framework which reflects the consensus view
of the international community. There is now need to develop suitable
Human Factors training for operational personnel and to look to the
improvement of Human Factors skills in operational contexts.
In 1988, ICAO established a Flight Safety and Human Factors Study
Group. This group is involved in the development of new Human Factors
training materials. Since 1991, ICAO also convenes semi-annual Human
Factors regional seminars, designed to promote the dissemination of Human
Factors knowledge.

Human Factors in training programs

The knowledge requirement

Given the predominant role of human failure in the accident record, it is no

surprise that the consequences of human deficiencies have been well
identified in Accident Reports and other publications. Licensing
requirements and the design of equipment, operational procedures and
training - including Human Factors training - are amongst those elements of
the aviation system which have changed steadily as a result of such
experience.
In retrospect, the KHUFAC program can be viewed as a milestone in
aviation Human Factors training. KHUFAC was the first effort to seek
solutions to safety deficiencies through formal training in aspects of human
performance. Subsequent Human Factors courses have built upon KHUFAC.
Table 6.1 identifies specific areas of knowledge included in contemporary
aviation Human Factors programs. The proposed training is directed
towards the prospective Airline Transport Pilot Licence (ATPL) holder.
With minor adjustments, it can be applied to training for the Commercial
Pilot Licence (CPL), the Instructor Rating and the Private Pilot Licence
(PPL). The curricula for the ATPL and Instructor Rating should include the
aspects of crew co-ordination, communications and small group dynamics.
The curricula for PPL and CPL should emphasize individual pilot judgement
or decision making.
It was originally envisaged that thirty-five hours would be required for
Human Factors training programs similar to that proposed in Table 7.1. In
practice, it appears that effective training can be achieved in about twenty-
five hours. This figure is based upon experience in the United Kingdom,
which was first to implement the Annex I requirements. A breakdown of the

100
 Human Factors Training in Aviation

recommended percentages of the total teaching time by topic has also been
provided in Table 7.1.

Table 7.1
ATPL training course in Human Factors

Module 1 Introduction 5% (1.75 hr)

Module 2 The Human (Physiology) 20% (7.00 hr)
Module 3 The Human (Psychology) 30% (10.50 hr)
Module 4 The Human (Fitness) 5% (1.75 hr)
Module 5 Pilot: Equipment 5% (1.75 hr)
Module 6 Pilot: Software 10% (3.50 hr)
Module 7 Interpersonal Relations 15% (5.25 hr)
Module 8 Operating Environment 10% (3.50 hr)

100% (35.00 hr)

Thirty-five hour pilot training course in Human Factors. It is

widely agreed that training for the ATPL and instructor rating
should explicitly address issues relating to crew communication,
coordination and management (Reprinted from ICAO Human
Factors Digest, No. 3. Training of Operational Personal in
Human Factors (Circular 227 - AN/136)).

Course content

Module 1 provides the rational for Human Factors training - considering that
three out of four accidents have had at least one contributory factor relating
to human performance. The relevance of the program is clarified, including
the fact that it is intended to be practical rather than academic training. An
operational orientation to the training course is strongly recommended.
Module 2 includes subjects such as hypoxia and hyperventilation,
pressure effects; limitations of the senses; acceleration effects;
disorientation; fatigue/alertness; sleep disturbances and deficits; circadian
rhythm and jet lag.
Module 3 discusses human error and reliability; workload; information

101
Aviation Instruction and Training

processing; attitudinal factors; perceptual and situational awareness;

judgement, decision-making and stress.
Module 4 addresses personal health; effects of diet/nutrition, alcohol,
drugs (including nicotine and caffeine), medications and blood donations;
effects of ageing; psychological fitness, stress management and pregnancy.
Module 5 explains basic concepts of controls and displays design, alerting
and warning systems; personal comfort; cockpit visibility, eye reference
position and motor workload.
Module 6 discusses standard operating procedures (rationale and
benefits); written materials and software (maps and charts); checklists and
manuals) and the operational aspects of automation.
Module 7 considers factors influencing verbal and non-verbal
communication with and between flight deck crew, cabin crew, maintenance
personnel, company management/flight operations control, air traffic
services and passengers; how verbal and non-verbal communication affects
information transfer and thus safety and efficiency of flight; crew problem
solving and decision-making and an introduction to small group dynamics/
crew management.
Module 8 discusses the internal physical environment (temperature,
pressure, humidity, noise, lighting, vibration, radiation and pollutants/
contaminants); the external physical environment (terrain, weather, other air
traffic); and the socio-economic environment.
Whatever the total amount of hours allocated to any given program, a
balanced introduction to Human Factors training should be attained if the
presented relative percentages are applied overall.

The skill requirement

The initial emphasis in aviation Human Factors training should be upon

knowledge and comprehension of basic Human factors concepts. However,
the need to develop appropriate operational behaviour and skills is of the
greatest importance. Training activities directed towards the development of
suitable skills should always be given the highest possible priority; they
should be offered, as part of an overall training package, from the beginning
of any operational career.
The most successful and widely implemented aviation Human Factors
skills development program to date is known as Crew Resource
Management (CRM). Although originally developed for the airline industry,
it has been successfully applied to the corporate and general aviation
communities. Latest developments of the concept will be applied to the air

102
 Human Factors Training in Aviation

traffic control and to the maintenance environments. An adapted extension

of the concept has been implemented by one civil aviation administration
(Transport Canada) and considered by others (Australia and the United
States) for pilot licensing purposes.
CRM (Figure 7.2) refers to the effective use of all resources available to a
crew or team to achieve safe and efficient aviation operations. At its
foundation, CRM deals with decision-making concepts in the context of
aeronautical operations and multi-crew environments, with a focus on
interpersonal communications.
Typical CRM programs demonstrate that decision-making skills required
in aviation can be developed through systematic training. CRM aims to
develop attitudes and seeks to enhance crew decision-making. A primary
consideration in CRM training is the relative emphasis given to both
relationships with other people, and the need to achieve the operational task
at hand. Since interpersonal communication is the means by which
resources in the work environment are managed, aspects of inter-personal
communications are a primary element in CRM training. Leadership, the
team concept and stress management are also addressed.
CRM programs promote understanding as well as provide a framework
within which to develop basic Human Factors skills. As with any other
skills, they must be practised on a periodic basis. The optimal vehicle for the
recurrent practice of Human factors skill is an instructional technology
known as Line Oriented Right Training (LOFT). LOFT uses high-
technology, real-life simulation to duplicate the operational environment for
flight crew training purposes. CRM and LOFT must be considered
inseparable. Figure 7.3 outlines a typical LOFT scenario.
Virtually every major airline in North America, the Pacific and South-east
Asia basin provide CRM training for their flight crews. Some airlines have
recently incorporated cabin crews into this training. Although the issue is
beyond the scope of this chapter, many contend that CRM is not a good
traveller, since its effectiveness is affected by social and cultural
preferences. In Europe and, partly, South America, most major airlines
provide Human Factors skills development programs derived from CRM,
though not having that name. LOFT, on the other hand, has been universally
accepted. Even airlines which have not instituted formal CRM training have
adopted LOFT for their simulator training.
The Human Factors skill development program developed and
implemented by Transport Canada is mostly directed towards general
aviation users operating in a single-pilot environment. Known as the Pilot
Decision Making Program (PDM), it was also originally developed by the

103
Aviation Instruction and Training

Figure 7.2
Domains for CRM training

(Reprinted from ICAO H u m an F a c to r s D ig e s t No. 2: 'Flight Crew Training - Cockpit

Resource Management (CRM) and Line-Oriented Flight Training (LOFT)', (Circular
217-AN/132))

104
105

Figure 7.3
LOFT simulation scenario

(Reprinted from The Impact of Cockpit Automation on Crew Coordination and

Communication, NASA Contractor Report - 177587, November, 1991)
Aviation Instruction and Training

US Federal Aviation Administration (FAA), and consists of the following

five modules:

1. PDM I is the foundation of good decision-making. It identifies pilot

attitudes towards risk-taking, shows how stress affects decision-making
and gives and approach to good decision-making.
2. PDM Π is a risk management program that applies the principle of
assertive behaviour in the cockpit.
3. PDM ΙΠ applies the PDM techniques to each phase of flight (preflight
planning, enroute, approach and landing).
4. PDM IV deals with basic communication and shows that good
information is the predecessor of good decision-making.
5. PDM V deals with safety awareness for flight instructors. This module
is meant to teach flight instructors how to deliver PDM to their
students.

The PDM program has been highly successful with the Canadian general
and corporate aviation community. The program is a good example that
motion and visual simulators, video recorders or other expensive training
tools are not a sine qua non for the successful development and
implementation of Human Factors skills development programs. It also
supports the case for the development and implementation of similar courses
elsewhere.
It is anticipated that training directed towards the acquisition of Human
Skills will be further developed in the coming years and that suitable training
techniques will be fully integrated into all pilot training activities. It is
especially undesirable that Human Factors training comes to be seen as an
academic exercise which lacks operational relevance. The development of
skills for practical application during flight operations is an important
evolution from theoretical Human Factors knowledge to actual operational
settings. Instruction directed as the acquisition of Human Factors skills is
the activity which is expected to yield the greatest benefits in the future.

Aviation Human Factors training and management

Although there will always be room for improvement in aviation safety

through Human Factors training for operational personnel, recent trends
indicate that the focus must broaden to include those responsible for
managing different parts of the overall aviation system.

106
 Human Factors Training in Aviation

Clearly, the ultimate responsibility for safety in any organization rests

with management. Manufacturers' management is responsible for the design
and manufacture of equipment. Operators' management determines
equipment acquisition. They then select and train the personnel who will
operate and maintain that equipment. State administrations' management
are responsible for establishing standards and criteria which define a frame
of reference within which operators and manufacturers conduct their
activities.
In addition, management is responsible for the organizational design and
functioning; management attitudes are very important factors is defining
organizational culture. Organizational design and corporate culture exert
powerful influences upon overall system safety and effectiveness. In the
future we can anticipate specific training for high-level decision-makers in
the Human Factors aspects of management and organization of high risk
industries.

Human Factors training and the organizational context

The model in Figure 7.4 integrates the overall training task and sets Human
Factors training within an organizational context. The first training step
involves the development of knowledge about the equipment (machines)
which are controlled by operational personnel, as well as about human (crew
member) performance. This is accomplished through the conventional
teaching of knowledge. The interface between equipment and operational
personnel necessitates the development of applied skills, which are acquired
through training (for example, in a flight simulator). Human Factors skills
must also be developed so that operational personnel can function in a multi­
crew context in a manner which enhances safety and efficiency. When
personnel have acquired the required knowledge and developed appropriate
skills, they can be regarded as proficient.
To translate this proficiency into day-to-day performance, operational
personnel must manifest attitudes consistent with corporate and training
objectives, so that operational behaviour is consistent with the overall safety
objectives of the organization. Operational personnel do not act in a
vacuum, but mirror the policies and practices of the organizations to which
they belong. When knowledge, skills and attitudes are combined, a
proficient individual can be considered a professional.
Organizational norms and values, as created and maintained by
management, may either support and sustain the training and development

107
108

Figure 7.4
Human Factors: the training task and the organizational context

(From Johnston, 1991)

 Human Factors Training in Aviation

task, or detract from it. An important point established by recent research

into the effectiveness of CRM programs is that successful implementation is
critically dependent upon active and visible support from the highest levels
of management. In the absence of such support, implementation will be
difficult and training dividends will be meagre.

The future

Building professional attitudes

Research supports the contention that there exists a strong relationship

between individual attitudes and professional performance. Attitudes are
significant predictors of behaviour. The importance of attitudinal factors
related to team or individual performance has been recognized by Human
Factors practitioners for some time. Although addressed by present Human
Factors training courses, the design of courses which effectively foster the
development of professional attitudes remains a challenge for designers of
the next generation of Human Factors courses. Desirable professional
attributes and dispositions conducive to good airmanship are among the
important considerations for this task. The piecemeal development of
professional attitudes may prove to be an insurmountable challenge.
Integrating attitudinal training into the overall training process remains
another big challenge for the nineties.

Integrating technical and Human Factors training

There is one common denominator among most Human Factors training

courses in aviation: they stand alone. They stand apart from technical
training, whether at ab initio level, or within an airline training system.
Human Factors training courses may range from a few hours to several days,
according to the context in which they are provided. This stand-alone nature
may create the undesirable situation whereby operational personnel may
perceive an artificial differentiation between technical and Human factors
training.
Desirably, the first step during transition training for a new aircraft type,
for example, would be to introduce and practice specific Human Factors
skills. Mastery of systems knowledge, normal and abnormal procedures
would be acquired as an integrated part of the overall training. Lastly,
through the use of carefully designed operational training scenarios, systems

109
Aviation Instruction and Training

and procedures would be integrated with those practised Human Factors

skills. The training tools required for this approach are readily available:
computer-based training tutorials, part-task trainers, table-top trainers or
fixed-base simulators.
Integrated Human Factors and technical/procedural training is still in its
infancy. The SPECTRUM training program of the University of North
Dakota (USA), along with the training programs at Massey University (New
Zealand) and the University of Newcastle (Australia) are examples of
highly-integrated ab initio pilot training courses. The SAS Training
Academy (Sweden) and Aer Lingus (Ireland) both have specialist bridge
courses for ab initio pilots transitioning directly to airline operations. Each
course endeavours to integrate Human Factors, CRM and crew co­
ordination training along with traditional airline technical and entry training.
The Boeing Company is considering a similar approach for transition
training to its next generation airliner, the Boeing 777.

Conclusions

The aviation industry has come a long way since the days in which Human
Factors was regarded as a branch of aviation medicine and the flight surgeon
was the only practitioner versed in the field. Gone, also, are the days in
which Human Factors training programs were comprised exclusively of
physiology and, perhaps, equipment design. A broader, systems approach to
the understanding and application of technology of Human Factors has come
of age. Human Factors is regarded today as the last major frontier in the
aviation system. The custodians of this frontier are managers and
operational personnel alike, the former the generals and the latter the ranks.
In properly training these custodians on Human Factors issues as they affect
system safety and effectiveness, the international aviation community should
rest assured that the frontier will remain properly manned.

Acknowledgements

I would like to express my deep appreciation to my colleague Captain Neil

Johnston for his invaluable advice in defining the orientation and contents of
this chapter, as well as for his well-informed and detailed review of its draft.
I am also obliged to Paul Lamy for his review and comments on the early
draft of the chapter.

110
 Human Factors Training in Aviation

References

Boeing Co. (1990), 'Cockpit management techniques', in Flight Operations

Safety Seminar, Boeing 737.

Bruggink, G.M. (1988), 'Reflections on air carrier safety', The Log, April,
pp. 11-15.

Haynes, A.C. (1991), 'United 232: Coping with the "one-in-a-billion" loss of
all flight controls', Flight Safety Foundation Accident Prevention Bulletin,
June.

Helmreich, R.L. (1990), 'How effective is Cockpit Resource Management

training?', Flight Safety Digest, May.

Hunt, G.J.F. (1990), An Abilities-Based Approach to Pilot Competency and

Decision Making, Paper presented to the ICAO Seminar on Human
Factors, Leningrad, 3-7 April.

ICAO (1984), Accident Prevention Manual, Doc. 9422-AN/923.

ICAO (1988), Annex I - Personnel Licensing, (8th ed.), July, 1988,

applicable 16 November 1989.

ICAO (1990), Flight crew training: Cockpit Resource Management (CRM)

and Line-Oriented Flight Training (LOFT)’, Human Factors Digest No. 2,
Circular 217-AN/132

ICAO (1991), 'Training of operational personnel in Human Factors', Human

Factors Digest No.3, Circular 227-AN/136.

Jensen, R.S. (1989), Aeronautical Decision Making - Cockpit Resource

Management, DOT/FAA/PM-86/46.

Johnston, A.N. (1986), Organizational and Motivational Aspects o f Air

Carrier Pilot Decision Making, Transport Canada PDM Conference,
Ottawa, Canada.

Johnston, A.N. (1992), CRM - Cross-Cultural Perspectives, (in press).

Ill
Aviation Instruction and Training

Johnston, A.N. and Maurino, D.E. (1990), Human Factors training for
aviation personnel', ICAO Journal, May, pp. 16-19.

Lauber, J.K. and Foushee, H.C. (eds.) (1981), Guidelines for Line-Oriented
Flight Training, Vol I and II, NASA Conference Publication 2184, Ames
Research Center, Moffet Field.

Maurino, D.E. (1991), 'Management and safety', ICAO Journal, October,

pp. 6-9.

Maurino, D.E. (1991),' Education is the key to progress in human factors',

ISASI Forum, (in press).

Maurino, D.E. (1991), Human factors aspects of management and

organization’, ICAO Journal, (in press).

National Transportation Safety Board, (1990), Aircraft Accident Report,

United Airlines Inc., DC-10, Report No. AAR-90-6, Washington.

Orlady, H. and Foushee, H.C. (eds.) (1986), Cockpit Resource Management

Training - Proceedings o f a NASAIMAC Workshop, NASA CP 2455, San
Francisco.

Reason, J. (1990), Human Error, Cambridge University Press, Cambridge.

Ruffell-Smith, H.P. (1979), A Simulator Study o f the Interaction o f Pilot

Workload with Errors, Vigilance and Decisions, NASA TM-78482,
Moffet Field.

Transport Canada (1988), Pilot Decision Making Programme (Modules I to

IV), Ottawa, Ontario.

Transportation Safety Board of Canada (1991), Aviation Occurrence Report

Number A89P0105, Hull, Quebec.

Vandermark, M.J. (1991), 'Should flight attendant be included in CRM

training? A discussion of a major air carrier's approach to total crew
training', The International Journal o f Aviation Psychology, vol. 1, no. 1,
pp. 86-93.

112
 Human Factors Training in Aviation

Wiener, E.L. (1990), 'The golden era of aviation human factors in

commercial aviation', Human Factors Society Bulletin, vol. 33, no. 8,
August, pp. 1-2.

113
8 Human factors in
training
J. Bent

The past

People have flown productively for less than one century out of their
evolution of many millions of years. During this amazingly brief period, they
have engineered a rapid technical evolution leading to commercial flight and
spaceflight. From their first use of tools, people have become progressively
more dependent on them. Recently their technology has evolved much faster
than their physiology, which is still estimated to evolve at about 1 per cent
per one million years! In fact, technical evolution has constantly accelerated
to a point where change is now exponential, driven by rapid advances in
computer science. A graph of technological change, extrapolated into the
future, would be likely to show that many science fiction plots of today are
unimaginative. The remarkable demands of change made on the aviator
during this century have eclipsed almost all humanity's previous experience,
and now, as the turn of the century approaches, new challenges are emerging
to further test its adapability.
So, confining ourselves to the century of aviation, how have aviators
coped? It is my contention that they have coped partly through the
establishment of rigid procedural systems within this changing environment.
Their overriding need for the familiar and the comfortable seems to have
prevailed. Systems tended to provide a protective shield against the changes
surrounding the aviator, instead of adaptive mechanisms for coping with this
change. Such a response is not unnatural for humans, trapped as they are in a
sluggish time frame of biological evolution. Only with technology have they
ventured outside this physiological straight jacket, inventing increasingly
brilliant tools. Hampering this process in earlier times, they had sometimes
 Human Factors in Training

been punished by their peers for new ideas which were too far ahead of the
acceptable 'group-think' of the society of the day. Now, they have become so
engrossed in technical achievements that they have largely ignored their own
biological limitations, becoming increasingly less able to cope with inventions
at a human level.
In the early days of the development of air transport, regulators set
stringent controls on the performance of pilots. This represented an early
system of defence against human error. Aircraft designers did not design
machines which were particularly 'user-friendly', or 'human-error-tolerant', but
instead expected the new gladiators of the sky to cope. If human error was
found routinely or under test, pilots were considered to have failed, and could
lose their license to fly. When accidents occurred, the cause was usually
found to be pilot error, and that was that. Those who apportioned the blame
did so in committee, with many days or months to consider decisions taken in
a mere split second by the aviator concerned. The marvellous advantage of
hindsight reinforced the tendency for a 'pilot error' verdict. Designers and
engineers were, for a very long time, less accountable for accidents than
those who flew their machines.
One consequence of this was that aviators were threatened with a standard
to maintain at all costs, for fear of losing their livelihood. Such motivation
was considered appropriate for the times. Although times have changed, the
authoritarian assumptions underlying this motivation live on today in the
management of some airlines. Standards were set, minimum expectations of
performance were established, and these were embodied into national
legislation. Examiners were paid to check that all licensed aircrew met these
standards, and the public was consequently protected by the responsible
authorities, who had duly done their duty.
The psychology which emanated from civil aviation regulation is
interesting. Young people who chose a career in aviation found that tfiey
were faced with a minefield of hurdles to overcome. Many did not make it
There developed in them an acceptance of the 'obstacle course' mentality as a
necessary part of the profession. The training they received was traditionally
target-based rather than method based. A standard was expected (what),
with little or no explanation of how best to achieve this standard (how).
Trainers were not expected to emphasize the human limitations inherent in
handling complicated flying machines proficiently. Rather, an expectation of
rapid competence through exposure underscored their philosophy.
The first few years in such an environment set the mind of the young
aviator into this format. The trainee soon realized that little guidance was
being given on the human side of the task, and that results would be

115
Aviation Instruction and Training

dependent on the individual's own motivation and hard work in order to

qualify on new aircraft types. Trainees often grumbled about the apparent
difficulties inherent in this process, but continued to accept the status quo as
part of a limited downside to an otherwise rewarding career. Later, the whole
process was to become to some extent self perpetuating, in that those aviators
who became trainers and checkers, sometimes without any instructional
training, continued to promote the system that they themselves had been
exposed to, and were most familiar with.
However, the traditional system did produce results, and aviators emerged
who could certainly do the job. Such was always the argument against
change. Regardless of this, three important questions have to be asked:

1. Was there always, and is there now, a more efficient and effective way
of producing competent aircrew?
2. Could training have been, and could it now be made more relevant to
accident prevention in pin-pointing the crucial human factors involved in
flying?
3. Were rigid thought processes being groomed at the expense of lateral
thinking - a quality which is so necessary in coping with change and the
unexpected?

I suspect that all of these questions would attract a 'yes' answer, and this in
itself would provide powerful motivation for improvement. Indeed, new
attitudes are now spawning in civil aviation. A number of airlines are
implementing training programs which embody increasing elements of human
factors training. The initial response is encouraging. The next section deals
briefly with some of the main ingredients emerging in such programs, and the
likely benefits which may accrue to the airlines concerned, and potentially to
the industry as a whole.

New directions

For the past fifteen years, some airlines have pioneered training programs
aimed at the human side of the man-machine relationship in modem transport
aircraft. Courses were started with such titles as 'Cockpit Resource
Management', 'Aircrew Team Management', n ig h t Team Management', but
they all shared a common interest, emphasizing the need to take a new look
at the human problems facing aircrew in aviation.
There is now a gradually developing awareness aviation-wide that more

116
 Human Factors in Training

attention to the human factors involved in any task will promote that cmcial
factor for a new trainee: the right attitude. This is now being recognized as
the element missing as a target for so many training programs. The
International Civil Aviation Organization (ICAO) is a main sponsor for this
awareness, having published and promoted a number of Human Factors
Digests since 1989.
It is often assumed that 'the right attitude' is to be expected of a trainee. To
a certain extent such an expectation is reasonable, since the trainee did
choose this career. However, in practice, misconceptions about training
objectives are bom and reinforced well before training programs commence.
It is the nature of any rapidly interacting group, such as the aircrew body of
an airline, that a small germ of fact can quickly grow into a larger germ of
fiction, even outside the organization concerned. It is, therefore, likely that
many aircrew in any prospective group of trainees have an inaccurate idea of
either the nature or goals of the training to come, unless these have been fully
explained, or unless a conditioning process of some sort has been started well
before training commences. Wrong concepts at the start of a training course
can lead to confusion and the requirement for a corrective effort later.

Integration of airmanship, judgement, training and

other human factors in flight training: a vital priority
(Photo: Swissair)

117
Aviation Instruction and Training

In extreme cases, the confusion fails to go away. Training extensions and

failures can occur which are a direct result of the limitations of the system
itself. This may be measurable in terms of significant expenditure for any
airline, and here lies a very real motive for change. It is also clear that a
person charged with the responsibility of commanding a modem airliner
worth as much as US$130 million and carrying 400 passengers, needs to be
accurately and effectively trained, with attention to human as well as
technical factors. To be really effective, a major objective of training should
enthuse and motivate the individual to the task. A training system without
such an objective can reduce motivation for the task, and loyalty to the
organization.
So it is increasingly recognized that the training received by aircrew,
especially at the start of their careers, or conversion to a new aircraft type,
lays down an important framework of attitudes for the future career of the
individual, and consequently may have a bearing on the fortunes of the
company as a whole. If training is interestingly presented, it will probably be
enjoyable for the recipient, and is therefore likely to be effective. Beneficial
spin-offs from this are probably in the area of productivity, loyalty and, most
importantly, flight safety.
This brings us to the frequently repeated connection between flight safety
and human factors. It continues to be a matter of concern to the master of
aviation that human factors remain the primary cause of accidents. It follows
that the greatest impact in the effort to minimize this factor can be found in
effective training. There would still appear to be a need for aircrew to
appreciate their own limitations more, and to recognize that human beings do
make mistakes and will probably always do so. A greater acceptance of this
fact is the first important step towards the development of effective systems
of defence against human limitations.
Human factors in accidents should be looked for more seriously beyond
the aircrew who fly the aircraft. It is false logic to only consider the period of
the flight concerned and therefore argue that 70 per cent of accidents are
human factor related. Such an hypothesis continues to focus attention only on
those who fly the machines. Well before the airliner took to the air with
passengers and crew, it had to be designed, developed, tested and approved.
It did not grow naturally or just arrive on the scene. All those talented people
involved with the development process are human. All those responsible
individuals involved in preparing the machine for flight are human. All those
careful personnel who provide a security shield are human. All those tireless
officers in Air Traffic Control who direct the crew through international
airspace are human (so far!). All the new highly developed electronic

118
 Human Factors in Training

systems involved in aviation have been designed by humans. It is vital to

understand that, barring 'acts of God', all accidents are caused 100 per cent
by human factors, and not 60,70 or 80 per cent.
This being the case, there is a very real need for human factors training for
designers, air traffic controllers (some have them already), engineers, airline
managers, and all those involved in shaping the industry. There is a pressing
need for much more communication of human factors between all groups
involved in producing commercial flight if safety records are to be improved.
Today, the majority of the major airlines of the world are employing
various types of Cockpit/Crew Resource Management programs to promote
an understanding of the human factors involved in aviation. The initial
development of these programs was very much on the principle of 'strap on'
courses. 'Let's have one of these and see how it works!'. The course was
usually run in isolation from other training programs. Only a minority of
airlines are applying these principles directly into mainstream training
programs such as type training, upgrade training, or trainer training.
For those airlines actively developing human factors training, it is too early
to see definitive results. The time interval between the application of new
training philosophy and any sign of quantifiable results must be measured in
many years. A broad change of approach can also be painstaking and
difficult to institute. Nevertheless, many of the human factor principles
referred to are not new to other industries, and have been substantiated in
practice.
However, for the time being, interested airlines will have to rely on the
basic rationale for the introduction of human factors, already outlined. Proof
of improving accident statistics will take time to seep through. In the interim,
those who take this route may enable the required operational standard to be
reached earlier, with consequent savings of time and money for the airline.
The expression human factors' attempts to round up all those aspects of
human behaviour with which we are all stuck. For aviation, the fact that we
all make mistakes has to be recognized, especially by aircraft designers. For
aircrew, an understanding of human limitations, stress, and effective
communication is vital. As in any profession, 'people skills’, 'empathy' and
'common sense', all play important parts, but much more so for aircrew
working closely together in a dark flight deck for many hours over many
times zones.
The first step towards developing these skills lies in understanding that
most of us were not necessarily bom with them in large measure. It has been
found that human factor skills can be developed rapidly in a small-group
workshop style environment which is deliberately relaxed, and which is

119
Aviation Instruction and Training

facilitated by peers. The interaction which results from this environment can
stimulate thought and assist learning significantly.
The challenges for aircrew in the immediate future include a broader crew
composition of culture and age, and glass cockpit FMS-based flight,
combined with ultra long haul. New solutions have to be found, and effective
human-factors-based training, applied to all training programs, is an important
one.

120
9 Pilots’ approaches to
learning
P.J. Moore and R.A. Telfer

When people go about learning, in any context, they bring with them a wide
range of individual differences. While there are demonstrated differences in
abilities (e.g. Kirby, 1984) and preferences for dealing with information (e.g.
Kirby, Moore and Schofield, 1988), there has been a growing concern for the
general predispositions that learners bring to any learning task. This
research, now termed 'approaches to learning', examines the ways in which
learners use motives and strategies when they learn (e.g. Biggs, 1987a; Biggs
and Moore, 1992; Bowden, 1986; Entwistle and Ramsden, 1983; Watkins
and Hattie, 1990). It is the purpose of this chapter to discuss the concept of
approaches to learning, and in particular the effects of approaches to learning
on pilot learning, training and instruction.
The literature shows that there are various approaches to learning which
have been identified through factor analytic methods. Ramsden and
Entwistle's (1981) research with university students confirmed three
approaches to learning: orientations towards personal meaning, reproducing
learning, and achieving. Extensive research by Biggs (1979; 1985; 1987a)
with high school and university students identified three similar basic
approaches to learning. Biggs (1987a) calls these 'Surface, Deep, and
Achieving Approaches', each constituting a set of motives and strategies.
Learners with a 'surface approach' are motivated to meet minimal course
requirements and achieve their goals by rote learning and strategies designed
just to reproduce what they have learned. On the other hand, learners using a
'deep approach' are more intrinsically motivated, seek to personalise their
learning and undertake meaning-oriented learning activities. Finally, learners
with an 'achieving approach' are motivated to seek high grades, to enhance
their egos through competition, and to organize themselves for learning.
Aviation Instruction and Training

There is some evidence that these different approaches to learning produce

different learning outcomes. Surface approaches certainly provide plenty of
detail but lack structure (Biggs, 1979); deep approaches tend to produce well
organized, high level responses (Watkins, 1983); and an achieving approach
tends to positively influence academic performance generally (Biggs, 1987a).
For pilots, the question which follows is whether or not particular approaches
are more appropriate for them, given the nature of their education and
training. Are there subjects or exercises in which a surface apprQach may be
more effective than a deep approach and vice versa? If, for instance, a topic
is relatively easy to learn, essentially detail oriented and not structurally
complex (such as learning a check list or approved operating procedures),
there might be a good case for employing a surface approach. If a topic is
structurally complex and requires substantial integration for meaning (such as
aerodynamics or meteorology), then a deep approach may prove more
beneficial. Biggs (1979) demonstrated that surface level learners recalled
more details from a report reading task than deep learners. However, the
opposite applied for understanding the aims of the report.
In addition, learners have to be flexible. They are required to examine task
demands, complexity, required outcomes and then to make a decision about
the approach which will gain maximum performance. While Biggs (1987a)
has argued that the notion of 'deep' implies flexibility, this has yet to be
shown. Another possibility, more quantitatively based, is the 'more-is-better'
hypothesis. Hattie and Hattie (1987) and Watkins and Hattie (1990) suggest
that individuals who are stronger on more than one approach, irrespective of
the nature of the separate approaches, perform better on academic tasks than
those lacking strength on any approach.
This 'more-is-better' hypothesis has been tested by way of classifying
commercial pilot syllabus subjects as high on none, one, two or three of the
approaches to learning (Moore and Telfer, 1990a) and more will be said
about this later in the chapter.
As noted above, most of the research on approaches to learning has been
conducted on school and university populations. However, as shown in
Chapter 14, aviation differs from these contexts in seven ways; effectiveness
versus efficiency; education versus training; the learning and instructional
environment; teachers versus flight instructors; theory versus practice; art
versus craft; and pedagogy versus andragogy. Given these differences, it
follows that the motivation to learn in aviation may be quite different from
that in schools and universities as there are clear personal goals to be
achieved. The first solo, the first cross-country and the four-barred epaulette
are high stakes indeed. Career prospects depend directly upon performance in

122
 Pilots'Approaches to Learning

training, and there are usually both time and financial constraints. In addition,
the aviation industry tends to encourage the use of mnemonics to remember
information, and the development of an instructor 'patter': both approaches
that arguably reflect a surface approach to learning.
The aviation pilot training context is also interesting in the differing
demands of subjects that have to be learned. While conducting preliminary
work, Moore and Telfer (1988), noted that trainees made comments about the
varying degrees of difficulty of topics. Topics such as Aviation Medicine
were invariably reported as being 'easy' to learn, while others such as Flight
Planning were perceived as 'difficult'. If a topic is perceived as easy to
learn, it may prove beneficial to employ a surface approach to learning.
However, when the material becomes more complex, it may be more
profitable to employ a deep approach.
Below we outline several studies addressing the relationships between
approaches to learning (Deep, Surface, Achieving) and pilot performance.
The first study was conducted with student pilots, the second with
experienced pilots undergoing training to fly wide-bodied commercial jet
aircraft. Two further studies with experienced pilots are discussed, although
in lesser detail than the first two.

The student pilot study

This study employed a sample of sixty-two trainee commercial pilots and

examined the relationships among approaches to learning (as measured by a
slightly modified version of Biggs' (1987b), Study Processes Questionnaire),
performance in ground school topics of differing degrees of difficulty, and
performance in the aeroplane as measured by hours-to-solo. The 'more-is-
better' hypothesis was also tested by examining whether or not trainees who
scored high on more than one of the approaches performed significantly
better than those with low scores (Moore and Telfer, 1990a).
The relationships between total ground school scores and Surface, Deep,
and Achieving Approach as well as the Surface Achieving and Deep
Achieving scores showed consistently that the adoption of rote learning
strategies and minimal-effort motives were harmful to overall performance.
These findings, in a more technical setting than has been examined
previously, are consistent with other research (e.g. Biggs and Moore, 1992;
Watkins and Hattie, 1990) suggesting that the adoption of such an approach
is more generally disadvantageous than had been previously demonstrated. It
is interesting that the rather negative effects of surface level approaches to

123
Aviation Instruction and Training

learning are seemingly neutralized by the addition of the Achieving

dimension.
In contrast to the overall ground school results are the findings for airborne
instruction. Surface Approach scores had a random relationship with time
taken to fly solo but Deep, Achieving and Deep Achieving were significandy
related (negatively) with time taken to fly solo. The most powerful effect was
for the Deep Approach. Approaching the task of flying with intrinsic
motives, a sense of inquiry and adventure proved an advantage. This leads to
a sense of personal satisfaction out of mastering the aeroplane and its
environment, a sense of absorption while flying, and a desire to understand
what happens, and why it occurs, when a task is undertaken in the cockpit.
In addition, the deep oriented flyer would be keen to monitor the effects of
such aeroplane manipulation, and to ponder on causes of those effects.
Locus of control factors could also play a role in the deep learner's
operation of the aeroplane. Deep oriented learners are more likely to be
internal, perceiving themselves as having control over their environments
(Biggs, 1987a). This would be beneficial to flying, allowing for the earlier
development of confidence in solo operations. Additionally, it would
promote the necessary autonomous attitude and self-sufficiency necessary for
command judgement and responsibility.
However, the results provided no support for the hypothesis that surface
level approaches could be beneficial for the learning of information that was
rated as relatively easy to learn. Perhaps even the topics perceived as easiest
to learn are still sufficiently complex to require the employment of strategies
beyond those characterizing the surface learner.
The method of assessment may also have been a factor. Biggs (1979)
showed that surface learners performed better on factual, detailed work. An
inspection of the types of questions asked in the tests of the 'easy' topics
showed in fact that the questions seek specific detail and do not, in general,
require problem solving or more complex thinking. The multiple choice
questions in the test require the retrieval of specific, unrelated information.
In contrast to the Surface findings were the Deep score relationships.
There was support for the notion that more difficult material benefits from
more meaning-oriented approaches. Success in a relatively difficult task
requires relating new and old knowledge, spending extra time finding out
more information about the topic, being absorbed in learning about the topic
and wanting to understand the information. Further, the assessment of the
more difficult topics focusses upon problem solving, judgement and
decisions.

124
 Pilots'Approaches to Learning

The 'more-is-better' hypothesis was not supported. There were no

significant differences between groups high on none, one, two or three of the
Approach scales on total ground school, nor on easy, moderate or difficult
topic scores, nor on hours-to-solo. These findings are in contrast to those
reported by Hattie and Hattie (1987).
Being high on surface scores was related negatively to performance in
general, being high on deep scores was helpful to flying and performance on
difficult topics. The achieving dimension had virtually a random relationship
with performance. To score high on both deep and surface implies a
neutralizing effect. High achieving scores reduced the negative influence of a
surface approach. Students who had high scores on several dimensions,
especially deep and surface, may have confusion in determining which
strategies to use in which contexts and consequently did not gain maximum
benefit from their range of strategy and motive options (cf. Cantwell and
Moore, 1991).

Pilots' approaches to learning differ.

It is vital that instructors are aware of such individual differences.
(Photo: Swissair)

It may be that the pressures in aviation do not allow sufficient time for the
learner to be reflective and ponder different ways and means of solving

125
Aviation Instruction and Training

different types of problems. It could also be that the training system itself,
with its high structure and efficiency-driven motives, does not permit the
flexibility of individualization.
This study raised questions about the efficiency of traditional methods of
training commercial pilots. Periodically, pilots are required to confirm or
upgrade their skills and knowledge or seek endorsement on new equipment.
With the cooperation of an international carrier, a group of newly-recruited,
experienced pilots were the subjects in a further study.

The airline pilot study

The 30 pilots were undertaking an introductory and orientation course before

commencing line duties. The results of Biggs' (1987a) Study Processes
Questionnaire (slightly modified) was linked with the pilots' results in the
initial test, type of aircraft and safety tests and rated performance in the
simulator. These results showed very little relationship betweeen approach
scores and the indices of pilot performance both in academic type learning
and simulated flying. The only significant relationship was between Deep
scores and the test examining particulars and principles of the aircraft that
they were being trained to fly (Moore and Telfer, 1990b). While the pattern
of findings with the experienced pilots and the student pilots differs, the
results are in theoretically defendable directions, especially the findings that
being deep is helpful for certain types of complex learning. Of course the
context of learning for the experienced pilots, and the backgrounds that they
bring to that learning, are vastly different from school-type learning and,
consequently, a study was undertaken to test whether there were aspects of
the experienced pilots' approaches to learning that the Study Processes
Questionnaire was not revealing.

Additional studies of airline pilots

Eleven commercial airline pilots were interviewed individually on their

approaches to learning (Telfer, 1991). The results demonstrated that their
motives, as probed by the SPQ, were the same, but that there were
differences in the strategies the pilots were employing. As busy professionals
and heads of households, they managed their time efficiently by shuffling
priorities, and they devised learning techniques to accommodate these
demands. They used such methods as notes in the kitchen, highlighted codes

126
 Pilots'Approaches to Learning

on the desk calendar, pocket cards, discussions over a beer, queries to

experienced colleagues, and simulated flight situations in a car seat or lounge
chair. Check flights, simulator schedules, CRM courses and so on had to be
scheduled among children's birthdays, wedding anniversaries, and school
vacations. To cope with changed circumstances, the pilots changed their
approaches to learning.
The interview data, and further discussions with the aviation industry, led
to the development of a more 'situated', aviation oriented version of the SPQ
so that it incorporated items focusing upon the strategies reported to be
successful by the experienced pilots. The motive items were also modified
and expanded so that they referred explicitly to the aviation context.
The new instrument, the Pilot Learning Process Questionnaire (PLPQ) was
based, like that of Biggs (1987b), on notions of Surface, Deep and Achieving
for both motives and strategies. The initial version of the PLPQ has 62 items
and respondents have been sought from a number of international airlines.
The first steps will be to examine the validity of the instrument via factor
analysis and assess item reliability. Emerging from these analyses should be
an appropriate instrument for examining experienced pilots' approaches to
learning. Following these steps is the obvious one of examining the
predictive ability of the instrument, that is its relationship with pilot
performance.
For the future, the PLPQ may have value as a further means of guiding
pilot selection. More importantly, it may have a diagnostic function to enable
airlines to assist pilots experiencing training problems, and it may assist
individuals to develop an awareness of their approach to learning and of ways
in which they may need to change that predisposition in order to be
successful in some aspects of training.
If, for instance, a pilot is an essentially surface oriented learner, training in
strategies such as setting personal goals, regularly reviewing material, testing
oneself on current knowledge state, and generating summaries to append to
notes taken during lectures and study sessions may prove beneficial. Overall,
though, the instructional focus at all levels of training (student to experienced
pilot) should be on the adoption of deeper learning approaches which can be
encouraged by allowing the learner to experiment with flying and realize a
degree of metacognition. In both ground school and in-aeroplane instruction,
there should be an encouragement of self-questioning, monitoring and
evaluating (these are major aspects of metacognition) in a relatively anxiety
free environment. For the instructor this implies constant monitoring of the
trainee, especially in the cockpit where cognitive overload is easily induced.

127
Aviation Instruction and Training

References

Biggs, J.B. (1979), 'Individual differences in study processes and the quality
of learning outcomes', Higher Education, vol. 8, pp. 381-394.

Biggs, J.B. (1985), 'The role of metaleaming in study processes', British

Journal of Educational Psychology, vol. 55, pp. 185-212.

Biggs, J.B. (1987a), Student Approaches to Learning and Studying,

Australian Council for Educational Research, Hawthorn, Victoria.

Biggs, J.B. (1987b), The Study Processes Questionnaire (SPQ): Users'

Manual, Australian Council for Educational Research, Hawthorn, Victoria.

Biggs, J.B. (1988), The lack of impact of research on improving student

learning: A question of methodology?', Paper presented at the 24th
International Congress of Psychology, Sydney, August.

Biggs, J.B. and Moore, P.J. (1992), The Process o f Learning (3rd ed.),
Prentice-Hall, Sydney.

Bowden, J.A. (1986), Student Learning: Research into Practice, Centre for
the Study of Higher Education, University of Melbourne, Melbourne.

Cantwell, R.H. and Moore, P.J. (1991), 'The management of learning: the
impact of strategy control and approach to learning on final year nursing
students', Paper presented at the Australian Association for Research in
Education Annual Conference, Gold Coast, November.

Entwistle, N.J. and Ramsden, P. (1983), Understanding Student Learning,

Croom Helm, London.

Hattie, J. and Hattie, J. (1987), 'Study skills and achievement', Paper

presented at the first AARE/NZARE Conference, Christchurch,
November.

Kirby, J.R. (1984), Cognitive Strategies and Educational Performance,

Academic Press, New York.

128
 Pilots' Approaches to Learning

Kirby, J.R., Moore, P.J. and Schofield. N.J. (1988), 'Verbal and visual
learning styles', Contemporary Educational Psychology, vol. 13, pp. 169-
184.

Moore, P.J. and Telfer, R.A. (1988), 'Study processes and trainee pilot
performance', in Proceedings o f the Workshop on Aviation Psychology,
Institute of Aviation, Newcastle.

Moore, P.J. and Telfer, R.A. (1990a), 'Approaches to learning: relationships

with pilot performance', The Journal o f Aviation/Aerospace Education and
Research, vol. 1, no. 1, pp. 44-58.

Moore, P.J. and Telfer, R.A. (1990b). 'Pilot learning styles', paper presented
at the Australian Psychological Society Silver Jubilee Conference,
Melbourne, September.

Ramsden, P. and Entwistle. N.L. (1981), 'Effects of academic departments

on students' approaches to studying', British Journal o f Educational
Psychology, vol. 51, pp. 368-383.

Telfer, R.A. (1991), 'How airline pilots learn', Paper presented at the Annual
Conference of the New Zealand Psychological Society, Palmerston North,
August.

Watkins, D. (1983), 'Depth of processing and the quality of learning

outcomes’, Instructional Science, vol. 12, pp. 49-58.

Watkins, D. and Hattie, J. (1990), 'Individual and contextual differences in

the approaches to learning of Australian secondary school students',
Educational Psychology, vol. 10, pp. 333-342.

129
10 Transfer for flight
simulation
G. Lintern and G. McMillan

The modem flight-training simulator is intended to provide a vehicle for

safe, economical and efficient instruction of flight control skills. At issue is
how a flight simulator can be designed to ensure the satisfaction of those
goals. The normal design approach is to seek high fidelity; that is to design
and tune the simulator so that it mimics to the closest possible extent the
perceptual sensations of real flight. The pursuit of high-fidelity can,
however, be a costly exercise. A more rational design approach would be to
evaluate what dimensions of these components impact training effectiveness.
Such an approach should result in the design of systems that are less costly
but equally effective.

Overview

The essential issue to be addressed here is whether a specific simulator

feature (i.e. an attachment, an added capability, or a modification) leads to
enhanced training effectiveness. From the behavioral perspective, the issue
is one of skill transfer. How well do skills learned in a simulator transfer to
the aircraft? On what design dimensions is it possible to relax the
requirement for high fidelity without compromising training effectiveness?
Four major components of a simulator stand out in terms of their potential to
impact training effectiveness and in terms of their potential to impact design
and production costs. These are visual systems that display representations
of out-of-cockpit scenes, whole-body motion cueing systems, flight
dynamics, and control systems. Transfer studies relating to these four major
 Transfer fo r Flight Simulation

design dimensions are reviewed under the categories of visual displays,

proprioceptive motion cueing, system dynamics, and control forces.
It is regrettable, but nevertheless a fact, that the bodies of experimental
literature to be reviewed leave much to be desired. Where possible, studies
that focus on flight control are emphasized. However, there are relatively
few of these, and it is necessary to supplement the review by consideration
of laboratory studies that employ basic tracking tasks. Data from nonflight
studies are useful in that they identify basic principles of transfer but such
studies must be followed by others that demonstrate the relevance of those
transfer principles to flight control. In most cases the essential flight data
are not available.

Experimental paradigm

The primary concern in this review is with differential transfer; namely

differences in transfer resulting from different training conditions. Figure
10.1 outlines the basic paradigm. The effectiveness of a visual system, for
example, could be assessed by training randomly selected groups of students
on one or the other condition. Tasks and training strategies should be as
similar as possible. Performance in the aircraft on the instructed tasks would
be used to establish relative effectiveness of the two training conditions.

Figure 10.1
The essentials of a transfer experiment to examine relative effectiveness
of different simulator configurations

131
Aviation Instruction and Training

Several of the studies to be reviewed differ from this paradigm in that the
students are not transferred to the aircraft. In some experiments, transfer is
assessed in a simulator. Often the same simulator used for training is
reconfigured for the transfer test. The term 'quasi-transfer' is used here to
distinguish these types of transfer studies from those in which transfer is to
an aircraft. Other studies examine transfer effects in the context of a basic
tracking task which, at best, represents only one dimension of a flight task.
While these experiments are elsewhere referred to as transfer studies they
are, in terms of the distinction employed in this review, appropriately seen as
quasi-transfer studies.

Qantas aircrew undergoing training in one of the airline's

Singer Link flight simulators at the Qantas Jet Base in Sydney
(Photo: Qantas)

Issue

Two transfer studies by Ritchie and his associates (Ritchie and Hanes, 1964;
Ritchie and Michael, 1955) provide a useful introduction to the issues for
flight simulations in that the data show trends that stimulate a number of
questions about simulation fidelity. Acquisition and transfer of flight skills
under instrument and visual flight (contact) conditions were examined in

132
 Transfer fo r Flight Simulation

both experiments (Table 10.1). Flight naive subjects learned to fly straight
and level to a criterion of three errorless trials, and then learned to make
level 180 degree turns also to a criterion of three errorless trials. Some
subjects first practiced under instrument conditions and transferred to visual
contact conditions, while other subjects first practiced under visual contact
conditions and then transferred to instrument conditions. Some subjects had
their initial instrument training in the aircraft while others had it in a C-3
Link Trainer. In addition, all subjects who started with instrument training
(either in the aircraft or the Link Trainer), completed a further series of
instrument trials in the aircraft following their visual (contact) training in the
aircraft.
Given the similarity of the two experiments it was possible to enhance the
stability of the data by combining comparable groups from the two different
experiments for reanalysis. The performance measure is total trials required
to achieve criterion in both tasks. The data are shown in Figure 10.2. The
only statistically significant difference is between groups in the final session.
Although the first- and second-session trends are not significant, the data
suggest that the Link Trainer was more difficult to control than the aircraft,
and that there was marginally less transfer to contact flight in the airplane
from the Link than from instrument flight in the airplane. Given that the
Link-trained subjects had approximately 30 per cent more instrument
practice prior to transfer to the airplane than those first trained on
instruments in the

Table 10.1
Experimental design for flight transfer experiments

Group
1 Link Trainer Airplane Airplane
Instruments Contact Instruments
2 Airplane Airplane Airplane
Instruments Contact Instruments
3 - Airplane Airplane
Contact H* Instruments

(Source: Ritchie and Hanes, 1964; and Ritchie and Michael, 1955)

In each phase, straight and level flight was instructed to a

criterion of three successive errorless trials, and then level
turns were taught to the same criterion.

133
Aviation Instruction and Training

Figure 10.2
Trials to criterion in transfer between instrument and contact
conditions

(Adapted from Ritchie and Hanes, 1964; and Ritchie and Michael,
1955)

airplane, this seems a noteworthy result. These two trends suggest that the
Link Trainer was deficient in some important respects. Statistically
significant support for this conclusion is found in the final session when
those subjects first trained in the Link were considerably poorer on
instruments in the aircraft than those whose first session was on instruments
in the aircraft. Unfortunately, there is insufficient information in the reports
of these experiments to identify or even to speculate about the particular
limitation.
There did, nevertheless, appear to be some advantage from prior
instrument training in the Link. In comparison to subjects who were not
given first-session training on instruments (within the aircraft or the Link),
the Link-trained subjects performed marginally better in contact flight, and
significantly better in the final flight session on instruments. Jacobs (1976)
has also shown the value of pretraining in a simulator; in this case a
modified Singer Link GAT-1 where transfer was to instrument flight in a
Piper Cherokee Arrow. An experimental group, pretrained in the GAT-1,

134
 Transferfo r Flight Simulation

required approximately 4.2 hours of flight time to achieve criteria on a

number of flight tasks versus approximately 6.5 hours for a control group
with no pretraining, although approximately 7.4 hours of GAT-1 time were
used to achieve this advantage.
In summary, the studies reviewed here show that an instrument trainer can
provide useful transfer to the air even if, as in the case of the C-3 Link
Trainer, it is demonstrably deficient in some respects. The central issue for
this review is to assess what physical differences between simulator and
aircraft might impact training effectiveness.

Visual displays

A comprehensive review of the early transfer research with visual flight

simulation is contained in Waag (1981). The bulk of that research involved
transfer studies in which the usefulness of pretraining in a ground trainer
with a visual attachment was tested against a base line of no simulator
pretraining. Most of the results are positive, thus demonstrating the value of
the visual simulation and ground trainer combination. Nevertheless, it is
well established that skills learned in a ground-based trainer (with no visual
display) can be transferred to the aircraft (e.g. Flexman, Roscoe, Williams
and Williges, 1972; Jacobs, 1976; Povenmire and Roscoe, 1971, 1973),
and it would be of considerable value to isolate the increment in transfer
resulting from the use of a visual system.
Of the experiments reviewed by Waag, only one by Browning, Ryan,
Scott, and Smode (1977) has made this comparison. Although the result
favored the addition of the visual system there were two confounds that
make the data difficult to interpret. The simulators used for visual and
nonvisual training were not identical, and the amount of training given in the
visual simulator was approximately twice that given in the nonvisual
simulator. Either of these factors could account for the superior transfer for
the visually trained group.
In some of the studies reviewed by Waag (1981), variations in the visual
scenes were compared. Day and night scenes were contrasted in two
experiments (Martin and Cataneo, 1980; Thorpe, Vamey, McFadden,
LeMaster, and Short, 1978) and field-of-view was manipulated in another
(Nataupsky, Waag, Weyer, McFadden, and McDowell, 1979). There was
no differential effectiveness of training with day versus night scenes or with
different fields-of-view. Thus these studies failed to establish any training

135
Aviation Instruction and Training

advantage for high-fidelity on two of the more prominent dimensions of

visual scenes.
Since Waag’s review, a few relevant studies have been conducted at the
United States Navy's Visual Technology Research Simulator. Westra
(1982) examined field of view and scene content (day versus night) in a
quasi-transfer study of carrier-landing training. Some temporary transfer
advantage was shown for the high-fidelity configurations of both of these
factors, but a subsequent transfer study (Westra, Lintem, Sheppard,
Thomley, Mauk, Wightman and Chambers, 1986) failed to replicate these
findings, and failed to show anything but the most transient of advantages
from the prior simulator training.
Lintem, Thomley, Nelson, and Roscoe (1987) used a quasi-transfer
design to examine scene content (moderate detail versus a schematic grid
pattern) for the acquisition of bombing skills. Transfer was tested to both
the moderate-detail and schematic scenes. Training with the moderate-detail
scene was better for transfer to both the moderate-detail and the schematic
scenes. This appears to be the first study to show a transfer advantage for
higher levels of scene content. In a follow-up study (Lintem, Sheppard,
Parker, Yates, and Nolan, 1989), manipulations of field of view and of scene
content (day versus night) did not differentially affect transfer to the
airplane, although the simulator training did enhance subsequent transfer to
the aircraft. To summarize these US Navy studies, the evidence that any
visual manipulations can impact transfer effectiveness is meager. Training
in a simulator with a visual system does result in transfer to an airplane but
there is no clear evidence that the visual system offers a useful increment in
transfer.
More recent research at the University of Illinois has examined issues
relating to the use of visual displays for the ab initio instruction of landing.
A quasi-transfer study by Lintem, Roscoe, and Sivier (1990) showed that
line-up control was learned more effectively with a pictorial versus a
symbolic display. Another quasi-transfer study by Lintem and Koonce (in
press) showed that descent-path control was learned more effectively with a
moderate- versus a low-detail pictorial scene. Although both studies suggest
that a moderately-detailed pictorial representation of an airport scene does
assist the instruction of landing skills, they leave open the question of
whether similar benefits would be observed in transfer to an airplane.
Lintem, Roscoe, Koonce, and Segal (1990) demonstrated positive transfer
to the airplane following landing instruction in a trainer with a moderately-
detailed pictorial scene. However, as in most previous research with visual
displays, the control subjects in this experiment received no simulator

136
 Transfer fo r Flight Simulation

training. Thus, the specific contribution of the visual scene could not be
assessed. Taylor, Lintem, Koonce, Kaiser, and Morrison (1991) sought to
make such an assessment in a transfer experiment in which control students
were given equal instruction in a simulator with no visual display. The
results showed no relative advantage for visual versus nonvisual instruction,
but both groups performed better than the control subjects of Lintem,
Roscoe, Koonce, and Segal (1990). It remains unclear whether the visual
system offers no enhancement at all for instruction of this visually-oriented
skill, or whether the extra visual and nonvisual training schedules in the
simulator engendered different but (coincidentally) equal enhancements in
transfer.

Summary

The weight of evidence suggests that skills learned in a simulator with a

visual system do transfer to the airplane. However, the most basic question
for visual simulation, 'Can the addition of any type of visual system increase
transfer?’ has received only the barest of attention, and the results are by no
means conclusive. Some quasi-transfer studies have demonstrated that at
least moderate levels of pictorial information are desirable. Otherwise it has
been difficult to demonstrate that even gross variations in the configuration
of a visual display can impact transfer. Despite the huge commitment by the
flight simulation industry to high-fidelity visual systems, there is only the
barest empirical support for the use of visual systems in flight training.
There are, of course, some visually-oriented tasks that would appear to
require a representation of the visual world; for example, VFR landings and
ground referenced maneouvers. The general assumption is that some form
of visual display is needed for practice on these tasks. Nevertheless, even
this type of assumption can be questioned. Students may leam most of what
is required by practicing the required control skills with reference to
instruments. Alternatively, it is possible that even the best of visual systems
will not enable effective instruction in the simulator of these visually
referenced tasks.
Rehearsal of complex missions is another activity that would seem to
require some form of visual display. There is, however, no systematic
research on the effectiveness of mission rehearsal. Thus the effectiveness of
mission rehearsal, the requirements for a visual display, and the optimum
levels of fidelity for a visual display are unknown at this time. Resolution of
the uncertainty surrounding the usefulness of visual systems can only be

137
Aviation Instruction and Training

achieved through systematic empirical evaluations that incorporate an

appropriate transfer design.

Proprioceptive motion cueing

An experiment by Jacobs (1976) offers the methodological model for

examining the transfer effectiveness of flight simulator motion cueing.
Groups of nine subjects each were taught instrument maneuvers in a Singer-
Link GAT-2 that had been modified to simulate a single-engine aircraft.
One group was taught instrument maneuvers with the motion system active,
and another group was taught the same maneuvers with the motion system
inactive. Subjects received a fixed number of simulator trials on each
maneuver regardless of performance1. On transfer to further instrument
practice in a Piper Cherokee Arrow airplane, groups trained with and
without simulator motion performed at an equivalent level, and were better
than a control group that had received no prior training in the GAT-2. Thus,
prior training in the GAT-2 speeded training in the airplane but the use of the
motion system provided no additional enhancement over that accruing from
the use of a fixed base.
Several tests of motion systems have been conducted at the Flying
Training Division of the United States Air Force Human Resources
Laboratory. These studies are reviewed by Waag (1981) and summarized in
Table 10.2. In not one of these studies was there a significant advantage for
training with a motion system. One quasi-transfer study has also been
undertaken in the United States Navy's Visual Technology Research
Simulator (Westra, 1982). A carrier landing task was taught with or without
activation of a six-degree-of-freedom motion system. On transfer to a test
with the motion system active there was no differential effect of training
with or without motion.
Only one flight transfer study has shown an effect of motion cueing.
McDaniel, Scott and Browning (1983) tested Naval Aviators who, having
graduated from basic flight training, were transitioning to the SH-3
helicopter, these students were trained to proficiency2 on SH-3 maneuvers
in a simulator with or without motion. They were then tested to proficiency
in the aircraft on the same maneuvers. The report of this research tabled
group means and standard deviations of aircraft trials to criterion on each of
nine maneuvers. Analysis by't' tests showed the differences between motion
and no-motion groups to be significant for four maneuvers. Flight
performances on two maneuvers were enhanced by prior training with

138
 Table 10.2
Flight transfer tests of motion in training

Authors Simulator Aircraft Visual Tasks Results*

Woodruff and Smith (1974) T-4, T-4G T-37 440(H)x280(V) Basic Contact nsd
Basic Instruments

Woodruff et al. ASPT - 300°(H)xl50°(V) Basic Presolo nsd

Advanced Contact
Instrument Navigation
139

Unpublished (Waag, 1981) ASPT T-37 300°(H)xl50°(V) Basic Contact nsd

Martin and Waag (1978a) ASPT T-37 300°(H)xl50°(V) Basic Contact nsd

Martin and Waag (1978b) ASPT - 300O(H)xl50°(V) Aerobatics nsd

Nataupsky et al. (1979) ASPT T-37 300<>(H)xl50O(V) Basic Contact nsd

Gray and Fuller (1977) ASPT F-5 300°(H)xl50°(V) Weapon Delivery nsd

nsd = no significant difference

Aviation Instruction and Training

motion but for two other maneuvers, were degraded by prior training with
motion.
An experiment by Levison, Lancraft, and Junker (1979; also reported in
Levison, 1981) is of particular interest because it provides additional data
which suggest platform motion cueing can effect transfer. Motion-platform
cueing and asynchronous visual-motion cueing were tested as quasi-transfer
variables with a one-dimensional, third-order, compensatory tracking task.
This task simulated the nulling of aircraft roll disturbances generated by a
pseudo-random forcing function. Subjects could be made aware of the roll
disturbances through a visual display and also proprioceptively via a motion
platform. Independent groups of subjects were trained with synchronous
visual and proprioceptive cueing (both delayed by 61 msec because of
irreducible delays in the simulation), with visual cueing only (also delayed
by 61 msec), or with asynchronous visual-motion cueing where delays of 80,
200, or 300 msec were added to the minimum motion-system delay of 61
msec. Subjects were trained to a criterion of asymptotic performance. All
except those trained with synchronous visual-motion cueing were then
transferred to the synchronous condition for further testing.
The results of that experiment, as assessed by mean square tracking error,
are shown here as Figure 10.3.3 These data show superior asymptotic
training performance with synchronous visual-proprioceptive cueing.
The delay of proprioceptive cueing by as little as 80 msec degraded
performance significantly. Delay by 200 msec or more, or complete
removal of proprioceptive cueing, seriously degraded training performance.
The effects in transfer were far less substantial. Here the 80 msec training
delay did not appear to have had any effect. In the first transfer session (4
trials of 165 seconds each) there were significant but slight degradations in
transitioning from proprioceptive delays of 200 msec or more and from
visual cueing only. By the end of testing the differences were considerably
reduced and were no longer statistically significant. Thus, delays in the
cueing system or absence of motion cueing had some detrimental effect on
transfer, albeit a relatively slight and transient one.
One feature of all reports of the flight transfer studies that might be noted
is that turbulence levels were unspecified, which presumably means that
there was no turbulence in the simulation phase of these studies. Only in the
quasi-transfer study by Westra (1982) was turbulence specified. Here a
turbulence/no-turbulence factor was crossed with the motion/no-motion
factor. Turbulence in training was set at a level regarded by pilots as a
maximum reasonable level, and in transfer was set at a moderate level.
There was however, no interaction between the two factors. While this

140
 Transfer fo r Flight Simulation

suggests that turbulence may not modify the effects of motion, there is some
reason to be cautious. One of the two studies to show a transfer effect for
motion (Levison et al., 1979) used a disturbance function, and Westra's
study examined only one task. Clearly, there is much to be tested before the
motion issue can be resolved.

Figure 10.3
Transfer from no-motion, asynchronous-motion, and
delayed-motion conditions to an asynchronous motion condition

(Adapted from Levison, 1981.) Visual cueing was always

delayed by 61 msec. Delays specified in the figure refer to
the motion cueing response in relation to the visual cueing
response. Data for the early trials of some conditions were
not provided in either of the two reports of this work.

A second concern for the studies reviewed lies in the visual-motion cueing
asynchronies. The simulator used in the studies reviewed by Waag (1981)
had visual-motion asynchronies of several hundred milliseconds (Gum and
Albery, 1977). There is no reason to believe that the system used by Westra
was any better. This must remain as a significant caveat on currently

141
Aviation Instruction and Training

available data because Levison et al. (1979) showed a significant transfer

disadvantage for visual-motion asynchronies in training, and showed no
difference between no-motion and asynchronous conditions.

Summary

Tests of motion as a transfer variable have been disappointing. Only one

result favors the use of motion platforms, and it is from a somewhat artificial
quasi-transfer experiment. The single flight transfer study to show an effect
of motion (McDaniel et al., 1983) serves to highlight the dangers of being
guided by intuition in the design of training devices. Contrary to an almost
universal belief in the value of motion, this study demonstrated that flight
training can actually be degraded by the use of a motion system. On the
other hand, there were also some benefits that could be attributed to motion.
These results suggest that motion can be advantageous in some
circumstances and disadvantageous in others. The remaining flight transfer
experiments show no transfer effects due to motion, and thereby suggest that
in even other circumstances, motion is neither advantageous nor
disadvantageous. The ambiguity of the body of results would appear to
provide strong justification for a more systematic and comprehensive
program of research into issues surrounding the use of motion in flight
training simulators.

System dynamics

Almost all of the data that bear on the issue of fidelity of system dynamics
are from basic tracking studies. Muckier, Obermayer, Hanlon, and Serio
(1961a, 1961b, 1961c) provide three reports that covered seven experiments
on the transfer of skill with one-dimensional, compensatory tracking.
Experiments 1 (Muckier et al., 1961a), 3, 4, and 5 (Muckier et al., 1961b)
are of central interest here. In these four experiments subjects were required
to track a 0.1-Hz sine wave. The tracking plant was a second-order
simulation of an aircraft's phugoid response (the natural frequency of system
oscillation in the longitudinal axis). Plant response could be adjusted in
frequency of phugoid response, in system damping (i.e. time to damp a
transient input to half amplitude) and system gain.
The experiments of Muckier et al. were quasi transfer-of-training studies
in which groups of ten subjects, drawn from a college-student population,
were trained under a particular plant condition for 30 trials (each of 30

142
 Transfer fo r Flight Simulation

seconds duration) and transferred to a criterion plant condition for 10 trials

(also of 30 seconds duration each). Time on target was used as the
dependent measure. Fifteen seconds of the maximum possible 30 seconds
of time-on-target was a benchmark level of performance that would be
obtained with the open-loop performance (i.e. no control inputs from the
subject).
In Experiment 1, phugoid frequency and damping of plant response were
fully crossed in training (3 levels of each to give 9 independent groups). All
groups were transferred to the medium values of frequency and damping.
Data from Muckier et al. (1961a) are presented here as Figures 10.4 and
10.5. They depict training and transfer performances for training
manipulations of frequency (Figure 10.4) and damping (Figure 10.5).
The phugoid frequency manipulation had a significant effect in training
and in transfer. Those trained with a low frequency of oscillation performed
very poorly in training (no better than open-loop performance). Performance
in transfer was poor for both groups transferring to the medium frequency in
relation to the performance of the control group that continued with the
medium frequency. Furthermore, the transfer performances of the two
experimental groups were not substantially better than early training
performance of the control group (Figure 10.4). This indicates that subjects
gained a marginal benefit at best from their prior practice with a higher or
lower frequency. Also to be noted here is that the high-frequency group
performed about as well as the medium-frequency group in training, but lost
that parity in transfer.
As a contrast, Figure 10.5 shows no performance differences either in
training or in transfer. It would appear that manipulation of damping had no
effect. This suggests that subjects could easily and spontaneously adjust
their own dynamic response to cope with changes in damping, while they
had considerable difficulty in doing so with changes in the natural frequency
of system oscillation. The remainder of the data from Experiment 1 of
Muckier et al. (1961a) support these observations. In general, changes in the
natural frequency of oscillation had an effect while changes in damping did
not. Where damping is changed, subjects perform as well as if they had
been trained on the transfer configuration. Where frequency was changed,
they perform as poorly as, or only slightly better than if they had no prior
training.
In Experiment 3 (Muckier et al., 1961b) gain was doubled from that used
in Experiments 1 and 2. Data from Muckier et al. (1961b) are shown here
as Figures 10.6 and 10.7. The natural frequency and damping conditions are
a set taken from Experiment!; high-frequency/heavily-damped, medium-

143
Aviation Instruction and Training

Figure 10.4
Transfer from a high and a low to a medium frequency
of phugoid response
30-

(Adapted from Muckier et al., 1961a) Phugoid response refers to the

natural frequency of system oscillation in the longitudinal axis.

Figure 10.5
Transfer from a high and a low to a medium damping ratio

144
 Transfer fo r Flight Simulation

Figure 10.6
Transfer between variation in frequency of phugoid response
and damping ratio at a low gain

Figure 10.7
Transfer between variations in frequency of phugoid response
and damping ratio at a high gain

(Adapted from Muckier et al. 1961b)

145
Aviation Instruction and Training

frequency/moderately-damped, and low-frequency/lightly-damped. Figure

10.6 shows data from the low-gain conditions of Experiment 1. The large
training and transfer differences noted for the low-gain conditions of
Experiment 1 are no longer evident with the high-gain conditions of
Experiment 3 (Figure 10.7).
Although gain had been varied in Experiment 3, it had not been
manipulated as a transfer variable. Subjects in Experiment 4 were
transferred from high- to low-, and from low- to high-gain conditions. The
medium-frequency/moderately-damped plant was used. Gain had an
important effect; those trained with low gain could perform as well on
transfer to the high-gain condition as those trained on the high-gain
condition, but those trained with high gain performed very poorly on
transfer to low gain. This latter group of subjects appeared to have learned
nothing of use for low-gain tracking from their prior experience with the
high-gain condition (Figure 10.8).
Muckier et al. (1961b) noted that their gain manipulation had modified
the time required for operators to detect the response of the plant to a
control input. Essentially, at higher gains, a plant will respond more quickly
so that the operator will notice the results of a control input sooner. While
most research with dynamic systems emphasizes the adjustment of
parameters of the dynamical equations, this notion of system responsiveness
implicates the perceptual capabilities of the operator. System
responsiveness may be adjusted through manipulations of delay, lag, order,
and damping as well as of gain.
The hypothesis forwarded by Muckier et al. (1961b) was that differences
in system responsiveness were responsible for at least some of the
differential transfer effects they had observed. The gains of a set of
frequency/damping conditions that had shown differential transfer in
Experiment 1 (Figure 10.6) were adjusted in Experiment 5 to equalize
system responsiveness. This adjustment eliminated the differential transfer
effects (Figure 10.9), thereby providing some support for the hypothesis that
differences in system responsiveness can have a major impact on transfer.
It also appears that the responsiveness of the transfer plant modifies the
effects of responsiveness as a transfer variable. As noted in the previous
discussion of Experiment 5, variations in responsiveness during training
produced differential transfer to a low-gain (sluggish) plant but not to a high-
gain (responsive) plant (Figure 10.8). In Experiment 5, Muckier et al.
(1961b) also tested the same frequency/damping conditions that had shown
differential transfer in Experiment 1, but with the higher gain value used in
Experiments 3 and 4. Although system responsiveness was not equalized in

146
 Transfer fo r Flight Simulation

Figure 10.8
Transfer from high and low to low and high settings of control gain
30-

(Adapted from Muckier, et al. 1961b)

Figure 10.9
Transfer between variations in frequency of phugoid response and
damping ratio at a low gain with responsiveness of the system equalized

(Adapted from Muckier et al., 1961b)

147
Aviation Instruction and Training

training for this set of high-gain conditions there were no differential transfer
effects. Thus, a general conclusion to be drawn from the work of Muckier
et al. (1961a, 1961b) is that variations in system responsiveness during
training will affect transfer to a sluggish but not to a responsive plant.
Plant gain has also been manipulated systematically as a transfer variable
in two experiments by Rockway and his associates (Rockway, 1955;
Rockway, Eckstrand and Morgan, 1956). A two-dimensional
compensatoiy, zero-order tracking system was used with a 150-msec
exponential delay and a forcing function which was the sum of 0.05-, 0.08-,
and 0.18-Hz equal-amplitude sine waves. The forcing functions for the two
dimensions were identical but out of phase. Time on target was used as the
performance measure. Rockway (1955) trained three independent groups of
subjects for 25 one-minute trials of this task with control-display ratios of
1:3, 1:9, and 1:27, where a l:x ratio indicates that one degree of control
movement produced 0.0625x inches (0.16x cm) of cursor movement on the
display screen. Transfer was to the 1:3 ratio for another 25 one-minute
trials. The experiment of Rockway et al. (1956) was identical except that
transfer was to the 1:27 ratio.
Data from these two experiments are shown in Figure 10.10. Differences
between training performances are statistically significant and there is
statistically significant positive transfer in both experiments. The only
differential transfer effect is found with the first transfer trial of Rockway
(1955). This shows a brief but significant performance disadvantage for
groups trained with the more responsive plants (i.e. higher system gain)
versus those trained with the transfer plant. Otherwise there was no
differential transfer effect of switching to a more or less responsive plant.
There have been few explicit tests of delay or lag as a transfer variable.
This is surprising given that it is inherent in all controlled plants, that there
are straightforward (albeit costly) methods of minimizing delays in training
systems, and that it is a straightforward variable to manipulate.
Furthermore, its significance is becoming increasingly evident with the more
widespread use of computer-based training systems, especially those with
real-time, interactive visual systems for instruction of vehicle control.
In what appears to be the first investigation of lag as a transfer variable,
Levine (1953) examined the transfer effects of the quasi-transfer effects of
an exponential lag4 on one-dimensional, zero-order, compensatory tracking.
A sum of 0.005-, 0.083-, and 0.183-Hz sine waves was used as the forcing
function. Time on target was the dependent measure. Transfer was tested
to a long delay (3000 msec) from delays of 15,150,900,2100, and 3000

148
 Transfer fo r Flight Simulation

Figure 10.10
Transfer from 1:3 or a 1:27 control-display ratio

(Adapted from Rockway, 1955, and Rockway et al. 1961b)

msec and to a short delay (15 msec) from delays of 15, 900, and 3000 msec.
Levine's data are shown in Figure 10.11.
The most striking observation to be made on Figure 10.11 is the
difference in transfer effects from short to long versus long to short lags.
Transfer from a sluggish to a responsive system poor in relation to control
subjects who continue with the responsive system. However, there are no
differential effects in transfer from a sluggish to a responsive system. In
contrast to the work of Muckier et al. (1961a, 1961b), differential transfer
effects show up in transfer to a responsive but not to a sluggish plant.
Riccio, Cress and Johnson (1987) examined the effects of transport delays
with two-dimensional compensatory tracking. Two sets of plant dynamics
were used to establish a nontransfer variable. They were simplified flight
dynamics that represented a sluggish and a responsive aircraft and were
first-order in the roll axis and second-order in the pitch axis. Sums of ten
non-harmonically related sine waves were used for the pitch and roll forcing
functions. Quasi-transfer was tested from 50, 100, 200, and 400 msec to
50msec. Delays of 200 msec or more degraded training performance as

149
Aviation Instruction and Training

Figure 10.11
Transfer from short to long and from long to short control-display lags

(Adapted from Levine, 1953)

measured by RMS error from the assigned heading and altitude. The
degradation in transfer was less substantial and was clearly evident only in
heading control of the sluggish plant for early transfer trials following the
300 msec delay. Thus these data are consistent with those of Muckier et al.
(1961b) but not with those of Levine (1953).
Goldstein and Newton (1962) tested moment of inertia and speed of
system response as quasi-transfer variables with a one-dimensional
compensatory tracking task, a 0.006-Hz sine wave forcing function and
simplified submarine longitudinal dynamics (a slow responding, second-
order plant). Goldstein and Newton appear to have defined their variables in
terms of coefficient adjustments in their plant equations. From a behavioral
perspective, increases in moment of inertia should decrease responsiveness
(increase sluggishness) and would thereby be similar to decreasing the speed
of system response. It is evident that manipulation of system response had a
strong effect on training performance but manipulation of inertia did not,
thus suggesting that the inertia manipulation was not a potent manipulation.
Given the nature of these manipulations, it can be assumed that they had

150
 Transfer fo r Flight Simulation

quantitatively (but not qualitatively) differing effects on the system and that
the inertia manipulation was less extreme.
Transfer, which was tested to both fast and slow responding plants and to
plants with high and low moments of inertia was always positive. The only
differential transfer effects were found in transfer to a slow responding
system with a large moment of inertia from fast responding systems with
high or low moments of inertia. Transfer performances for these two groups
were poorer than for the respective control groups. Thus, these data are also
consistent with those of Muckier et al. (1961a, 1961b).
Some experiments have manipulated plant order5 as a quasi-transfer
variable. Briggs, Fitts, and Bahrick (1958), Dooley and Newton (1965), and
Holland and Henson (1956) examined the transition from first-, or
combination zero- and first-order compensatory systems to second-order
compensatory systems. Briggs et al. (1958) used a two-dimensional
tracking task with transient disturbances to one of four positions on the
display while Dooley and Newton (1965), and Holland and Henson (1956)
used a one-dimensional tracking task with a sine wave forcing function of
0.05-Hz. Transition to second-order tracking from lower order tracking
always resulted in some, but not complete transfer (Figure 10.12). Dooley
and Newton (1965) and Holland and Henson (1956) also transferred groups
trained on second-order tracking to the lower-order tasks. In this case there
was no differential transfer. In that lower-order plants are more responsive
than higher-order plants, these data are consistent with those of Muckier et
al. (1961a, 1961b).
Two quasi-transfer experiments have examined manipulations of system
order in a flight simulator (Lintem, Roscoe and Sivier, 1990; Lintem and
Koonce, in press). Subjects were taught visual landings with a zero-order,
normal first-order, or sluggish first-order roll response and were then tested
on the normal first-order roll response. Transfer from the zero-order
response was poor, but there was no difference in transfer following training
with the sluggish versus the normal first-order response. Thus, there is
evidence here of an assymetry. There was a loss in transfer effectiveness
followed training on a system that was more responsive than the criterion
system but not following training on a system that was less responsive.

Summary

The data of Muckier et al. (1961a, 1961b) indicate that phugoid response
(or frequency of natural oscillation) can be a significant transfer variable at
some levels of system gain. With low-gain systems it appears that fidelity in

151
Aviation Instruction and Training

Figure 10.12
Transfer between systems of different order

(Adapted from Holland and Henson, 1956)

the phugoid response is crucial. Transfer is poor from either a high to

medium or from a low to medium frequency. High gain can, however,
eliminate the effect. Muckier et al. (1961b) forwarded the hypothesis that
variations in system responsiveness during training can generate differential
transfer effects. Their data do suggest, however that these effects will be
found only with relatively sluggish transfer plants. Most of the data
reviewed here follow this pattern. The failure of Rockway (1955) and
Rockway et al. (1956) to find differential transfer effect may have been due
to the use of relatively responsive transfer plants although, given the
difficulty of assessing differences between different sets of experimental
apparatus, the characterization of these plants as relatively responsive must
remain a conjecture. Only the data of Levine (1953) do not follow the
hypothesis forwarded by Muckier et al. (1961b). There is, however, a
pattern in the data reviewed here that implicates responsiveness of the
training plant and of the transfer plant as important to transfer.

152
 Transfer fo r Flight Simulation

There is also evidence from Lintem, Roscoe, and Sivier (1990) and
Lintem and Koonce (in press) of asymmetric transfer effects from
adjustment in responsiveness of the training plant. The results showed
degraded transfer from a highly responsive to a moderately responsive
system. It is possible that the control of sluggish systems requires the
development of prediction or lead generation skills that cannot be learned
during practice with responsive systems. Thus operators may transfer
relatively poorly in the responsive to sluggish direction because critical new
learning is required. In contrast, those transferring in the sluggish to
responsive direction should not have to acquire new skills.
In summary the general conclusion to be drawn from the transfer research
with system dynamics is that variables which affect system responsiveness
affect transfer. There is, however, a possibility of asymmetry in the trends,
with degradation in transfer being more likely in transition from a responsive
to a sluggish system than from a sluggish to a responsive system.

Control forces and system dynamics

Muckier and Matheny (1954) examined control friction as a quasi-transfer

variable with a one-dimensional pursuit tracking task, a sine wave forcing
function (frequency unspecified), and a displacement control. Time-on-
target was used as the dependent measure. The frictional force in the
control was independent of magnitude or rate of displacement. Seven
independent groups of subjects were trained to a criterion6 of three errorless
(no time off target) cycles of the sine wave with 0.16, 2, 4, 6, 8, 10, or 12
pounds of frictional force, and transferred to 6 pounds of frictional force.
Transfer from lower or from higher frictional forces was reported as
showing slight but nonsignificant decrements, thereby suggesting the
interpretation that frictional variations (at least in the tested range) do not
affect transfer (Figure 10.13).
There are, nevertheless, two dimensions of concern with this study. One
is that friction was not shown to affect training performance. As friction is
known to affect tracking performance under at least some circumstances
(Poulton, 1974) it is possible that the manipulation used in this experiment
was not sufficiently potent to explore the issue of frictional effects on
transfer. In this case die variation of frictional forces was substantial, but it
is possible that a more complex or a different type of tracking task would
show some effects. A second concern is that, unaccountably, the statistical
test was performed on a data set that did not include the control group.

153
Aviation Instruction and Training

Inspection of Figure 10.13 suggests the possibility at least that an

appropriate test would have shown significant differences. Unfortunately,
the data provided by Muckier and Matheny do not permit reanalysis.

Figure 10.13
Percent transfer to a control with 6 lbs of frictional force from
higher and lower values of frictional force

(Adapted from Muckier and Matheny, 1954)

Matheny, Williams, Dougherty, and Hasler (1953) tested control force as

a transfer variable. Flight naive subjects were taught constant-rate climbs
and constant-rate descents in a P-1 flight trainer with either m in im u m
elevator control force or with an elevator control force that corresponded
closely to forces found in the North American T-6 SNJ aircraft. Training
was continued to a criterion of three errorless trials. Both simulator groups
were later tested in the T-6 aircraft as was a control group with no prior
experience in the P-1 trainer.
Only with the descent maneuver was there clear evidence of transfer to
the aircraft. With this task both simulator groups outperformed the control

154
 Transfer fo r Flight Simulation

group in the aircraft. There was no significant difference between the two
simulator groups either with the climb or the descent. Thus these data
indicate that the pressure manipulation did not affect transfer. As in the
Muckier and Matheny (1954) test of frictional forces, a concern is raised
that the variation in pressure did not affect training performances.
Nevertheless the task was relatively complex, and does represent the type of
tasks of concern in this review. If the extreme changes in forces tested by
Matheny et al. (1953) have no affect either on training or on transfer in this
situation, it seems unlikely that poor simulation fidelity on the stick pressure
dimension will have any adverse effects on transfer.

Summary

While the data relating to control forces are meager, they permit the
tentative conclusions that frictional and pressure variations have no effect on
transfer. It is, however, possible that flight experience is an important
variable here, since the control-force to aircraft-response relationships may
be subtle and require extensive experience to learn and use. This is a
neglected area and there remains a definite need for further exploration of
transfer effects from control manipulations.

Discussion

Considerable expense in terms of design and purchase of equipment is

required to ensure that a simulation provides a high-fidelity representation of
the operational vehicle. It is reasonable to inquire about the need for that
expense. Unfortunately, the answers that can be drawn from this review do
not have clear implications for the design of simulators. It is apparent that
there are large gaps in the data base. Not all effects of interest have been
satisfactorily examined, and there is the possibility of important interactions
between critical variables. Very few interactions have been examined and
many (indeed most) potential combinations of variables remain untested. If
transfer research is to contribute anything to simulator design, more
systematic and more controlled tests of important variables are essential.
In spite of the obvious limitations of the data, they at least suggest that
distortions of fidelity will not have an all-pervasive effect on training
effectiveness. The most obvious conclusion to be drawn is that the simple
view of transfer as being directly related to degree of physical similarity or
fidelity is untenable. Similarity and fidelity may be conceptually appealing

155
Aviation Instruction and Training

as basic behavioral design principles but they impose considerable costs and
technical demands on engineering development. It is apparent from this
review that uncritical adherence to principles of fidelity and similarity will
not necessarily maximize transfer. Noticeable departures from physical
fidelity or similarity, even along dimensions that seem central to the task, do
not always lead to a decrement in transfer.

Acknowledgments

Preparation of this paper was supported under USAF contract F33615-86-C-

0532 to Charles River Analytics. We were assisted by the comments of
John Flach, William Levison, Gary Riccio, and Greg Zaccharias, who
reviewed earlier versions of this paper.

Notes

1 This is a key methodological point that appears to have eluded some who
are involved in transfer research. Interpretation of transfer experiments
that use the alternate procedure of training to proficiency is problematic
because the criterion will be reached more quickly with an easy version
of the training task. A transfer disadvantage for that group could be due
either to less practice or or to the variable that has been manipulated.
The training data may be examined to ascertain whether or not there is a
potential confound, but the results of such an examination will sometimes
force a conclusion that the patterns in the transfer data cannot be
interpreted.
2 See Note 1. McDaniel et al. (1983) report correlations between amount
of simulator training and time to proficiency in the aircraft, the
correlations were small, thereby suggesting that the potential confound
between motion and amount of training in the simulator did not impact
the results.
3 The design of this experiment does not conform to the ideal for a test of
differential transfer in that the five independent groups were not given
equal training (see Note 1) and the control subjects were not tested in
transfer. Nevertheless the conclusions drawn from the data do not
appear to be severely compromised by these methodological variations
because the results do not follow a trend that could be explained by the
observed variations in the duration of training.

156
 Transfer fo r Flight Simulation

4 The time constant of the exponential lag was defined as that time taken to
achieve 63 per cent of the total output from a step input.
5 In this review, quickening is viewed as a manipulation of order. There
are important technical distinctions between these concepts that are,
however, irrelevant in synthetic tracking systems when there is no actual
vehicle output (Lintem and Gopher, 1978).
6 See Note 1. This potential confound is not a problem with the data of
Muckier and Matheny (1954) because the group differences in training
trials to criterion were slight and unsystematic.

References

Briggs, G.E., Fitts, P.M. and Bahrick, H.P. (1958), Transfer effects from a
single to a double integral tracking system’, Journal o f Experimental
Psychology, vol. 55, pp. 135-142.

Browning, R.F., Ryan, L.E., Scott, P.G. and Smode, A.F. (1977), Training
Effectiveness Evaluation o f Device 2F87E, P-3 Operational Flight
Trainer, TAEG Report No. 42, Training Analysis and Evaluation Group,
Orlando.

Dooley, R.P. and Newton, J.M. (1965), 'Transfer of training between

quickened and unquickened displays’, Perceptual and Motor Skills, vol.
21, pp. 11-15.

Flexman, R.E., Roscoe, S.N., Williams, A.C. Jr. and Williges, B.H. (1972),
'Studies in pilot training: the anatomy of transfer’. Aviation Research
Monographs, University of Illinois, Aviation Research Laboratory,
Champaign, vol. 2, no. 1.

Goldstein, D.A. and Newton, J.M. (1962), 'Transfer of training as a function

of task difficulty in a complex control situation’, Journal o f Experimental
Psychology, vol. 63, no. 4, pp. 370-375.

Gray, T.H. and Fuller, R.R. (1977), Effects o f Simulator Training and
Platform Motion on Air-to-Surface Weapons Delivery Training, AFHRL-
TR-77-29, Air Force Human Resources Laboratory, Brooks Air Force
Base, Texas.

157
Aviation Instruction and Training

Gum, D.R. and Albery, W.B. (1977), Time delay problems encountered in
integrating the advanced simulator for undergraduate pilot training',
Journal o f Aircraft, vol 4.

Holland, J.G. and Henson, J.B. (1956), 'Transfer of training between

quickened and unquickened tracking systems', Journal o f Applied
Psychology, vol. 40, no. 6, pp. 362-366.

Jacobs, R.S. (1976), Simulator Cockpit Motion and the Transfer o f Initial
Flight Training (ARL-76-8/AFOSR-76-4), University of Illinois at
Urbana-Champaign, Aviation Research Laboratory, Institute of Aviation,
Savoy.

Levine, M. (1953), Transfer o f Tracking Performance as a Function o f a

Delay Between the Control and the Display, Technical Report 55-237,
Wright Air Development Center.

Levison, W.H. (1981), Effects o f Whole-Body Motion Simulation on Flight

Skill Development, Report No. 4645, Bolt Beranek and Newman, Inc.
Cambridge.

Levison, W.H., Lancraft, R.E. and Junker, A.M. (1979), 'Effects of

simulator delays on performance and learning in a roll-axis tracking task',
Fifteenth Annual Conference on Manual Control, Dayton.

Lintem, G. (1987), 'Flight simulation motion systems revisited', Human

Factors Society Bulletin, vol. 30, no. 12, pp. 1-3.

Lintem, G. and Gopher, D. (1978), 'Adaptive training of perceptual motor

skills: Issues, results, and future directions', International Journal o f
Man-Machine Studies, vol. 10, pp. 521-551.

Lintem, G. and Koonce, J.M. (in press), 'Visual augmentation and scene
content effects in flight training', The International Journal o f Aviation
Psychology.
Lintem, G., Roscoe, S.N., Koonce, J.M. and Segal, L.D. (1990), 'Transfer
of landing skill in beginning flight training', Human Factors, vol. 32, pp.
319-327.

158
 Transfer fo r Flight Simulation

Lintem, G., Roscoe, S.N. and Sivier, J.E. (1990), 'Display principles,
control dynamics, and environmental factors in pilot training and transfer’,
Human Factors, vol. 32, pp. 299-317.

Lintem, G., Sheppard, D.J., Parker, D.L., Yates, K.E. and Nolan, M.D.
(1989), 'Simulator design and instructional features for air-to-ground
attack: A transfer study', Human Factors, vol. 31, pp. 87-99.

Lintem, G., Thomley, K., Nelson, B. and Roscoe, S.N. (1987), 'Content,
variety, and augmentation of simulated visual scenes for teaching air-to-
ground attack', Human Factors, vol. 29, pp. 45-59.

Martin, E.L. and Cataneo, D. (1980), Computer Generated Image: Relative

Training Effectiveness o f Day Versus Night Visual Sense (AFHRL-TR-
79-56), Operations Training Division, Air Force Human Resources
Laboratory, Williams Air Force Base.

Martin, E.L. and Waag, W.L. (1978a), Contribution o f Platform Motion to

Simulator Training Effectiveness: Study I - Basic Contract (AFHRL-TR-
78-15, ADA-058-416), Flying Training Division, Air Force Human
Resources Laboratory, Williams Air Force Base.

Martin, E.L. and Waag, W.L. (1978b), Contribution o f Platform Motion to

Simulator Training Effectiveness: Study II - Basic Contract (AFHRL-
TR-78-52, ADA-064-305), Flying Training Division, Air Force Human
Resources Laboratory, Williams Air Force Base.

Matheny, W.G., Williams, A.C. Jr., Dougherty, D. and Hasler, S.G. (1953),
The Effect o f Varying Control Forces in the P-1 Trainer Upon Transfer
o f Training to the T-6 Aircraft, Technical Report 53-31, Human
Resources Research Center, Lackland Air Force Base.

Muckier, F.A. and Matheny, W.G. (1954), 'Transfer of training in tracking

as a function of control friction', Journal o f Applied Psychology, vol. 38,
no.5, pp. 364-367.

Muckier, F.A., Obermayer, R.W., Hanlon, W.H. and Serio, F.P. (1961a),
Transfer o f Training with Simulated Aircraft Dynamics: I. Variations in
Period and Damping o f the Phugoid Response, WADD Technical Report

159
Aviation Instruction and Training

60-615(1), Behavioral Sciences Laboratory, Wright-Patterson Air Force

Base.

Muckier, F.A., Obermayer, R.W., Hanlon, W.H. and Serio, F.P. (1961b),
Transfer o f Training with Simulated Aircraft Dynamics: II. Variations in
Control Gain and Phugoid Characteristics, WADD Technical Report
60-615(11), Behavioral Sciences Laboratory, Wright-Patterson Air Force
Base.

Muckier, F.A., Obermayer, R.W., Hanlon, W.H. and Serio, F.P. (1961c),
Transfer o f Training with Simulated Aircraft Dynamics: III. Variations
in Course Complexity and Amplitude, WADD Technical Report 60-
615(111), Behavioral Sciences Laboratory, Wright- Patterson Air Force
Base.

Nataupsky, M., Waag, W.L., Weyer, D.C., McFadden, R.W. and

McDowell, E. (1979), Platform Motion Contributions to Simulator
Training Effectiveness: Study III - Interaction o f Motion with Field o f
View (AFHRL-TR-79-25), Flying Training Division, Air Force Human
Resources Laboratory, Williams Air Force Base.

Poulton, E.C. (1974), Tracking Skill and Manual Control, Academic Press.
New York.

Povenmire, H.K. and Roscoe, S.N. (1971), 'An evaluation of ground-based

flight trainers in routine primary flight training’, Human Factors, vol. 13,
no. 2, pp. 109-116.

Povenmire, H.K. and Roscoe, S.N. (1973), 'Incremental transfer

effectiveness of a ground-based general aviation trainer', Human Factors,
vol. 15, no.6, pp.534-542.

Riccio, G.E., Cress, J.D. and Johnson, W.V. (1987), 'The effects of
simulator delays on the acquisition of flight control skills: Control of
heading and altitude', Human Factors Proceedings.

Ritchie, M.L. and Hanes, L.F. (1964), An Experimental Analysis o f

Transfer Effects between Contact and Instrument Flight Training,
Federal Aviation Agency, Washington.

160
 Transfer fo r Flight Simulation

Ritchie, M.L. and Michael, A.L. (1955), 'Transfer between instrument and
contact flight training’, Journal o f Applied Psychology, vol. 39, no.3, pp.
145-149.

Rockway, M.R. (1955), The Effect o f Variations in Control-Display

During Training on Transfer to a 'High' Ratio, WADC Report No. 55-
366, United States Air Force, Ohio.

Rockway, M.R., Eckstrand, G.A. and Morgan, R.L. (1956), The Effect o f
Variations in Control-Display Ratio During Training on Transfer to a
Low Ratio, WADC Report No. 56-10, United States Air Force, Ohio.

Taylor, H.L., Lintem, G., Koonce, J.M., Kaiser, R.H. and Morrison, G.A.
(1991), 'Incremental transfer study of scene detail and visual
augmentation guidance in landing training', Proceedings o f the Sixth
International Symposium on Aviation Psychology, The Ohio State
University, Columbus.

Thorpe, J.A., Varney, N.C., McFadden, R.W., LeMaster, W.D. and Short,
L.H. (1978), Training Effectiveness o f Three Types o f Visual Systems fo r
Kc-135 Flight Simulators (AFHRL-TR- 78-16, ADA-060-253), Flying
Training Division, Air Force Human Resources Laboratory, Williams Air
Force Base.

Waag, W.L. (1981), Training Effectiveness o f Visual and Motion

Simulation (AFHRL-TR-79-72), USAF Human Resources Laboratory,
Operations Training Division, Williams Air Force Base.

Westra, D.P. (1982), Simulator Design Features fo r Carrier Landing: li.

In-Simulator Transfer o f Training (NAVTRAEQUIPCEN 81-C-0105-1),
Naval Training Equipment Center, Orlando.

Westra, D.P., Lintem, G., Sheppard, D.J., Thomley, K.E., Mauk, R.,
Wightman, D.C. and Chambers, W.S. (1986), Simulator Design and
Instructional Features fo r Carrier Landing: A Field Transfer Study,
(NAVTRASYSCEN 85-C-0044-2), Naval Training Systems Center,
Orlando.

161
Aviation Instruction and Training

Woodruff, R.R. and Smith, J.F. (1974), T-4g Simulator and T-4 Ground
Training Devices in USAF Undergraduate Pilot Training (AFHRL-TR-
74-78, ADA-008-197), Flying Training Division, Air Force Human
Resources Laboratory, Williams Air Force Base.

162
11 The technology of
part-task training
D.J. Allerton

In the past, flight simulators were constructed to resemble actual aircraft and
not surprisingly, were usually operated in a manner almost identical to the
operation of an aircraft. For procedural training (particularly in airline
operation), generating the illusion of flight can obviously improve the
acceptance of a training device and arguably, the quality of training. There
are also further advantages in using flight simulators which include increased
safety and potential reductions in training costs.
Flight simulators have, therefore, become accepted in airline training
because they reduce the cost of training, or they are safer, or they are more
effective than training in aircraft. However, flight simulators have been used
not only as an alternative to airborne training, they have also been used for
specific training tasks which are not necessarily best suited to training in an
actual aircraft. For example, engine-start procedures and engine fire drills
occupy a significant proportion of a training program and clearly, these
training tasks do not require a fully-equipped flight simulator with motion,
visual system and fully functioning avionics. For specific tasks it is
therefore possible to devise training equipment which is designed to provide
training which is optimized for that task (and is, therefore, inappropriate for
other tasks). These trainers are termed 'part-task' trainers.
There are two primary reasons for adopting the use of a part-task trainer.
Firstly, the training costs for a specific task may be significantly less than the
cost of equivalent training in a full simulator. If the part-task trainer needs
less equipment (and the reduction in fidelity is justified), then the
concomitant reduction in equipment costs can be reflected in the training
costs. Secondly, a part-task trainer may provide a higher quality of training
for a specific task than a full simulator, or even an aircraft.
Aviation Instruction and Training

This latter point is contentious: there is a natural assumption in aviation

that an aircraft is the ideal place to learn to fly. However, the aircraft
cockpit does not compare favourably with a classroom environment; there
are distractions from radio traffic and other aircraft, the cockpit is noisy and
subject to vibration, fuel management must be rigorously observed, aircraft
separation must be maintained and regulatory procedures must be observed.
These activities of airmanship may occupy the student's attention almost to
the exclusion of learning.
Although part-task trainers have been widely adopted, in order to provide
a cost-effective part-task training device, it is essential to match the
technology of the trainer to the training requirements. It is therefore
important to combine an understanding of the principles of flight training
with an appreciation of the capability of the computer systems and
electronics which constitute part-task trainers. The technology inherent in
simulation can then be viewed simply as a set of tools which facilitate the
design of flying training courses.

Computer technology

There have been remarkable developments in computing since 1970 in many

areas of technology:

- processor speeds;
- the capacity of memory chips;
- the capacity and transfer rates of secondary storage, particularly disks;
- system reliability; and
- a wide range of support chips for graphics, arithmetic and
communications.

More remarkably, these improvements have occurred in parallel with equally

dramatic reductions in the cost of computer systems. In the future, there
appears to be no apparent reduction in the rate of progress and very high
performance computer systems will be available for office applications
which are directly amenable to flight training applications.
These market-driven developments have given rise to the ubiquitous PC
and also to the workstation for CAD which have been adapted for flight
training applications where their low cost is combined with the processing
capacity and high performance computer graphics necessary for flight
simulation.

164
 The Technology o f Part-Task Training

For self-paced and non real-time learning, computer-based training (CBT)

systems using optical disk storage of both digitised and computer-generated
information provide a means of offering high resolution textual,
photographic and graphical information combined with a limited degree of
animation. The mainly static images afford teaching methods to present
repetitive and factual information. Examples of this approach are aircraft
engine-start procedures and descriptions of specific aircraft sub-systems.
The rate of learning can be monitored automatically and the information can
be structured and paced in a manner appropriate to the application.
However, the scope of CBT systems is limited by the lack of real-time
response. In many training instances, it is more beneficial to allow the
student to acquire knowledge in a 'sitting and thinking' mode. However, in
training situations where the pilot is ϊη-the-loop' and needs to respond to
situations in real-time, the limitations of CBT systems caused by lack of
graphics performance or computer processing power may be counter­
productive.
On the other hand, systems which have sufficient processing and graphics
capability, offer a much broader range of applications in flight training.
These part-task training systems provide dynamic models of aircraft
systems, animated displays and possibly visual systems. They bridge the
gap between the classroom (ground-school) and the full simulator (or
aircraft). Real-time part-task trainers may lack the fidelity of full simulators
but their simplicity offers two advantages. Firstly, the software (or
courseware) can be readily programmed and secondly, the software can be
designed and optimized for very specific tasks.
Microprocessor technology, therefore, offers the opportunity to design
low-cost part-task trainers with much of the capability of full simulators but
with this capability restricted to specific tasks and with an acceptance that
fidelity is compromised by the reduction in cost.

Training analysis

Prior to the advent of the part-task trainer, flight training devices were
limited to very restricted classes of training systems. Typically, a training
organization procured a training device within their financial budget and
then adapted their training courses to match the characteristics of the device.
With the availability of part-task trainers, the capability of these devices is
more diverse and there are more degrees of freedom in specifying the
desired training characteristics. Consequently, there has been a change of

165
Aviation Instruction and Training

emphasis from matching a training program to a specific device towards

matching the characteristics of a training device to the training requirements.
This is an important development as it leads to three consequences.
Firstly, it assumes that a training organization has the experience to analyse
and specify its training requirements. While training organizations may be
experienced in the provision of a wide range of flight training courses and
meeting the training requirements of regulatory authorities, the skills needed
for course design are not covered in conventional courses for flying
instructors, which tend to concentrate on the presentation of material and
methods of instruction.
Secondly, it is necessary to evaluate the training benefits and limitations
of a new training device, possibly incorporating state-of-the-art computer
systems. For example, what is the effect on training of a less-than-rigorous
flight model or a 'low-cost' control-column that binds? The scope and
applicability of training devices is influenced by visual content, visual
resolution, effects of aliasing in visual systems, effects of texture, update
rates, fidelity of avionics and so on. It is very difficult to isolate these items
and to relate their effects to specific training applications. Part-task trainers
are complex systems and the underlying technology is changing rapidly.
Accordingly, it is necessary for training organizations to closely 'track'
advances in training technology.
Thirdly, it is important to estimate the training effectiveness of a synthetic
training device. It is unlikely that the manufacturer can offer this
information and moreover, the effectiveness depends not only on the
characteristics of the equipment, but also on the manner in which it is
operated. The standard method of assessment of training effectiveness is to
conduct training trials, but this is unlikely to be feasible at the start of a
procurement program. One further disincentive in undertaking a training
analysis is the consideration of the cost of trials. It may appear irrelevant to
undertake trials on a $10,000 part-task training device despite the possibility
that the use of the device could have a significant negative effect on a
training program.

Training equipment

Part-task trainers are manufactured by simulator companies who have

developed their product to meet the market needs. Training organizations
are confronted by a mass of technical specifications and promotional
information so that it is far from clear how the advantages of a part-task can

166
 The Technology of Part-Task Training

be assessed or more significantly, how two part-task trainers can be

compared. This section reviews the components of a part-task trainer and
outlines their significance in the context of training effectiveness.

Trainer environment

It is recognized that the quality of the teaching environment has a direct

impact on the quality of training. It is essential to minimize distractions to
enable students to concentrate on a learning task. For example, in a CBT
classroom, the main objectives are to provide a comfortable environment in
terms of noise, temperature and light to enable a student to concentrate on
information displayed on a PC monitor.

Computer technology in flight instruction

(Photo: Swissair)

With CBT systems, the student is interacting with a software package and
its user interface. For example, a student may need to turn a radio knob
displayed on the screen and this can be achieved by moving a mouse and by
selection of mouse buttons. The emphasis is to replicate the functionality of
a system which is appropriate to the specific task. The user interaction with

167
Aviation Instruction and Training

the CBT package should have minimal impact on subsequent transfer to

actual equipment. In this case, students are not being trained to turn a knob,
they are being trained in the procedures of frequency selection. Of course, if
the actual selection and operation of a radio was significant (possibly
selecting a hidden knob or using a multi-function switch on a side-stick),
then it would be appropriate to incorporate synthetic radio controls.
If the training involves flying procedures, it is necessary to provide an
acceptable aircraft seat and 'realistic' flying controls. It is also important to
exclude extraneous external references, particularly noise, unnecessary light
and distractions. This is usually achieved by the provision of an aircraft-like
cockpit. Unlike an aircraft cockpit, it must accommodate an instructor
together with the instructor station, and provide straightforward access. A
common assumption is that aircraft parts can be readily adapted for use in
simulation. Typically, this is not the case, because the equipment may
require an unusual power supply or the costs of using equipment which has
been designed to satisfy airworthiness considerations may be prohibitive.
Crashed aircraft do not make low-cost simulators and (thankfully) there is an
acute shortage of written-off aircraft.
The level of appointment also has a drastic impact on cost. Unless the
trainer is required for type-specific training, it may well be sensible to equip
the trainer as a generic trainer. For example, a park-brake can be provided
in the form of a panel push-pull switch rather than a mechanical lever.
Irrelevant equipment can be omitted and the equipment can be adapted (or
matched) to the training requirements.
However, it is difficult to establish an acceptable balance between cost
and acceptable fidelity, and this is mainly influenced by the application. For
example, there are no clear training benefits in replicating a curved
windscreen whereas the cost of fabrication of a curved windscreen may add
considerably to the costs of manufacture. A further example is the use of
engine levers. Sliding potentiometers can be used to reduce the cost of
engine levers, but they offer a very different feel and layout from
conventional engine levers and, if engine operation constitutes an essential
part of the training, it is essential to provide representative engine controls.
However, if the trainer is simply used as a navigation trainer, where the
simulated aircraft is operated in the climb, cruise or descent modes, then the
engine controls can be simplified.
One further misconception is that costs can be saved by the use of an
office seat rather than an aircraft-like seat. Pilots spend most of their flying
hours in an aircraft seat and considerable attention is given to the quality of
aircraft seating. The same situation applies to a flight trainer and there are

168
 The Technology of Part-Task Training

no benefits in reducing the quality of a seat. Most aircraft offer vertical and
fore/aft adjustment and this feature must also be incorporated in a trainer
seat. The mechanical considerations to accommodate 'percentile' pilots are
formidable and lead to very expensive seats. The provision of arm rests is
also frequently overlooked. Aircraft are provided with arm rests to reduce
pilot fatigue and to improve control sensitivity as the hand is then pivoted
about the elbow rather than the shoulder. From observation, a flight trainer
without an arm rest is much harder to fly, in terms of accuracy, than a trainer
with an adjustable arm rest.

Balancing cost and acceptable fidelity in part-task training

(Photo: Cranfield Institute of Technology)

Flight controls

Flying can be viewed as a feedback control system. The pilot observes the
aircraft outputs, that is, the visual cues and aircraft instruments. Analysis of
these outputs by the pilot results in the application of inputs to the aircraft
controls to control the aircraft attitude and performance.
There are two points to be drawn from this simplistic model. Firstly, a
pilot-in-the-loop is part of a real-time control system. Pilot response to

169
Aviation Instruction and Training

certain visual cues may be less than one tenth of second. Secondly,
considerable attention is given to the mechanical design of aircraft controls
to assure that they are sufficiently robust to cater for large inputs while
providing adequate sensitivity for small inputs. Feedback in an aircraft
comes from a variety of sources including visual reference, aircraft motions
which are detected by a pilot's sensory organs and tactile feel of the primary
flight controls (particularly from the control column). A flight training device
must replicate these feedback cues in applications where they are
appropriate to the training requirements.
However, for fixed-base trainers the only sense of motion is derived from
visual cues. In addition, the loading forces imposed upon a control surface,
as a result of a manoeuvre or the dynamic pressure resulting from the
airspeed, generate forces on the control column. In large simulators, these
forces, which are termed 'stick forces', are replicated by means of
computer-controlled hydraulic systems. In a part-task trainer, it is not
economical to install a complex hydraulic system and stick forces are
invariably replicated by simple spring-damper mechanisms.
As the column is pulled or pushed or rotated, the tension on a spring is
increased and this replicates the aerodynamic resistance of a control surface.
Similarly, control surfaces consist of relatively large surfaces and the inertia
of these surfaces is replicated by damping applied in parallel with the spring.
Although this arrangement is commonly used in part-task trainers, there are
several important limitations with this approach:

1. The dynamic pressure or 'q-feel' cannot be replicated by this means

unless the spring length can be altered to simulate changes in dynamic
pressure. This arrangement is complicated and it is difficult to provide
small changes of dynamic pressure over a large range. The alternative
solution is to provide a representative range of control column feel
covering aircraft speeds from zero to Vne and consequently, many
spring-damper versions of an aircraft control column are representative
of the feel of an aircraft under cruise conditions.
2. The relationship between stick position and stick force is very much
simplified by a model based on a linear spring arrangement and
typically, the match between stick force and displacement is simply
achieved by the selection of an appropriate spring.
3. Trimming an aircraft is a fundamental part of basic flying. In a
simulator, the position of a control surface is converted from a
mechanical displacement to an electrical signal which is read into the
flight model computer. For some trimmed state, there is a 'hands-off or

170
 The Technology o f Part-Task Training

null position of the control column where the effect of the primary
control is compensated by a trim control. In a part-task trainer, this is
accomplished in one of two ways. Either the trim wheel moves the
mechanical assembly to relieve the spring forces on the control column
and the mechanical displacement of the column is transferred to the
flight model computer as an input signal. Alternatively, if the trim
wheel is mechanically decoupled from the control column, the number
of turns of the trim wheel can be converted to an equivalent mechanical
displacement which is added to the displacement signal from the
control column. The latter method, although significantly cheaper,
necessitates that the control column always returns to the same
mechanical datum point, whereas, for most aircraft, the trimmed
position depends on the aircraft manoeuvre.
4. In a trimmed state, and this represents the major part of flying, pilots
make very small inputs, often returning the control column to the
trimmed 'hands-off position. If the control column of the trainer binds
or sticks, then it cannot return accurately (and repeatedly) to this
trimmed position, resulting in an undesirable increase in pilot workload.
5. In an aircraft, flight controls are coupled to control surfaces by
mechanical linkages. In a flight trainer, the control input displacements
are converted to voltages which are input to the flight model and errors
can also arise as a consequence of the resolution of analogue-to-digital
conversion, delays in sampling the controls and electrical noise.

The effects of under estimating the importance of flying controls in a

real-time flight trainer can be dramatic. If the control forces are inadequate,
pilots tend to 'push-and-pull' the controls as a result of lack of mechanical
feedback. If the sensitivity of trimming cannot be felt at the control column,
pilots will apportion an unrepresentative amount of time to trimming. If the
control column binds, particularly around the centre datum where the spring
force is negligible, there will be a noticeable reduction in flying accuracy.
These observations apply mainly to the control column, but they also
apply to rudder pedals, brakes, flap levers and engine levers. Care must be
taken that negative transfer does not result from control forces in the trainer
which are very different from the control forces in an aircraft and to ensure
that the method of trimming is compatible with the aircraft.

Flight model

The dynamics of the aircraft and its engines are modelled by means of a

171
Aviation Instruction and Training

computer program. This operation is repeated at a relatively high and

regular repetition rate (typically 30 to 60 times per second) so that the
motion of the aircraft appears to be continuous and to closely follow pilot
inputs. This requirement for real-time response is critical, and is often
overlooked in part-task trainers. The corollary of the requirement for real­
time operation is that the flight model computer must have a very high
performance to sustain the repetition rate.
A lack of real-time response is evident in many computer games, claiming
to provide a flight simulator on a home computer'. Experienced pilots have
difficulty in flying these 'simulators', primarily for the reason that the rate of
update is often less than 2 Hz. As a consequence, pilot inputs are not
executed (or fed back) for at least a second or two, by which time extra
input has been applied to correct the previous input, and so on until the
situation becomes unstable, particularly in critical phases such as landing.
Interestingly, non-flying enthusiasts often perform remarkably well with
these computer games, partly as a result of the hours of practice, but also
because they learn to compensate for the lags by the application of
feed-forward, albeit in the form of manual input.
The other requirement of the flight model is accuracy, or fidelity. There
are three causes of inaccuracy in flight modelling:

1. Incorrect models (or equations) of aircraft dynamics: this may come

about from a lack of understanding of aerodynamics or simplification of
the flight model owing to limitations of computing power. These
effects may be obscure and the limitations may not be relevant to the
training application. However, poor flight models lead to irregularities
in the apparent handling qualities of the flight trainer. Usually, it is
difficult for a flying instructor to isolate the problem - the instructor
simply knows that the simulator does not 'fly like the aircraft'.
Generally, these problems can only be detected by trained test pilots
and simulator engineers and even then, can be difficult to quantify.
2. Lack of data: this may necessitate estimation of certain critical aircraft
parameters. A very simple flight model requires at least fifty numerical
values which are substituted into the flight equations. These values
may not be available from the aircraft manufacturer or from flight trials
or wind-tunnel data and if they are, they may be prohibitively
expensive. This data should also include test data for validation of
aircraft response and performance, in order to give some degree of
confidence in the fidelity of the flight model. There is an assumption
that all this information is contained in an aircraft handbook.

172
 The Technology o f Part-Task Training

Unfortunately, this is not so and good aircraft data is difficult to acquire

and may possibly cost far more than the trainer!
3. Numerical computing errors resulting from the computational methods
used to solve the equations of motion: these errors include algorithm
approximation (for example, look-up tables for complex functions),
rounding errors and truncation errors.

It is important to stress that flight models are complex and, although they
can be simplified for certain training applications, to cater for all modes of
engine operation, ground handling, stalls, operation near the flight envelope,
effects of flaps and gear, atmospheric effects and so on, leads to very large
software program. Added to this complexity to solve all the equations and
axes transformations is the need to maintain the solution rates at 50 Hz and
to integrate the flight model with the visual system, instruments displays,
noise generation system, navigation systems and avionics. To summarize,
flight models are complex software programs and, for certain training
applications, their fidelity may be critical to the transfer of training for a
specific task.

Instrumentation

Aircraft instruments take the form of either electro-mechanical instruments

designed to operate in a simulator environment rather than an aircraft or
computer displays to mimic aircraft instruments in the form of computer
graphics. The former approach offers a high degree of fidelity but has the
drawback that these instruments are expensive in terms of unit cost and
maintenance. The latter approach has the attraction that displays can be
readily programmed and that there are no 'moving parts' in a CRT display,
while they suffer from the drawback that, even with high resolution graphics,
a CRT display cannot fully replicate a mechanical instrument.
Generally, the functionality of an instrument can be readily incorporated
and complications only arise where there is a need to simulate a lag (such as
a vertical speed indicator), or the instrument has unusual failure modes or
the instrument has several knobs for setting and selection. Increasingly,
CRT displays are being used in part-task training because they offer a
degree of flexibility that is simply not available with 'conventional'
instruments.
The primary disadvantage with the use of computer-generated instruments
lies in the graphical presentation. An instrument such as an HSI (Horizontal
Situation Indicator) requires a considerable amount of computation and

173
Aviation Instruction and Training

graphical operations. For example, in rotating the HSI by one degree, in

order to simply redraw the compass card, 36 lines and 18 characters must be
erased and then re-drawn in a new position. These operations necessitate
the use of special-purpose graphics processors in order to update the display
at the minimum rate to maintain the 'smooth' movement of the instruments.
If the displays are not maintained at these rates, the movement of a
pointer may appear to be jerky and a pilot's instrument scan will be
disrupted by sudden unexpected movement. An update rate of twenty five
times per second (25 Hz) is the minimum rate to avoid a noticeable level of
flicker or jerkiness in displays, and this generally precludes the use of
proprietary PC (VGA) displays from applications of real-time part-task
training. The display update rate should not be confused with the video
refresh rate. CRT displays are refreshed to 50 Hz for non-interlaced
displays and 25 Hz for interlaced displays (in the USA, the respective
refresh rates are 60 Hz and 30 Hz). Although update rates may be 'slaved' to
refresh rates, the refresh rate influences overall screen flicker, whereas the
update rate influences the level of flicker of animated objects.
A second problem arises from the use of computer graphics to represent
aircraft instruments. A screen has a finite resolution in terms of the grid of
dots (termed pixels) which constitute the display. All lines are drawn on this
implicit grid and lines of shallow angle may exhibit a jaggedness, termed
aliasing. A single line may appear to be made up of two, or three or four
segments and the number of segments will change as the slope of the line
changes. This problem can be disconcerting, and moreover, anti-aliasing
hardware to overcome this problem is expensive. For example, the number
of jagged segments of the zero degree pitch line of an attitude indicator will
increase as the angle of bank increases. A student could therefore use this
information as a method of 'tuning' the angle of bank and unwittingly
introduce negative transfer into instrument flying tasks.
Again it is essential to match the facilities of a part-task trainer to the
training requirement. Even if the pointer of an airspeed indicator appears
jagged from time to time, provided the student can read airspeed to the
nearest knot, this limitation should not prove restrictive. Similarly, it is not
possible to mount the barometric pressure knob of an altimeter on the glass
screen of a display, yet it is still possible to use a graphic display of an
altimeter with a separate knob for barometric pressure in order to offer the
same operational procedures as an aircraft.

174
 The Technology of Part-Task Training

Functionality in instrumentation in the Aerosoft-200

(Photo: Cranfield Institute of Technology)

Visual systems

Visual systems in large-scale simulators have, up to the last few years, been
developed as special-purpose systems by simulator manufacturers. The
typical cost per channel (or per windscreen) may exceed one million dollars.
These displays are based on surveyed databases of specific airports. They
include effects to mimic haze, fog, visibility and the time of day. The
addition of texture considerably enhances the realism of objects and
anti-aliasing hardware eliminates jagged edges. Typically, images are
rendered at 50 Hz and may contain several thousand objects. Interestingly,
many of these features are demanded by airlines who operate their
simulators for the majority of the time where the pilot is flying under IFR
conditions and not 'looking out of the windscreen'.
Visual systems are used in part-task trainers and, in order to reduce the
cost of the visual system to a reasonable level, several technical
compromises are made. Firstly, the number of objects can be reduced to a

175
Aviation Instruction and Training

level that is computationally feasible. A display of fifty objects can offer a

remarkable degree of realism for certain phases of training. Although
texture enhances depth cues, it can be omitted and effects of aliasing are
naturally reduced by 'blending' colour boundaries.
Taking all these considerations into account, together with recent
developments in VLSI chip sets for graphics, real-time visual systems for
under $50000 are widely available, as a result of technical developments in
workstations and PCs. Although the application of real-time visual systems
will spread rapidly as costs continue to fall, the characteristics and
limitations of visual systems in full flight simulators are the same as those in
part-task trainers:

1. The computations inherent in three-dimensional graphics and the

problems of rendering an image, particularly in-filling (colouring)
displayed objects are demanding and lead to a compromise between the
hardware costs and the number of displayed objects, in order to
provide a low-cost visual system and a reasonable update rate.
2. There is an unavoidable lag between pilot input and the actual display
of the image which may be as large as one tenth of a second. These
lags, termed transport lags, may be reduced for specific visual systems
but for certain applications may be unacceptable.
3. The display is presented to the pilot as a two-dimensional
computer-generated image. In order to add depth to the image it is
possible to project the image (either by collimated optics or direct
projection) so that the focal length of the displayed object is effectively
infinite. If an image is generated directly on a monitor, the depth cues
perceived by a pilot are misleading and tend to confuse. However, for
all methods of projection, it is imperative that the angular field-of-view
subtended by the display corresponds to the physical field-of-view seen
by the pilot.
4. The visual system must be fully integrated with the navigation systems,
atmospheric model and the flight model. For example, in breaking out
of cloud during an approach, the runway should appear to have the
correct perspective, orientation and size, in terms of QDM, length,
width and altitude. The incorporation of a full database of runways and
navigation aids for a large geographical region is non-trivial and in a
few trainers, the navigation database is unreasonably reduced to a
handful of 'artificial' runways and navigation beacons.
5. One channel of visual system is the equivalent of one windscreen and
typically subtends a horizontal angle of forty to fifty degrees. The

176
 The Technology o f Part-Task Training

provision of extra channels necessitates the addition of additional

display hardware. There are problems with the use of collimated
projection systems in that the pilot must be positioned close to the
centre-line of the display. Although these limitations have been
overcome in recent projection systems for full-scale flight simulators, a
limited field-of-view can restrict the applications and give an unnatural
(even claustrophobiac) effect in the trainer cockpit. However, a visual
system can extend the range of applications in flight training and is very
much an asset in part-task training. Nevertheless, the technical features
and limitations of a visual system can have a major impact of the effec­
tiveness of training and careful consideration must be given to matching
the technology of a visual system to the training requirements.

Instructor stations

Instructors have a range of roles in flight simulators and trainers:

1. They instruct the pilot under training in very much the same way that a
flying instructor instructs in an aircraft - by demonstration and
explanation. The existence of a 'freeze button' in a simulator adds an
extra dimension to the effectiveness of training.
2. They can set up a training scenario which is appropriate to the specific
task. This requirement ranges from establishing an aircraft on a
specific approach through to the introduction of turbulence, winds and
engine failures.
3. They monitor the performance (and progress) of the pilot under
instruction. This may include following an aircraft track during a hold
or an approach or simply observing pilot handling and the aircraft
instruments.
4. They 'perform' specific roles to enhance the training situation, for
example as a first-officer or an air-traffic controller or other aircraft.

There is a need is to incorporate the instructor in the training situation in an

optimal manner and considerable attention is therefore given to the features
provided at the instructor station, which nowadays is likely to be a PC or
workstation with a touch-screen or a mouse. There is no merit in cataloguing
all the functions to be found in a typical instructor station, but it is important
to understand the needs of the simulator instructor and the underlying
considerations of the facilities in the instructor station:

177
Aviation Instruction and Training

1. The user interface should be friendly and easy to learn. For a part-task
trainer, it may be most effective for many instructors (and possibly
some students) to use the trainer without the supervision of a simulator
instructor. The selection of commands should be straightforward and
regular, that is, commands should be sensibly grouped and the means
of selecting related commands should be consistent.
2. The commands offered should enable the instructor to perform all the
functions required in a simple and straightforward manner. Consider
an example; setting a cloud base of 500 feet QFE. The choice of menu
commands and selection should take only a few seconds. In this
environment a keyboard is counter-productive and touch-screens with
hierarchical pop-up menus and sliding scales are commonly used.
3. The instructor station should not interfere with the training role. In
other words, it should not intrude into the training session and the pilot
under instruction should be unaware of the instructor station and its
operation by the instructor.

Instructor stations have, for many years, been very much an after-thought in
simulator design. They had a reputation for being difficult to learn and
confusing to operate. Recently, with the availability of windows, menus and
touch-screens, their effectiveness in the simulator environment has increased
significantly. There is a trend manufacturers to continually add new features
to the instructor station, while most instructors generally use a handful' of
common commands. It is difficult to quantify the effect of an instructor
station on training but it can certainly have a major impact and is another
component of the training tools that need to be matched to the training
requirements.

Applications

In introducing a part-task trainer into a flying training organization, there are

two initial stages of analysis that need to be undertaken by the training
organization. Firstly, the characteristics of the part-task trainer must be
assessed and fully understood so that it can be best incorporated in the flight
training programs. Secondly, the syllabus must be reviewed to examine
areas of the syllabus where the use of part-task methods can improve the
quality of training. This latter change can lead to a major review and
re-design of both the training program and also the training methods.
One other consideration in embarking upon a part-task approach to

178
 The Technology o f Part-Task Training

training, is that computer-based teaching support can be progressively

removed as the student progresses, up to the point where all artificial
prompts are removed and the student is operating with training displays
which mimic actual systems, (in terms of function) facilitating subsequent
transfer to specific aircraft equipment. An example of this activity is to
provide a navigation display in the form of a plan view containing sufficient
information to enable a student to orientate directly from the display.
Eventually, prompts showing intercept tracks, wind vectors and so on are
removed to the point that the display corresponds directly to an HSI. By
means of a programmed set of exercises, a student can acquire a new skill
without being aware of the sub-tasks needed to acquire this skill.
This section illustrates several applications of a part-task across a broad
spectrum of applications of flight training to give an insight into the potential
of this relatively new approach to flight training. The examples are based on
the Aerosoft-200 flight trainer, which is a fixed-base part-task trainer with a
computer-generated instrument display and a relatively simple collimated
visual system. The set of exercises was developed in conjunction with
training organizations using the Aerosoft-200 and make use of animated
diagrams, based on the concept of part-task training as part of an integrated
flight training course.

IFR training

This is the most widespread application of the synthetic trainer. The flight
trainer is operated as an aircraft under IFR conditions. The obvious benefit
of this form of training is that many hours spent 'chasing needles' can be
replicated in the trainer rather than in a twin-engine IFR equipped aircraft
flying in controlled airspace. However, even with the one-for-one
replacement for flying hours (and in the UK, the Aerosoft-200 is accredited
by the CAA towards logged IFR time), a synthetic trainer can be used
effectively as a part-task training device in this role with a few very simple
additions:

1. A 'freeze' button allows the instructor to suspend a session and brief the
pilot under instruction and then to resume the session at the interrupted
point. This mode combines real-time flying (with its attendant
pressures during instruction) with non real-time briefing, in a relaxed
and constructive situation which is clearly not feasible in an aircraft.
2. The use of a simple auto-pilot facility can significantly reduce pilot
workload during instruction. For example, a hold pattern can be flown

179
Aviation Instruction and Training

in 'speed-hold' and Tieight-hold' modes so that the pilot under

instruction can concentrate on the task of turning at the appropriate
parts of the hold.
3. The aircraft track and position can be displayed for the pilot under
instruction so that initially he is fully aware of his orientation. For
example, the aircraft autopilot can select speed, height and heading
holds, so that a student making an intercept to a VOR can observe the
localiser response at different angles of intercept and at different
distances and relate these situations to the aircraft position displayed on
a chart.
4. Progressively, workload is increased until the student pilot can cope
with cross-winds, turbulence and simulated ATC requests to change
altitude and so on. In this way, the difficulty of the training task is
matched to the student's rate of progress.

Effects o f controls

A part-task trainer can be used to introduce the basic elements of flying.

This does not imply that the simulator is replacing flying time - it is simply
that 'intelligent' use of a simulator (in effect, as a briefing tool) can enable an
instructor to make more effective use of airborne time. Consider the
objectives of the first lesson:

- To illustrate the effects of 'stick and rudder’.

- To introduce the rudiments of attitude flying, based on the maxim that
'power + attitude = performance' (Kilowatts + radians = metres per
second!).
- To introduce the need for trimming.
- To practice the basics of attitude flying, typically for straight-and-level
flight.

A simple visual system consisting of distant hills, a horizon between land

and sky and a few fields provides sufficient visual cues for these training
tasks. By setting the autopilot heading-hold, it is possible to concentrate on
pitch response. Furthermore, by overlaying pitch lines, the aircraft pitch
attitude can be directly observed.
The student can experiment with stick displacement fore and aft to
comprehend the effects of elevator input. Then, by selecting desired pitch
attitude (by placing the horizon on a specific pitch index) it is possible to
illustrate that a specific attitude (with a constant power setting) will lead to a

180
 The Technology o f Part-Task Training

desired performance and that for every attitude there is a unique resultant
performance. Moreover, this steady-state condition occurs after several
seconds and there is no benefit to be gained from 'chasing' a performance
instrument. Indeed, in order to demonstrate the limitation of poor flying
technique, the student can be allowed to experiment by 'chasing' an airspeed
or rate of descent, under the supervision of an instructor in the trainer.
The student is then asked to select an attitude to meet a specific
performance. Progressively, as improvement takes place, the heading-hold
can be removed and then the additional pitch reference lines can be
withdrawn. Finally, the student is asked to select specific attitudes to obtain
a desired aircraft performance with the aircraft in a trimmed state.
During these exercises, which can now extend to simple turns, the hills
provide heading cues, the 'perfect' horizon provides an ideal attitude
reference and the trainer is used in a mode which is closer to ground-school
briefing than actual flying. However, in the first airborne lesson, a student
exposed to this set of part-task exercises will be looking for the horizon on
the aircraft windscreen as the attitude reference, will select attitudes rather
than chase performance instruments and the student will be keen to ensure
that the workload is kept to a minimum by maintaining the aircraft in trim.
To summarize, the student's quality of flying in that first lesson will increase
dramatically; the student is consolidating knowledge in the aircraft rather
than acquiring it!

Turns

During a level turn, the pilot is balancing the force generated by the lift from
the wings with the mass of the aircraft. If a diagrammatic view of the
aircraft (as seen from behind) is displayed, with the lift force depicted as a
vector perpendicular to the wings, then the instantaneous resolution of the
vertical components of force can be readily understood. This diagram
illustrates the need for extra back pressure on entering a level turn to
increase the vertical component and similarly, reduction of back pressure
coming out of a turn. It also shows the high amount of lift needed to sustain
level flight at high angles of bank.
There are two points to observe from a diagram of this form Firstly, the
student is participating in the 'experiment' in real-time and secondly, it
introduces concepts of force, mass, velocities and attitude, bridging the gap
between ground-school and flying.

181
Aviation Instruction and Training

Lift and drag forces

If a 'side-on' view of the aircraft is displayed, with the lift vector pointing
upwards, the mass vector downwards, the engine thrust vector to the left and
the overall drag vector to the right, then the balance of forces can be
demonstrated for climbing, descending, accelerating and de-accelerating
flight. Moreover, if the angle of the aircraft to the horizontal depicts the
angle of attack, then aircraft attitude in stalls and in different configurations
can be readily demonstrated. The availability of a dynamic lift vector also
introduces the concept of load factor. For example, if the aircraft is set up in
a high speed stall and then frozen, the lift vector at the time of the stall can
be shown at different airspeeds to allow a student to relate stalling to angle
of attack rather than airspeed.

The approach

Many students find lining up and maintaining a straight-in approach one of

the hardest aspects of basic training. The provision of a runway with a
correct aspect ratio can introduce the basic idea of runway perspective. For
example, the instructor can deliberately position the aircraft too low or too
high. This training aid can be further enhanced by displaying an extended
centre-line emanating from the runway and by displaying the resultant
aircraft heading vector on the 'ground'. By this means, a student can practice
the task of lining up on an approach and start to understand the idea of
regaining the centre line by small intercepts.

Aim point

The concept in 'The approach', outlined above, can be further extended by

adding a 'velocity vector' to the display to illustrate the actual point of
impact. During an approach, the runway should be continuously in the centre
of the 'visual flow' perceived by the pilot. By overlaying the visual display
with a velocity vector, students have confirmation that they are flying a
consistent approach. Moreover, changes in attitude and changes in power
are readily observed by changes in the point of impact.
This support does not teaching flying technique, it gives students
confidence that they are performing the task acceptably. Arguably, this form
of visual feedback can be removed when the students are flying within
desired limits and then re-introduced when they are outside these limits.
The velocity vector can also be augmented to demonstrate the effects of

182
 The Technology o f Part-Task Training

cross winds. In addition to the velocity vector derived from the aircraft
heading, the vector caused by the wind can be added to the heading velocity
vector to illustrate the resultant track vector. For example, students can fly
an approach with different cross winds and lay off the drift by placing the
resultant cross on the runway. The diagram also serves to illustrate the
need to maintain the 'crab into wind' on the transition from VFR to IFR when
the runway will appear visually towards one side of the windscreen.
Again, these part-task methods re-enforce basic training in cross-wind
landings. The flying techniques and procedures are still taught in the aircraft
but the knowledge that the student has gained in a part-task flight trainer are
transferred to the airborne exercise.

ILS approaches

Students learning to fly Instrument Landing System (ILS) approaches have

to grasp two basic concepts. Firstly, the presentation of an ILS display
gives only an instantaneous indication of the position of the aircraft relative
to the desired (or ideal) position. Even if the localiser and glide-slope
needles are centred, unless the aircraft has the same heading as the runway
QDM and the same rate of descent as the glide-slope, the aircraft will
quickly diverge from the glide path. Secondly, when the aircraft is off the
localiser or the glide-slope, the pilot is required to 'calculate' the change in
heading or rate of descent to re-acquire the ILS.
For many students, these simple ideas are far from straightforward and
there is an almost overwhelming desire to 'centre the needles' by following
them blindly as commands. In an aircraft some ten miles from an actual
runway, it is difficult to demonstrate the spatial changes relative to the glide
path. However, in a part-task trainer it possible to overlay the visual scene
with an actual projection of the localiser and glide-slope. Indeed, computer
graphics enable the projected 'cone' to terminate 100 metres in front of the
aircraft. A student can therefore start to fly ILS approaches by reference to
the projected ILS cone and appreciate the technique to intercept both the
localiser and glide-slope. It also illustrates the relationship between distance
and sensitivity of the ILS display.
As the student progresses, the cone can be removed and the cloud base
set to a realistic minima. As cross-winds are added the cone can be
re-introduced to illustrate the effects of cross-winds during an ILS approach.
This form of part-task training also extends to selective removal of
augmented visual feedback. If the student is within one degree of the
localiser and within five degrees of runway QDM, say, the cone could be

183
Aviation Instruction and Training

withdrawn. These criteria can be adjusted according to the training program

and the rate of student progress.
This example of part-task training illustrates the potential for very simple
augmentation to a visual display to reduce the conceptual threshold for
particular skills. Of course, for some students there is no conceptual
difficulty in relating the theory of ILS taught in ground-school to the practice
of flying an ILS. This form of part-task training is not proposed as a
panacea or a mandatory component of training, it is simply that part-task
training can be used in a wide variety of ways which include remedial
training for students having problems with a specific training task, to
supplement ground-school lectures with practical assignments and to enable
able students to grasp (and demonstrate) new skills more quickly and
therefore make better use of airborne training time.

Twin engine training

The diagrams outlined in the sections on 'Turns' and 'Lift and drag forces',
above, can be extended to training in twin engine aircraft by displaying the
aircraft in a 'plan-view' form with the engine thrusts and resultant aircraft
moments displayed in a real-time diagrammatic form. As one engine is
failed, the difference in thrust vectors is visible and the resultant moment (or
couple) is presented in the form of an arrow on the nose of the aircraft with
length proportional to the magnitude of the moment.
If the aircraft is orientated in wind axes so that the angle of sideslip is
displayed, the relationship between turning and sideslip is directly obvious
to a student and the effect of compensatory rudder input can be observed as
the sideslip is reduced. In addition, effects of engine feathering and limita­
tions of Vmc are apparent. To summarize, students build on their
knowledge obtained from ground-school by means of a set of directed
exercises and will make use of this practical knowledge during airborne
exercises. Equally, students having trouble grasping these ideas in airborne
exercises can follow remedial exercises in a part-task trainer under the
supervision of an instructor.

HSI/RMI orientation

HSI and RMI displays are widely used in EFR-equipped aircraft. To a

student with low time on instrument flying, they can be bewildering. To the
pilot used to flying with HSIs and RMIs they are instinctive and provide a
compact and thoroughly comprehensible method of navigation. The

184
 The Technology o f Part-Task Training

difficulty for the instructor is taking the student from this initial phase of
trepidation to the latter phase of competence and confidence.
A part-task trainer offers a solution to this recurrent problem in flight
training. It should be pointed out that there is not necessarily a need for a
real-time HSI/RMI trainer and arguably no need for conventional aircraft
controls in this trainer. The student needs to concentrate on learning the
instruments rather than flying the aircraft at this stage of training and a
simple autopilot function is adequate to relieve the student's workload during
this training. In the Aerosoft-200, this task is supported by two additional
training diagrams which can operate in real-time, under either manual or
autopilot control.
The display for the RMI is configured to show the NDB (ADF) station
together with a radial selected by the instructor. The RMI compass is
displayed and indicates the aircraft position relative to the beacon and of
course, the aircraft heading. A green line is drawn from the centre of the
compass to the station to represent the green needle on the RMI. The
aircraft is drawn in the form of an extended white lubber line on the compass
to indicate aircraft heading. Finally, the display can include the
instantaneous desired intercept to the radial and the wind vector can be
added to the heading vector to indicate the resultant track vector. Although
this arrangement sounds complicated, it gives a chart-like appreciation of the
situation, but also, after some 'exploration' by the student, mainly contains
information that is available on the RMI instrument.
Exercises can be run in an hands-off mode, allowing the student to select
headings and watch the situation develop at the same time as watching the
RMI needle and compass movement. Progressively, information is removed
from the display until it only contains information which is also available (in
the same format) on the RMI display. A similar exercise has been
developed for the HSI, typically to train VOR tracking and intercepts. The
VOR station is displayed together with a radial selected by the instructor.
This display also incorporates the plus and minus ten degree CDI limits from
the selected radial. A frill HSI instrument is displayed to indicate the aircraft
position and heading. This HSI display operates in parallel with the actual
HSI display on the instrument panel. Intercept tracks and wind vectors can
also be added to this display. The objective for this display is the same as
the RMI display. At first, the student can readily comprehend the map-like
navigation situation. As some of this information is removed, the student
starts to appreciate that all the essential information is available in a very
similar format on the actual HSI. This realization comes about as a
consequence of a set of directed exercises, paced at the student's rate of

185
Aviation Instruction and Training

progress, and introducing variables such as winds, ATC instructions,

procedures and manual control of the aircraft at stages of the training when
the student can accommodate the increase in workload.

Summary

There have been numerous advances in computer technology which are

appropriate to part-task training and there is no apparent limit to these
developments. It is certain that processor performance of desk-top
computer systems and the computer graphics demanded for CAD
applications will continue to extend the scope of part-task training devices.

External use of part-task trainers requires the matching

of training needs with available technology
(Photo: Swissair)

As the fidelity and capability of part-task trainers increases and their role
in flight training becomes more accepted, there will undoubtedly be a change
of emphasis in flight training away from the single flight simulator towards a

186
 The Technology o f Part-Task Training

range of training devices optimized for specific training programs. This does
not imply the demise of the flight simulator which will continue to have a
major role in aviation training. It does, however, infer that low-cost real­
time part-task trainers will be used in an increasing proportion of training
applications.
In order to optimize the capability of part-task trainers, it will be essential
to analyse training requirements and to match these requirements to the
available trainer technology. This will entail evaluation of the training
effectiveness of specific trainers for particular training roles and also take
into account the cost effectiveness of these forms of training.
As part-task training devices proliferate, it will be essential to regulate
their operation, partly to avoid negative transfer of training caused by
incorrect selection of equipment or inappropriate use of training equipment
for a specific task. This problem is further compounded, because training
effectiveness is not simply a function of the training device, the regulatory
authorities will need to take into account the training course, the training
syllabus and the manner in which the part-task trainer is integrated into a
training program.
At present, the application of part-task trainers is constrained by the lack
of fidelity resulting from limited computer power and the lack of software to
support flight training programs. As these problems are overcome, part-task
trainers will encroach increasingly into flight training programs and it will be
necessary to set down guidelines for their effective operation, based on
analysis of training studies involving part-task trainers.
No doubt, part-task trainers will be used incorrectly and manufacturers
may make unjustified claims for their training effectiveness. But the
advances in computer technology open up a very wide range of new
applications in flight training and the fundamental issue for training
organizations will be to match the technology to the training requirement, by
monitoring these technical advances and by assessing the effectiveness of
part-task trainers on flight training programs.

187
12 The evaluation of
training
K.C. Given

'Competency Based Training' (CBT) is now an integral part of the workplace

trainer's vocabulary. But this has not always been the case; CBT is a
relatively recent phenomenon in the training world which previously was
dominated by formal course-work and qualifications. However, for very
many years CBT has been the rationale behind the Systems Approach to
Training (SAT), favoured, in slightly varying forms, by air forces throughout
the western world. The emphasis on competency has been most obvious in
flying and flying operations training such as ab initio pilot training and air
traffic control training, and is readily illustrated by the widespread use of
increasingly realistic simulators in the development of operational
competencies. Less obviously, but of equal importance, the use of CBT has
also been the rationale behind the training of air force support personnel
including the technical, supply and clerical job groups.
The SAT reflects a very simple philosophy namely, that for any training to
be completely efficient and/or effective it must satisfy the needs of 'on the
job’ performance. That is, the competency being trained for should have as
immediate as possible an application to the workplace. The ideal situation is
to train a for a competency today and apply that competency in the workplace
tomorrow. As the ideal is rarely, if ever attained, a process is needed to
ensure that the training provided matches job performance needs as closely as
possible. This process is the evaluation (validation) of training. Therefore,
the trainer who is committed to the development of competencies, and to a
systematic approach to training, cannot avoid the issue of evaluation that this
chapter will seek to address.
 The Evaluation o f Training

The process of evaluation is not unique to the aviation industry, it has

global application in all training environments so, rather than adopting a
specific aviation focus, the thrust of this chapter is general in nature. There
is much more to the process than the relatively narrow task of matching the
training provided to the job requirements, but this is a useful starting point.
In the broader sense, evaluation implies an assessment of the total value of a
training course both to the individual being trained and to the organization
sponsoring the training. Many other influences are at play and a number of
them will be discussed in detail later in the chapter.
The bracketing of evaluation with validation will undoubtedly raise some
eyebrows among training theorists (and perhaps some practitioners also), but
the dichotomy between the two is often blurred. Evaluation, in general,
appears to refer to the efficiency of training (does it provide value for
money?) and validation appears to refer to the effectiveness of training (does
it produce the competencies desired?). As this chapter will seek to address
both the efficiency and the effectiveness of aviation support training, the term
evaluation will be used in a generic sense to cover both aspects. Space will
not however, allow an exhaustive treatment of the subject therefore, a few of
the more important issues will be highlighted. First, the place of evaluation
in the training cycle will be detailed, followed by an examination of the
current theory and practice relating to evaluation. Finally some suggestions
on possible future developments will be made.

The place of evaluation in training

The evaluation of training is not something that is done for its own sake, it is
done in a particular context, and that context is the training system or cycle.
In fact, training itself also needs a context and that context is the world of
work, or in terms of this book, the aviation industry. For any training to be
successful it must be conducted in a supportive and committed environment
which is consistent and systematic in its approach. It follows then that in an
industry dedicated to professionalism, as the aviation industry most certainly
is, that the professionalism must extend to the training of all its personnel, not
just the 'operational elite'. Aviation support training needs to be provided not
in an ad hoc manner, but in a rigorous, planned, and regulated manner
underpinned by a rational theory of training. It was to meet such demanding
standards that the SAT emerged.

189
Aviation Instruction and Training

The training cycle

The simplest model of a systematic approach to training is shown at Figure

12.1. Within each of the four main elements there are a number of
sub- elements. For example, among other things, 'Analysis' involves an
examination of the job to be performed, a review of the training currently
provided, and an appreciation of the knowledge, skills, and attitudes which
the training will seek to inculcate.

Figure 12.1
The training cycle

'Design' of a training course will include a method and media analysis to

determine the best means of developing the required competencies, deciding
on the most logical sequencing of the training objectives, developing relevant
test specifications, and accessing appropriate training resources. 'Conduct' of
a training course includes audio visual production, the preparation of
instructor lesson plans, the crucial instructor/trainee interface, and the
acquisition by students of competencies or performance skills. Finally,
'Evaluation' includes assessing the level of competency attained by the
trainees, determining the appropriateness of the training provided, and
identifying any deficiencies in the training.
In this simplified model the most important concept, the one which binds
all the elements together, is that of the closed loop. Put briefly, the elements

190
 The Evaluation o f Training

of the model are interdependent and the model will fail to produce the desired
training outcomes if any one element of the model is omitted. Again, this can
be readily illustrated. Assume for a moment that, in a situation where an
urgent training 'fix' is demanded, the 'Analysis' element of the model is
omitted . Often the end result of such an action is that training solutions are
developed to solve what are, in effect, management problems. Only by
undertaking a thorough analysis of a perceived problem or need can
appropriate training, if that is the correct solution, result. For example, a
training 'fix' is often sought when staff turnover is high, the tendency is to
train our way out of the problem. However, a thorough analysis of the
reasons for the turnover may suggest that a management fix', which
encourages the retention of current staff, is more appropriate and effective
that a possible training solution.
The omission of the 'Design' element is equally fraught with problems.
Improper sequencing of instruction or inappropriate testing of performance
will result in disjointed or irrelevant training. A training program that is
inadequately designed has the potential to cause more problems than it solves.
Of course, the model would not exist at all, except as a theoretical concept, if
the 'Conduct' element was omitted. Therefore, attention must be paid to the
sub-elements of conduct; for example, if visual aids are inappropriate or
instructor lesson plans inadequate then little is actually being achieved even
though the training is still going on. It is the 'Evaluation' element of the
model which seeks to ensure that the shortcomings briefly highlighted above
do not occur and more will be said about this in the remainder of this chapter.
To some extent I have laboured the point about the interdependence of the
elements of the system for, while it may appear to be obvious, and hopefully
logical, all too often the obvious is being ignored for a variety of reasons.
These include the cost in financial and manpower resources of the full
implementation of the SAT, the time taken to develop an ongoing training
package when all elements of the system are properly implemented, and the
belief that such an approach is too inflexible in the dynamic world in which
we live.

The place o f evaluation in the cycle

Despite these reservations, generally speaking, in at least a limited way, the

analysis and design elements of the training system are usually undertaken by
most organizations wishing to conduct their own training. It was in support
of the 'Analysis' element of training that the US Air Force developed the
exhaustive Comprehensive Occupational Data Analysis Programs (CODAP)

191
Aviation Instruction and Training

job analysis method which is now widely used in military and civilian
organizations. CODAP analyses jobs in terms of task level performance data
and task factor training criteria. As far as the design element is concerned,
there is considerable emphasis on the development of trade standards and test
specifications in particular, and the more general need to design training in
such a way as to extract maximum return for the dollars invested.
However, the commitment to the element of evaluation has been much less
obvious despite its equally important place in the training system. As
indicated previously evaluation needs to be seen as the closing of the training
loop and if the loop is not closed then there can be no confidence in the
effectiveness or the efficiency of training. This is unacceptable to any
organization but particularly in an industry such as the aviation industry where
both operational and training costs are high and safety is a significant
consideration.
Evaluation, like analysis and design, involves the collection of information.
In the case of evaluation it is information about the training provided and the
ability of the trainee to perform in the workplace. The information source
may therefore, be from the training organization itself (internal) or external to
it (from the workplace). Whatever the source of the information, the purpose
behind the gathering of the data is to respond to a number of leading
questions such as:

- Is the trainee being provided with the competencies required to

satisfactorily perform in the workplace?
- Are there any shortcomings in the training provided?
- Are the most appropriate training methods being used?
- Are all aspects of the job requirement being satisfactorily addressed by
the training?
- Is any over-training, under-training, or duplication of training taking
place?

In essence, evaluation is the litmus test of the effectiveness of the training

being conducted in that it provides an assessment of the match between
training and job performance. As illustrated in Figure 12.2, the ideal situation
is achieved when there is a perfect match between the two.
There are a number of well tried procedures for gathering the necessary
information. As far as internal information is concerned, formative and
summative assessment (testing), course critiques, instructor input through
course reviews, and training management reports, all serve to improve the

192
 The Evaluation o f Training

training process. Externally, post course evaluation surveys of both course

graduates and their supervisors, training organization visits to the workplace
to conduct interviews with graduates and their supervisors, and on-the-job
proficiency testing, all provide information relevant to the evaluation of
training.

Figure 12.2
Training versus job performance

Summary

Evaluation then, is the cement that holds the elements of the training cycle
together and unless all elements of the cycle are addressed we cannot claim
that our training is systematic. The demoralising, and grossly inefficient and
ineffective, impact of mismatched training, over-training or under-training can
only be avoided if evaluation receives the same attention as the other
elements of the cycle. Bearing this in mind, the next logical step is to
examine the theory and practice of evaluating training.

The theory and practice of evaluation

The theory

Evaluation suffers from the all too familiar cry, That's all right in theory, but it
won't work in practice.' To a large extent, the theory behind evaluation and

193
Aviation Instruction and Training

its crucial place in the training system has already been detailed above. There
are however, a number of recent additions to training theory which need to be
addressed because they will impact on training quality control practices
(evaluation) in the immediate future.
As mentioned previously, evaluation can be thought of as internal or
external and for the sake of clarity it is worth noting the relationship between
the two as illustrated in Figure 12.3. The link between them is the
determined competencies, that is the workplace competencies identified
during the analysis phase which become part of the training objectives.
Internal evaluation assesses if the training course output matches these
training objectives, and external evaluation assesses if the competencies
provided match the job requirement. This hopefully, sounds like a circular
argument as that is exactly what it is supposed to be, with the competencies
required on the job being the focus of the debate.

Figure 12.3
Internal and external evaluation

However, evaluation, particularly internal evaluation, is not just about the

competencies developed through training vital as they are. Evaluation
assesses the whole gamut of the training process including instructor
effectiveness, the quality and appropriateness of training aids, the physical
environment and study facilities, and the administrative support given to the
trainees. In fact, theoretically, evaluation is an assessment of the other
elements (analysis, design and conduct) of the training system.
The development and use of valid and reliable evaluation processes are
also important aspects of evaluation. These processes can range from a full
blown occupational analysis at the one extreme (using a very structured
approach such as CODAP), to informal feedback on an opportunity basis at

194
 The Evaluation of Training

the other. Despite the best efforts of the evaluator to gather valid and
reliable data, this is often the most daunting of tasks as the training provided
tends to change so often that no two training courses are exactly the same.
Therefore, asking course graduates for their perceptions of training courses
such as: 'What percentage of the training you received is applicable to your
present job?' is affected by a number of variables which it is hard to control,
not the least of which is changing course content. If course content has
changed then it is not possible to compare two different courses unless we
want to compare apples with oranges. The end result of such comparisons is,
unfortunately, of little real value in assessing the applicability of training to
on-the-job performance.

The competency approach to training emphasizes early exposure

to the practical requirements of the job
(Photo: Royal Australian Air Force)

In the age of CBT, another theoretical issue is becoming increasingly

important. This issue is that of assessing competencies and is closely
connected with other issues such as the Recognition of Prior Learning (RPL)
which is also referred to as the Recognition of Acquired Learning or the
Recognition of Current Competencies. In short, the current emphasis in
training theory is on output (what a person can do at the conclusion of a

195
Aviation Instruction and Training

training program, or as the result of work experiences) as opposed to input

(what a person has been exposed to in formal training programs). There is a
recognition that people develop a range of competencies, often without any
formal training, and this results in a multi-skilled workforce which is able to
perform a broad range of activities. By way of illustration, in the air force's
military culture, technical airmen develop management skills in addition to
their technical skills as they progress through the ranks. The question of
assessing such competencies remains.
The competency approach can also be associated with a top down (just in
time) approach to training. That is, trainees are exposed to the practical side
of their job requirements (and to the job itself) as early as possible in the
training rather than being exposed initially to a mass of peripheral knowledge
before gaining any practical experience. This approach also involves smaller
bites of training being provided at more regular intervals during an
individual's career rather than treating a student like a camel and filling him up
at the start of a twenty year career with all the technical information and skills
he needs to complete that career. The up front approach ignores the reality
of the attrition of acquired skills and knowledge when they are not put into
immediate practice; the guiding principle is 'use it or lose it'.
A final factor currently impacting on training (and all other areas) in the
Royal Australian Air Force, where it is referred to as RAAF Quality, is that of
Total Quality Management (TQM). TQM, which is growing in influence in
the wider Australian industrial community, stresses continuous process
improvement by a systematic and data based approach so that the customer's
needs are met. The evaluation of training is not only consistent with this
management philosophy, it could be argued that it, in fact, pre-dated this
highly logical approach. By closing the training loop, evaluation makes the
training process a continuous one, and by constantly seeking to make the
training more efficient, evaluation improves the training. In such
circumstances both the trainees and the organization sponsoring the training
(each of whom can be regarded as customers) have their needs met.
Evaluation is therefore, completely consistent with, and an example of, TQM
in action. To omit evaluation from the training loop is to suggest that all is
well with the training and that there is no room for the continuous
improvement in the process which is fundamental to a TQM philosophy.

The practice

Turning to the practice of evaluation, we are faced with the immediate

problem of personalities. Theory would suggest that the evaluation process

196
 The Evaluation o f Training

would best be conducted by a neutral, someone with no vested interest in the

other stages of the training system. In reality however, that is seldom, if
ever, the case as the most likely scenario is that the people who have analysed
the training need, designed the appropriate course, and conducted the
necessary training, are also charged with evaluating the training. That is,
people are expected to critically examine their own work and human nature
being what it is such criticism is seldom forthcoming. Perhaps TQM offers a
possible solution to this problem stressing as it does the concept of the
customer and his needs, not the process itself.
A second problem is that people only have so much time at their disposal
and in the scheme of things, evaluation tends to come up in the rear. If the
elements of the training system had to be placed in order as they impact on
the trainer, then course conduct will take pride of place followed by course
design, analysis of the problem, and finally evaluation if any time is left for
such activity.
Evaluation is also a costly pursuit. From the point of view of the training
establishment, the choice may be between conducting training and
undertaking evaluation as the same people are likely to be involved in both.
Even very large organizations would baulk at the costs involved in employing
dedicated evaluation personnel. In practice, the training organization will do
what it has to (that is conduct training) before it will do what is perceived to
be nice to do (that is evaluate training). Additionally, from the workplace
point of view evaluation may involve taking people away from their
productive work to complete a survey or to take part in an interview. The
workplace supervisor will naturally give priority to job performance before
paying attention to the demands of evaluation. There will be a natural 'what's
in it for me' attitude in the workplace which will impact on the level of
support given to the evaluation process. Without the support of the
workplace the task of evaluation becomes extremely difficult.
Another demanding aspect of the process of evaluation is the need to
document the findings and then to do something constructive with the
information gathered. There is a tendency for the information collected as
part of the evaluative process to disappear into the black hole of the training
organization with no obvious changes to, or improvements in, the training
process or the graduates of the process. (In part this explains why the
workplace is reticent to support evaluation surveys in the workplace because
there is little feedback or reward). In addition, there is a perception that the
training process is often guilty of re-inventing the wheel; recycling old
solutions. It is for this reason that there is a need for the evaluation process to
be amply documented. Again however, documentation requires resources,

197
Aviation Instruction and Training

and if the initial indications of an evaluation study turn up no significant

recommendations in terms of changes or improvements, it is likely to remain
un-recorded.
There is the equally daunting question as to what course of action is
required when the data collected suggests that a radical change to the training
being provided is needed. Every trainer knows that there is inherent inertia in
the training system and, despite the evidence of documented information and
a thorough going evaluation of the training, the very thought of the effort and
costs involved in changing a training program tend to encourage this inertia.
It is generally much more comfortable to nibble around the edges of the
problem than to perform the radical surgery that is really needed.
The problem of reliable and valid evaluation was mentioned earlier and
there is a tendency to hide behind these problems and take the easy way out
by not doing evaluation at all, or by paying Up service to the process. This
half hearted approach is often seen in practice, where the trainer, convinced
of the importance of closing the training loop by evaluating the training,
nevertheless accepts a standard of evaluation much lower than the standard of
analysis, design or conduct which would be acceptable.

Assessment of competency is an im portant aspect of instructional design

(Photo: Royal AustraUan Air Force)

198
 The Evaluation o f Training

With the growth of CBT the training, role, and responsibilities of the
competency assessor need to be clearly defined, as do the assessment
standards. One of the issues which needs to be resolved before CBT wins
the commitment which it deserves is, in fact, the control of competency
standards. In Australia, the National Training Board is encouraging the
development of competency standards for the major industries (such as the
hospitality industry) against which trainee performance can be assessed. The
measurement of trainee performance in defined situations is what competency
assessment is all about. Trained and registered assessors will become the
custodians of the competency standards which all trainees must satisfy in
order to gain recognition of their abilities and eventually employment in the
relevant industry.
In practice however, the assessment of competencies is not yet well
developed and there is a continued reliance on the tried and true approach to
training namely, classroom and theory instruction supported by practical
lessons. This is the comfortable, economic, and safe approach for the
training organization and the training staff. But there are adventurous people
who are developing concepts such as the top down approach to training
which is forcing trainers out of the comfort zone. Facilitators are replacing
instructors, experiential learning is replacing rote learning, self paced learning
is replacing common paced learning, and the investigation of techniques
which may enhance training effectiveness such as Neurolinguistic
Programming (NLP) and accelerated learning is continuing.

Evaluation into the next century

Training in general faces an exciting decade as the use of technological

advances opens up avenues for trainers which were inconceivable even a few
years ago, and as exciting developments in training processes become more
broadly acceptable. Inevitably developments in the evaluation element of the
training cycle will also proceed apace. Firstly, the emphasis on CBT will
require a corresponding emphasis on the place of assessors and assessment in
the process of evaluation. This will be driven by economic and personnel
related reasons. The economic push will require training to be as efficient as
possible, and the role of evaluation in that area is obvious. Evaluation can
also reduce the duplication that currently exists in the training provided, it can
ensure that training is closely aligned to job requirements, and it can
therefore, improve the efficiency of the whole training cycle.

199
Aviation Instruction and Training

The personnel push will be driven by the desire to obtain broad based
recognition of the training being provided by private organizations (and this
includes the military training organization) for their course graduates. This is
already of some importance within the Australian Defence Forces and is
liable to gain popularity with other training organizations also. Not only is this
a sensible move for the military to take, it is a logical move for government to
support as the military training organization is supported by government
expenditure and, through the production of highly skilled personnel,
especially in technical fields, can make a substantial contribution to the
overall skill levels of the nation.

Evaluation completes the training loop

(Photo: Royal Australian Air Force)

Training will be modularized and given in smaller portions, and just in time
training will become as common as just in time inventory keeping. The
training modules will be competency based and will require separate
evaluation. A whole raft of opportunities will become available, both in the
development of the modules and in the evaluation of the reliability and
validity of each module. The main question will be whether training

200
 The Evaluation o f Training

organizations will apply their limited resources equitably between all

elements of the training cycle or will the development of modules take
priority over the development of evaluation tools.
The recognition of prior learning will result in further training efficiencies
as individuals acquire skills through life experiences rather than through
formal training programs. Assessors will evaluate the acquired competencies
of people and with greater recognition of acquired competencies, less formal
training should be required. This will see people gaining more recognition for
being multi-skilled and should result in a more broadly skilled workforce
where fewer people can do the tasks currently requiring a larger number of
people. The balancing act here will be in avoiding the 'Jack of all trades,
master of none' syndrome. The role of evaluation in this whole area is likely
to increase dramatically.
The major challenge to be addressed concerning evaluation is the
development of an economical, reliable and valid approach to the whole
process so that it becomes attractive to undertake the exercise. In this
regard, the development of PC versions of CODAP (see above) may well
become useful not just in job analysis but also in the evaluation process.
Suffice to say that there is a strong feeling in the training fraternity that more
attention needs to be paid to the final element of the training loop; the loop
must be seen to be closed, evaluation must be seen to occur.

Conclusion

The arrival of the era of CBT may well see the emergence of the element of
evaluation as more central to the training cycle than it has been in the past. It
is certainly unlikely that it will be able to be ignored to the extent that it has
been previously. The reason for this is quite simply that it is the evaluation
of the training process which is the quality control on the effectiveness and
the efficiency of the training. Evaluation is therefore, consistent with the
demands of, and the processes associated with, TQM. For that reason alone
its influence is assured.
A growing emphasis on this aspect of the training cycle should be
welcomed by the aviation industry. By improving the efficiency of the
training provided, evaluation should contribute to the bottom line in company
accounts. By improving the effectiveness of the training, evaluation should
contribute to the maintenance of the high standards which already exist, and
which must be maintained, in an industry so reliant on the safety of its
operations for its success.

201
Aviation Instruction and Training

Evaluation is one part of the training cycle, and training is but one aspect of
the multi-faceted activities of the aviation industry. The industry's focus will
always, and quite properly, be on operations. Training cannot dominate the
industry and evaluation cannot dominate training. However, evaluation is a
technique which can improve training and therefore, by inference, it should
improve operations. In the search for a competitive, or safety edge, the use
of a systematic evaluation process, which assesses the efficiency and
effectiveness of the acquisition of proficiencies, appears to be a reasonable
investment for the future.

202
Section 3

A b-In itio Pilot Training

and Instruction

(Photo: Qantas)
13 Introduction to section
3: ab initio pilot
training and
instruction
R.A. Telfer

The eight chapters in this section constitute a strong argument for the
professionalization of flight instruction. They also argue for innovation
provided there is a consolidation of aspects of ab initio pilot training which
are serving the industry well.
In the first two chapters it is argued that aviation instruction is a specialist
form, different from other types of instruction. Apart from obvious
differences between, say, school teaching and flight instruction, it is possible
to make other distinctions such as the nature of the stake at risk. There is
also the need for efficiency; the training element; the roles of theory and
practice; the aviation culture; the guidance of androgogy rather than
pedagogy.
If flight instruction is specialist in nature, how valid is the application of
the findings of educational research? The second chapter of the question
takes up this point, pointing out that valid application depends upon the
nature of the research. It goes on to identify relevant research which can
guide aviation instructors towards more effective learning by their students.
These results can be applied in three stages: presage (pre-instruction),
process (during instruction) and product (post-instruction). Research
findings can also be applied to instructor preparation (as in the example of
Cathay's instructor workshop discussed in the Introduction to Section 2).
The teaching of skills and the value of practice are important instructional
elements at the ab initio stage. Theoretical bases for instructional
approaches are presented, with details of the conditions need for the learning
of a skill. Chief Ground Instructor at the British Aerospace Flying College
Aviation Instruction and Training

in Scotland, Brian Blyth, suggests that theoretical training has been given an
unjustified subordinate role in pilot instruction. He will undoubtedly find
support for his argument that minimum standards required by examination
are not what instructors should use as criteria. The British Aerospace
Course and the Lufthansa 'Futura' scheme are provided as examples of
programs designed to train professional pilots. The aims of the courses
include skills, knowledge and an integration of the two.
Design and phasing of the practical and academic components are critical
to ab initio training, Blyth argues. He presents a ground training curriculum,
then the syllabus and lesson plan which evolve. In his discussion of the use
of CBT, he reinforces the importance of the instructor as part of educational
environment.
Prior to joining the Army Personnel Research Establishment at
Famborough in 1992, Chris Elshaw had an active interest in RAF instructor
preparation. In the fourth chapter of this section he discusses the potential
unreliability of the training process, and ways of preparing instructors to
overcome this problem. What ten reasons would you give for students
failing basic flying training? You can compare yours with those Elshaw's
research revealed, and see how skill acquisition, motivation, stress and
anxiety link to an explanation of these reasons. He concludes that there are
persuasive reasons for including such aspects of the psychology of learning
in pilot instructor training.
The total learning experience is the focus of Professor Hank Lehrer of
Embry Riddle Aeronautical University, Florida, where he was the founding
editor of the Journal of Aviation/Aerospace Education and Research.
Following foci on instructional effectiveness, the academic component, and
instructor preparation, it is timely in the fifth chapter to examine
instructional design. Lehrer provides a checklist of key questions which
lead to the Aviation Curriculum Design Matrix which is later applied to two
different settings in aviation. The ACDM provides a useful aid for those
developing curricula in aviation.
Dr Len Lambeth combined his experience in the RAAF, his flying, and
his medical and psychiatric background to devise the aviation medicine
course conducted in the BSc (Aviation) at the University of Newcastle. In
his chapter he traces the history of aviation medicine, distinguishing civil
aviation from the military flight surgeon's role. By means of short case
studies, a range of medical problems in aviation are presented: hypoxia,
hyperventilation, decompression sickness, acceleration, noise and vibration,
fatigue, alcohol, smoking, and medication which affect pilot performance.
Lambeth's contribution parallels a more general interest in human factors
in medicine which has been traced over the past twenty-five years. It has
been argued that because medical technology may be outpacing our

206
 Introduction to Ab Initio Pilot Training and Instruction

knowledge of its effects and usability, there continues to be a major need for
human factors involvement in health care and medical applications (Sind,
1992). This view supports Lambeth's conclusion of the value of regular
involvement of a medical practitioner in a professional pilot's career.
'Professional' in this sense does not necessarily mean a commercial or
airline transport pilot. It implies one who takes a pride in airmanship, and
bases aviation practice on a set of high personal values. In the next chapter,
Professor Bill Pippin, Chairman of the Department of Aviation at the Ohio
State University, calls upon his forty-five years of aviation experience to
compile an education program for an aviation professional. Such a person is
seen by Pippin as one with both competence and conscience.
Pippin identifies the problems in changing traditional pilot preparation
which may be recreating the past, or tend to be incestuous in its promotion
of managers. He justifies a university preparation for pilots, arguing that
'total system considerations must be added to the study of component parts if
aviation practitioners are to be truly educated for the future'. This future
includes the evolution of a global aviation system, for which an evolutionary
change in aviation education is required.
A differentiation of pilot learning into skill-based (relying on stored
routines learned through practice), rule-based (for which a procedure has
been learned), and knowledge-based (for which no procedure has been
established) helps to distinguish training and education (Green et al, 1991).
The change Pippin describes involves education - knowing why, as well as
training - knowing how.
Such a development is described by Dr Andre Droog, psychologist at the
KLM Flight Academy. The academy's facilities and equipment are
described in the final chapter of this section, prior to a summary of the KLM
ab initio training program. It is the attention given to non-technical training
which makes this program different. The non-technical training emphasizes
team leadership, decision-making, communication, aviation psychology and
physiology, captaincy and cross-cultural dynamics.
Based on a substantiated rationale that crew performance is a variable
'over and above individual performance' (O'Hare and Roscoe, 1990), these
modules in non-technical training are introduced systematically according to
the nature of the trainee's flight training at the time.
As part of the individual's self-development an Outward Bound course is
included early in the training program, while sports and fitness are continued
throughout the two year course. Droog foresees that human factors training
and the development of non-technical skills will become common in future
flight training: a benefit for both pilots and the airline system.
It can be concluded that the contributors to this section foresee an ab
initio training which extends beyond the minimum needed to obtain a pilot

207
Aviation Instruction and Training

licence. The nature of the elective component differs according to the intent
of the program. For example, KLM seeks team leaders while Ohio State
seeks an educated professional. Regardless of the specific goal, programs in
ab initio pilot preparation will include greater attention to human factors,
and instructor preparation will have to reflect that change.

References

Green, R.C., Muir, H., James, M., Gradwell, D. and Green, R.L. (1991),
Human Factors fo r Pilots, Ashgate, Aldershot.

O'Hare, D. and Roscoe,S. (1990), Flightdeck Performance: The Human

Factor, Iowa State University Press, Ames.

Sind, P.M. (1992), 'Human factors in medicine: quo vadisT Gateway, Crew
System Ergonomics Information Analysis Center, University of Dayton,
Dayton.

208
14 What makes aviation
instruction different?
R.A. Telfer

Where is the theory of aviation instruction?

It should be possible to provide guidelines drawn from tested research,

enabling any flight instructor to plan, present and evaluate instruction in an
effective and efficient manner. It would be effective in that it achieved its
goals, and efficient by doing so in the minumum time and at the minimum
cost.
It is logical that the aviation industry would expect research to provide the
same guidance in instruction as it has in ergonomics, engineering and
navigation. After all, millions of dollars of taxpayers' money have been
expended internationally to find clear indices of effective teaching and
learning. There remains, however, rather pessimistic views of what
researchers and practitioners have to offer each other:

Researchers and practitioners might follow the lead of games theorists,

whose rule of rationality is codified in the so-called 'mini-max' principle:
minimize the maximum loss you can suffer. If both communities do this,
and also expect the loss, then anything else that happens must be a pleasant
surprise. By expecting little from each other, occasionally both the
researcher and the practitioner may be in for a little treat. (Phillips, 1980,
P· 20)

That is an extremist viewpoint. Fortunately a balancing perspective is

available:

Education is not a dismal science, though many act as if it were. In

Aviation Instruction and Training

educational research, of course, there is still much to find out, but

education is not essentially mysterious. (US Department of Education,
1987, p. vii)

This chapter leans towards the latter, optimistic view. There is a body of
theory available in psychological and educational research. Some of this
theory has validity when applied to aviation (or, say, driver education or
sports training) and some does not. Think of it as a series of overlapping
circles, each representing a specialist training area and each in turn
overlapping with a portion of larger circle representing the body of theory
available (see Figure 14.1). The basic problem is that there is no clear
indication of the limits of application of the larger body of theory.

Figure 14.1
Aviation instruction theory

A contributing factor has been a confusion of description (what is) and

prescription (what should be) by both producers and users of theory. The
limits of descriptive research are the lack of specific recommendations or
guidelines for practical implementation.
In psychology, pure research has been given more recognition than
applied, consequently it is preponderate. The knowledge produced by pure
research is certainly generalizable: the method, sampling and statistical

210
 What Makes Aviation Instruction Different?

analysis are all designed to ensure this. Problems of flight instructors,

however, are in the immediate and in the particular. They are clouded by
priorities, needs, values and goals.
In the same way that a psychiatrist describes each patient as a 'universe of
one', or a physician says that '85 per cent of the problems he sees are not in
the book', instructional practice in aviation is characterized by unique events
(Schon, 1986). This is a challenging situation in which to call on theory for
guidance or solution, and is compounded by the nature of what is termed
'research'.
Research can range from a summary of what others say is true, to critical
reviews of work which has been subjected to stringent scrutiny by experts,
and to scientific inquiry under rigid guidelines and supervision. There are
grades of quality within research, and the source one chooses as a reference
determines the nature of the solution which will be offered.
Description certainly provides data, but there remains the task of
determining if and how the data relate. Most importantly, there is the
question of whether one variable causes another. In other words, there has
been a tendency to attribute to instructional research qualities of prediction,
testability and guidance it may not possess.
Theory represents reality: and some work which has been seen as a
theoretical contribution has simply been a description of observations which
lack application to other settings. Grade 3 History with thirty students in
Minneapolis is quite different from six British Airways trainees at Prestwick
studying principles of flight. There is a need for an eclectic selection of
instructional theory applicable to aviation.
After reviewing one hundred and seventy five publications on teaching
effectiveness, a recent reviewer concluded, 'What empirical evidence is
available does not necessarily generalize across subject matter, grade level,
students or classroom culture' ( Reynolds, 1992).
Table 14.1 shows some important differences between instruction in
educational institutions and that in commercial aviation. Compared to the
structure in which the classroom teacher operates, flight instructors are more
in a vertical than a horizontal flow of information. Their mission is more
immediate, requiring prompt accommodation of any changes in employment
requirements. Their focus is squarely upon the transfer of their instruction to
real-life sitations, while the teacher has the luxury of delayed transfer.
Accountability for the teacher is less immediate.
The flight instructor's budget forces an emphasis on the maximum trainee
competency at the minimum company cost. Training is perceived as being
non-productive for airlines, and a critical cost for flying schools and their

211
Aviation Instruction and Training

clients. The teacher, however, can stress the difference between preparation
for life and living, as opposed to earning a living. There is no need to rush,
and effectiveness is more important than efficiency. The flight instructor
has only limited flexibility, with a rigid syllabus and tight scheduling. The
teacher, however, usually has choices available.

Table 14.1
Commercial aviation instruction and public education

Variable Aviation Education

Structure Centralized, with Decentralized

Instructors in loop Teacher autonomy

Mission Knowledge and skills Transmit culture

Socialization

Focus Transfer of learning Delayed performance

Budget Efficiency Effectiveness

Flexibility Limited Choice available

Teacher Temporary/transient, Career

rarely career Method and content

(Adapted from Burkman, 1987, p. 436)

Finally, the professionalization of teaching has occurred, but flight

instruction is still perceived as a transitory role rather than as a career path.
Flight instructors are usually temporary experts, anxious to get back to the
line. Career teachers have the advantage of specified periods of preparation
in both method as well as subject content.
The flight instructor thus operates in several ways which are quite different
to those of other teachers. It follows that we cannot simply take instructional
research from schools and apply it to aviation without critical scrutiny.
Similarly, findings from laboratory studies on work load, learning skills,
whole task versus part-task instruction, or spaced instruction versus massed

212
 What Makes Aviation Instruction Different?

or block instruction can be unequivocal in terms of results on a keyboard or a

specially designed task, but of little help for a flight instructor seeking an
answer for the next lesson with a waiting student. From a scientific
viewpoint, there is much that we do not know about aviation instruction.
Instructional theory is relatively youthful, as shown by a review in the late
sixties indicating a century of research had produced about two thousand
publications (Snelbecker, 1967). Several false starts, proving fmitless, have
occurred. Many of the early contributors saw themselves in more prestigious
pursuits, such as unravelling the complexities of cognitive processing.
Contributing to instructional theory was a corollary.
Thus evolved general views about instruction and an expanding body of
facts which were not systematically related. Without a stable body of theory
to underpin them, instructional methods peaked and waned. Programmed
instruction and the use of teaching machines expanded in the fifties.
Computer assisted instruction was the new hope of the sixties. For example,
PLATO (Programmed Logic for Automatic Teaching Operation) at the
University of Illinois went through a four-stage development between 1960
and 1972. By that time there were 950 terminals at 140 sites. Eight thousand
hours of instructional material had been produced by some three thousand
authors. Yet there is no 'compelling evidence that PLATO has had either a
positive or negative effect on student achievement' (Saettler, 1990, p. 309).
Similarly, in aviation, what evidence is available to support audio-visual
displays in individual study carrels as the optimal means of providing type-
endorsement instruction for pilots?
In the absence of reliable answers to questions such as these, the industry
seeks expertise from within its own ranks when it comes to the design of new
training programs. Experience does not always bring expertise, but the
aviation industry perceives a greater probability of transfer from a company-
designed program to the performance of trainees on the job. Why is this the
case?

The specialist nature of aviation instruction

Effectiveness vs efficiency

One of the major differences between instruction in most educational

institutions and that which occurs in the aviation industry is the demand for
efficiency. Public education, certainly, places effectiveness (achieving the
instructional goal) above efficiency (the costs of goal achievement).

213
Aviation Instruction and Training

Compulsory education in most industrialized nations occupies at least twelve

years. Instruction in aviation has to occupy the minimum time necessary to
achieve the specified level of competency.

Training vs education

A second difference is the emphasis on training rather than education. Whilst

ab initio aviation instruction encompasses all the domains of learning
(cognitive or knowledge; affective or attitudinal; psychomotor or skills)
there is an implicit curriculum hierarchy in which skills are pre-eminent. How
often have we heard about the importance of being a 'stick-and-rudder' person
first and foremost? There is an essential difference between education for
living in society, and training to fill an occupational or professional role in
that society.

The environment o f teaching and learning

Further, there is a major difference in the learning and instructional

environment. Although some aviation instruction will occur in a highly-
designed context, such as the well-equipped classrooms found in the top
flying schools, universities and airlines, vital periods of instruction will also
occur in a cramped, noisy, mobile setting with unpredictable distractions.

Teachers vs flight instructors

Fourthly, teaching is usually a career choice with professional characteristics

such as a licence to practice; specified qualifications over at least three
years, mostly four; an active professional association; and a code of ethics.
As discussed above, and summarized in Table 14.1, there are major
differences between classroom teaching and flight instruction. Flight
instructors lack the esteem and career opportunities of the line pilot, and the
position tends to be transient. There are clear limits on the amount of time
and money a young instructor will devote to professional development as an
educator, knowing the ultimate goal is flying.

Theory [Link]

Fifthly, until recently instructional effectiveness has been the focus of

researchers, not flight instructors or pilots. This has widened a perceived gap
between the talkers and the doers, and a real gap between research priorities,

214
 What Makes Aviation Instruction Different?

realities of the job, and available theory as a guide to action. There are some
encouraging signs of this gap being bridged by academics' activities in flying
schools and airlines, by more universities becoming involved in providing
academic aviation qualifications, and by the employment of more graduate,
and some post-graduate qualified pilots.

Art vi craft

Sixthly, the culture of flying is distinctive. It may not be unique, but the
nature of the 'right stuff, the kinship of pilots, the achievements, norms,
values, language, attitudes, dress and language all contribute to an unusual
instructional setting. There is undoubtedly an a rt as well as a craft of
aviation instruction (Telfer and Bent, 1992). Some individuals have the
advantage of special experience, abilities, insight and nature which gives
them a considerable personal advantage as an instructor. There is
undoubtedly a 'wisdom of practice' and most instructors develop an ability to
do better at some things than at others (Brandt, 1992, p. 16). This is a lottery
choice across the industry, however, in designating strengths and
weaknesses.
In order for the industry as a whole to advance there needs to be a solid
body of theory, and based on this recognized foundation a program of
preparation which will enable all instructors to reach a specified level of
competency. Ongoing studies of effective aviation instruction will enable the
body of theory to expand so that it encompasses planning, preparation,
presentation, assessment and management. Concurrently, the instructor
preparation program will become more efficient.
Case studies provide valuable source materials for instructor training, and
more research can be undertaken by instructors with their own instruction as
the subject. The best case studies are those drawn from the experience of
colleagues. Most flight training organizations could quickly collect some
local cases to utilize in instructor training courses, and a workshop or
conference could draw them together in a form suitable for shared use by
industry as a whole.

Andragogy vs pedagogy

Finally, there is an important matter of the appropriate source of underlying

theory for aviation instruction. Trainee pilots are mature (or maturing)
learners. Unlike school-teaching, aviation instruction occurs with adults.
Young adults, in some cases, but above the age where pedagogy (teaching

215
Aviation Instruction and Training

the young) rather than androgogy (teaching adults) should provide the
guiding principles.
Given these differences, it is apparent that some research findings from
classrooms and laboratory settings will not (indeed, cannot) apply to aviation
instruction. Not all the results are irrelevant, however, as the following
section of this chapter will demonstrate. If this core accumulation of theory is
ignored, the aviation industry is destined to restrict its own growth and
efficiency to the contributions of individuals within the spans of their personal
careers.
The need is for a body of theory (more likely, for some time, to be in the
form of a collection of theories) with direct application to aviation as a guide
to practice and as a nucleus for the growing contributions from increased
instructional research, and as flight instruction becomes professionalized. For
scientific enquiry we need to identify the concepts we are working with in
flight instruction, then systematically isolate the laws linking the variables.
From these laws, eventually, will come the theory to guide aviation
instruction.
Without such a body of theory there is a danger that aviation instructors
will continue to follow what sociologists term 'folkways': unwritten patterns
of behaviour practitioners consider to be correct (Buchmann, 1987) such as
the 'patter' which traditionally accompanies particular points of airborne
lessons. Instructors can appear effective by demonstrating the 'right stuff, but
may lack vital expertise. Worse, if they have intuitively found a formula for
effectiveness, it will die with them. A more scientific culture can be
gradually refined transmitted from generation to generation of flight
instructors. This a challenging venture. 'Everyone should be cautious and
remember that teaching is a tremendously complex set of processes that
continually departs from fixed recipes and formulas' (Arends, 1988, p. 47).
For example, to what extent do we take into account prior knowledge by
individualizing pilot training? Recruits vary widely in extent and nature of
their flying hours. A recent view is that prior knowledge is vital in
determining a learner's readiness for learning and depth of understanding.
Prior knowledge enhances the trainee's ability to process information and
interconnect concepts (Leinhardt, 1992). How can airline courses be adapted
to take into account variations in prior professional experience?
Further, there is growing evidence of identifiably different approaches to
learning by pilots (see Chapter 9 by Moore and Telfer). There are different
motives to leam, and different strategies which result in varying levels of
depth of knowledge. Given multiple kinds of knowledge, the aviation industry
would undoubtedly seek to gain depth. Yet ’... neither teaching simple

216
 What Makes Aviation Instruction Different?

hierarchies of actions nor simply having students work with hands-on

materials ... will result in deep learning' (Leinhardt, 1992, p. 24).
With appropriate caution, what can we draw from existing educational and
psychological research? Fortunately, we do not have to go back to each of
the separate studies, as some broad reviews have provided summaries and
analyses. In the next chapter, this potential theoretical base for aviation
instruction will be presented and discussed.

References

Arends, R.I. (1988), Learning to Teach, Random House, New York.

Brandt, R. (1992), 'On research on teaching', Educational Leadership, vol.

49, no. 7, April, pp. 14-19.

Buchmann, M. (1987), 'Teaching knowledge: the lights teachers live by',

Issue Paper, vol. 87, no. 1, Michigan State University Institute for
Research on Teaching, East Lansing.

Burkman, E. (1987), 'Factors affecting utilization', in Gagne, R.M. (ed.),

Instructional Technology: Foundations, Lawrence Erlbaum, Hillsdale, pp.
429-455.

Leinhardt, G. (1992), 'What research on learning tells us about teaching',

Educational Leadership, vol. 49, no. 7, April.

Phillips, D.C. (1980), 'What do the researchers and the practitioners have to
offer each other?' Educational Researcher, vol. 9, pp. 17-20.

Reynolds, A. (1992), 'What is competent beginning teaching? A review of

the literature', Review o f Educational Research, vol. 62, no. 1, Spring, pp.
1-35.

Saettler, P. (1990), The Evolution of American Technology, Libraries

Unlimited, Englewood.

Schon , D.A. (1986), Educating the Reflective Practitioner, Jossey Bass,

San Francisco.

217
Aviation Instruction and Training

Telfer, R. and Bent, J. (1992), 'Producing a workshop for training airline

instructors’, The Journal o f Aviation!Aerospace Education and Research,
vol. 2, no. 2, Spring, pp. 31-38.

Telfer, R.A. and Biggs, J.B. (1988), Psychology and Flight Training, Iowa
State University Press, Ames.

US Department of Education, (1987), What Works - Research about

Teaching and Learning, (2nd ed.), Washington.

218
15 Effective aviation
instruction
[Link]

Although aviation is quite a distinctive teaching context, it is possible to

identify a theoretical base to promote effectiveness and efficiency.
Concurrently, there is a need to review needs and task analysis to improve
instructional design. In this chapter, key research findings are reviewed to
assist in the planning, presentation and follow-up of instruction in aviation.

A starting point

In a vocational training context such as aviation, two strands of learning

theory are most relevant: behavioural theory which relates instructional
objectives with performance, and cognitive theory which deals with short and
long term memory (Ryan, 1990). In themselves, however, these two strands
of theory fail to provide a complete foundation for practice. We now know
why theory has not helped as much as it should have in the pursuit of more
effective and efficient aviation instruction. First, we are aware of the
significant differences that make aviation a specialist instructional
environment (see Chapter 14). Second, it appears that we may have
concentrated too much on the method and too little upon the task. There is
much more information available about how to instruct than there is about
how to divide the instructional task so as to teach it most effectively. This is
especially true of teaching the skills of flying.
Research into part-task training, adaptive training and flight simulation
gives insight into why theory has been unable to actually improve aviation
instruction. Task analysis is the key: being far more important than has
been appreciated. It is vital that we are able to analyse comprehensively the
Aviation Instruction and Training

needs and the instructional task in order to isolate the essential components
and teach them effectively. If the analysis is incomplete, so will be our
instruction. Despite this weakness, we cannot, however, simply dismiss
theory. Without theory,

... uncritical application of ideas developed from intuition will generally

produce disappointing results ... It remains as a critical issue for training
research that there is not yet a widely accepted theoretical perspective that
can guide research and integrate the research findings. (Lintem, 1991, pp.
189,187)

In the absence of a body of research on aviation instruction, the option is to

turn to the literature on instruction in general, keeping in mind the need for
caution described in Chapter 14. We also need to keep in mind that any
review of research (including this chapter) may reflect the subjectivity of the
author.
It is quite impractical, however, to analyse the methods and results of each
of the two thousand or so separate studies which may be relevant to aviation
instruction. At increasingly closer intervals, scholars have provided
authoritative publications which synthesize research findings in an objective
way. A simpler approach is to examine these position papers, and to extract
the implications for aviation, while being aware of their limitations.
For example, the American Educational Research Association periodically
publishes a Handbook o f Research on Teaching, now in its Fourth Edition.
There are other publications to clarify the scientific basis for instruction
(what we know as demonstrable fact, as opposed to subjective opinions
based on personal experience or opinion) such as the synthesis provided by
Dunkin and Biddle in 1974, and the encyclopaedic approach by Dunkin
thirteen years later. More recently, there have been informative reviews by
Cruickshank (1990) and by Waxman and Walberg (1991). Also, there are
published papers such as a recent review of one hundred and seventy five
publications on teaching effectiveness (Reynolds, 1992) and another which
considered two hundred and eighty eight studies (Wang, Haertal and
Walberg, 1990). Together, these sources can point to a scientific basis for
the craft of aviation instruction.
Basically, the current evidence supports the view that aviation instruction
and training should be trainee-centred rather than instructor-centred. In
other words, the instructor considers what the trainee needs to do (or know,
or value); what the trainee can do (or knows or values) prior to instruction;
and plans with the trainee to enable the goal to be attained. It is a

220
 Effective Aviation Instruction

collaborative approach in which the trainee is treated as an adult with a

growing awareness of personal approaches to learning (Telfer and Biggs,
1988) and who can be accountable for his or her performance on set tasks.
As a start, instruction can be divided into three integrated phases:

- presage (pre-instruction);
- process (of instruction); and,
- product (post-instruction). (Biggs and Telfer, 1987)

Following are some clear indicators, summarized in Tables 15.1, 15.2 and
15.3, of the steps instructors can take to make each of these phases effective.

Table 15.1
Effectiveness in the pre-instructional phase

- Cooperative planning with the trainee to establish

goals and collaborative setting of objectives.
- Preparation of instructional aids.
- Pre-testing to establish entering ability.
- Organization of the trainee in advance by discussion
of syllabus, schedule, test, instructional planning and
procedures.

Involving the student so that learning becomes co-operative produces

superior results to those achieved by competitive or individual approaches.
The optimal arrangement is when the instructor and the trainee(s) work
collaboratively in a team approach. This enables the objectives to be jointly
set, with common expectations, and mutual understanding of the criteria that
will determine success or failure. The extent of such instructor/trainee
discussion contributes to growth in knowledge and change in attitudes.
Students who feel that they are able to set their own learning goals and rate
of progress usually experience greater achievement. To some extent, this
becomes a guided self-teaching approach in which the trainee is told what is
to be learned; is assisted to obtain the information and to assimilate the

221
Aviation Instruction and Training

knowledge, attitude or skill; and knows the criteria which test whether or not
the exercise has been mastered.
Making instruction behaviourally-based assists trainee and instructor.
There are clear guidelines for both training and assessment if the objective is
expressed in terms of student behaviour. Consider this example: 'After an
introduction to the effects of controls, the student will be able to return the
aircraft to straight and level flight from slight turns, dives and climbs’. This
objective states what the student has to do; under which circumstances; and
gives a clear criterion of success or failure.

A cooperative approach to learning is superior to that achieved

by competitive strategies
(Photo: Swissair)

Giving the trainee advance organizers (knowledge of what is to be

covered in the lesson) helps both learning and retention. Instructors can
commence by stating, 'At the end of this exercise you should be able to ...', or
'Here are some of the questions you should be able to answer at the end of
this lesson ...', 'Have you ever wondered why ... ?' At intervals, a short

222
 Effective Aviation Instruction

summary can indicate what has been attained to that point, and what is yet to
come.
A related skill is structuring, in which the instructor links new learning
with what the student already knows, and with the overall goal (such as
successfully accepting the responsibility of a pilot in command). An
instructor structures material by providing a series of points which lead the
student to the answer, without actually supplying it. If the student is able to
attain the goal, the way points are reviewed more explicit reference to the
contribution each has to make towards the solution. Often this can be
achieved by questioning, 'Remember the wind direction when we flew
yesterday? How did we compensate in each leg of the circuit? What is the
wind direction today? How should we compensate today?’
Pre-testing alerts both instructor and, more importantly, trainees to what
they do not know. This discovery is the first step towards learning.
Individuals are able to focus on areas of ignorance, and the instructor is able
to ensure that the lesson has appropriate coverage and no needless content.
Pre-tests also help the instructor to model the type of learning being sought.
For example, if students are not using manuals as they should, the instructor
could quite deliberately refer to a manual during the lesson, indirectly
showing students the way professionals seek the answers. At times
indirectness, even though it takes longer, may be more effective (as in the
structuring example above).
It is obviously less time-consuming just to tell the student. One of the
basic tenets of effective teaching is that the focus is on the process of
learning, not simply the product of that instruction. The focus is the student,
not the instructor, nor the content of the lesson. It requires a patient and
student-centred instructor to approach a problem with a series of questions or
exercises that gradually produce the sought student response or insight.
In the planning phase, instructors can ensure that their expertise is so
evident that it should not be doubted. Effective instructors are task-oriented
(getting the job done), organized, and know their subject (and their trainees)
very well. Methods and materials are varied to suit the needs of the students
and to provide novelty in approach. For example, the interaction will vary
from one-to-one, to small group and larger group activities. Information will
be presented in a variety of ways: actual aircraft components, videos,
overhead transparencies, films, charts, programmed instruction sheets,
display boards, cartoons, chalkboard, photographs, models, television and so
on. The emphasis here is clearly on the visual approach to learning.
More visual teaching, making maximum use of visual aids, provides
advantages over conventional chalk and talk approaches. With the visual

223
Aviation Instruction and Training

approach there is evidence that students have higher achievement, rate the
course higher, and more stay to complete it. It is patently impossible for
instructors working to tight schedules to revamp their lesson notes to
incorporate all of the visuals listed above. Indirect means work here, too.
The student lounge can have several poster boards with displays regularly
changed. One airline has a monthly quiz competition, prepared by the Safety
Department, on display behind the coffee machine. Solutions and winners
are also posted publicly.
To assist both instructors and students, an aviation resource centre can be
developed. It need start only as a store-room and can progress to a library
facility in which helpful publications, teaching aids, regulatory and legislative
documents and other resources are available. If the budget stretches to a
coffee machine, refrigerator, a few subscriptions for aviation magazines and
airshow videotapes; and if these are located in the same room, that may be all
the incentive needed for students and instructors to start calling regularly.

Table 15.2
Effectiveness in the instructional phase

- Exploiting the full range of types and applications of

questions.
- Motivation, especially when the trainee initiates
worthwhile contributions or responses to questions.
- Clarity in communication and expectations.
- Flexibility, prepared to take advantage of valuable
but unplanned circumstances.
- Enthusiasm: it's infectious.
- Indirectness, time permitting, may be better than
direct instruction.
- Structuring instruction by linking with prior
knowledge and ultimate goals.
- Criticism only selectively and with self-evident
justification.

In presenting the lesson, the effective instructor will be clear (in

communication, examples, questions and requirements), enthusiastic, and
flexible enough to deviate from the lesson plan to adapt to trainee needs or

224
 Effective Aviation Instruction

unexpected events. To do so, the instructor will be aware of what is actually

going on, being able to empathise with the student and read signs of
frustration, boredom, distraction, inability or enthusiasm. Instructors who
achieve high results from their students are also likely to improve the self
concepts of these individuals. Setting and helping students to achieve high
standards helps to develop the individual while improving task performance.
As Socrates showed, and research since has confirmed, questioning is a
key instructional skill. Questions need to be varied so that they probe both
understanding and opinion as well as factual knowledge and concepts. A
useful technique is to redirect the question to other trainees so that responses
can be compared before instructor comment is provided. This is especially
useful when group discussion is sought, and the use of trainee ideas and
responses may be the best form reinforcement for mature-age students.
Praise improves student achievement by reinforcement, especially when a
'Because ...' extension is added. In other words, the instructor's motivating
comment is not simply: 'Good, Bloggs. It becomes, 'Well done, Bloggs, that
turn was nicely balanced because you coordinated your use of mdder and
aileron'. The use of praise (and destructive criticism or sarcasm) links
directly with the trainee's self-concept. Simply, the higher the self-concept
the higher the likely achievement and the greater the student's own control of
learning.

Instructor preparation

To train instructors to achieve these skills, a microteaching approach has

been shown to be successful. The name is derived from the fact that trainee
instructors start with a short less (about ten minutes) on a very simple topic
with a small group of peers or students (usually about three as a maximum,
initially). The lessons also concentrate on only one skill at a time, such as
questioning, or reinforcing (motivating) students. The length of the lesson,
difficulty of the topic, size of the group, and range of teaching skills are
gradually expanded. Videotaping (usually with private viewing by supervisor
and trainee instructor) and evaluation forms completed by a peer observer
and students provide feedback. The basic plan is to augment natural
strengths (or the art of instruction) with the skills drawn from the craft or
developing science of instruction. The aviation resource centre, discussed
above, is a vital adjunct for instructor training courses.
Simulation is as valuable a teaching device in instructor training as it is in
pilot training. It can range from a 'cockpit' of two chairs behind a table to

225
Aviation Instruction and Training

role-playing, critical incidents and case studies. For example, one airline
uses, with great success, a scenario in which a 'second officer' (another
instructor acting the part) is quite distracted by the training captain’s body
odour or bad breath; another trainee may be under stress from family
problems; a third could be having language comprehension difficulties. The
'trainee' and the observers know the designated problem, not the 'instructor',
who has to present a designated lesson while identifying the trainee’s concern
as soon as possible.
To make such case studies realistic and applicable, they are can be
gathered from experienced instructors in the company. The scenarios can be
stored in a flexible file to which instructors are encouraged to add as they
come across situations which they feel have application in the training course.

Table 15.3
The post instruction phase

- Discuss performance analytically.

- Link attainment with the overall goal of on-the-job
performance.
- Motivate.
- Make sparing use of criticism: evaluating by means
of observable and appropriate criteria.
- Reinforce the trainee's self-concept.
- Link by advance organizer with the next lesson.

Teaching aviation skills

A characteristic which distinguishes aviation training from education is a

concentration on skills. Sometimes termed 'motor control' (see Schmidt,
1988) or the 'psycho-motor domain' of learning, skills are basic to a pilot's
competence. There are several theories of motor learning, of which two have
been influential.
The first theory is the closed-loop explanation (Adams, 1971) which sees
us developing a blueprint (the 'perceptual trace') of a skill, largely from
feedback we are given. 'Much research suggests that information about
performance is the single most important variable for motor learning (except

226
 Effective Aviation Instruction

for practice itself (Schmidt, 1988, p. 454). When an instructor rewards a

good landing, the student does not receive only motivation, but information to
add to the blueprint. It is vital, then, that the instructor ensures that the
blueprint has no errors.
There is also an interesting dilemma for the instructor. On the one hand,
plenty of help and guidance will enable the trainee avoid errors and anxiety.
However, the greater the guidance the less the opportunity for discovery
learning. One solution is to vary the amount of help, concentrating guidance
early in the period of instruction when the skill is new, then gradually
permitting greater freedom to discover.
The second approach is schema theory (Schmidt, 1975) which is based
upon two states of memory: recall (which produces the skill) and
recognition (which tests the execution of the skill). Slow movements are
said to be feedback-based, and fast movements are program-based. Schema
are rules which pilots leam to allow them to generate novel responses (such
as using differential thrust for directional control when other controls are
lost).
What these two theoretical approaches have in common is the view that
skills are learned through a combination of feedback and perceptual
control. In both views, we first perform skills under verbal guidance, then
gradually refine them by detecting errors ourselves and discriminating
between levels of quality in our performance:

The formation of highly refined discriminations among task and

environment features (’percepts') as filters for input processing and as
guides to motor learning has, over time, gained impetus in the study of skill
transfer ... Such percept formation may be a major explanatory concept in
both the improved use of simulators for training ... (Lane, 1987, p. 94)

The good news in our knowledge of the ways skills are acquired is that
there is surprising concurrence in the theory. Whilst a number of theorists
(Fitts et al., 1967; Rasmussen, 1979; Anderson, 1982) have labelled the
developmental stages differently, there is general agreement that there are
three steps. First is the cognitive level in which the student discovers what is
required. Then comes the fixative level when the students follows the
instructor through the procedure, practises, gets feedback, and finally perfects
it. The last stage is the autonomous level when the skill is automatically
executed when it is needed. It is believed that as one becomes more skilled,
one needs to rely less and less on the cognitive and more on the motor.

227
Aviation Instruction and Training

Control of the skill increases at each level. The working conditions needed
for this three level model are summarized in Table 15.4.

Table 15.4
Conditions for learning a skill

Level 1: Cognitive
- Motivation.
- Recognition of the importance of the skill.
- Clear instructions.
- Verbal description or pre-training.

Level 2: Fixative
- Accurate demonstration.
- Rehearsal.
- Timely and helpful feedback: extrinsic first (from the
instructor) then intrinsic (from the execution of the
task).
- Modelling.
- Provision of cues and criteria to check level of
performance.

Level 3: Autonomous Level

- Periodic testing.
- Practice in a variety of circumstances and sequences.
- Random introduction of the skill in practice.

Practice

'The "traditional" method of continuous drill on a particular action (i.e.

practicing a skill repeatedly until it is correct) is probably not the most
effective way to learn' (Schmidt, 1988, p. 400). It is argued that blocked or
'massed' practice is best for performance: random or distributed practice is
best for learning. From that point, however, the evidence is less precise in its
guidance for the instructor.

228
 Effective Aviation Instruction

The choice of proportion of practice and rest, and distribution of practice,

will depend upon the actual exercise. The instructional skill of variability is
still important, especially with adults. Not only is interest retained, but the
learner has the opportunity to see the skill applied in various situations and
generalize the learning.
Again, task analysis is important. In the design of practice, the instructor
will be attempting to ensure that the processes are the same as those to be
used in airline operations. The circumstances need not be identical: hence
the success of part-task trainers. However, the amount of transfer will
depend upon the similarity between the practised task and that required in the
aircraft. Because any differences between the practice and the complexities
of the real thing (operation and feel of the controls, feedback, resistance, and
so on) reduce instructional effectiveness, approximated simulation would be
better suited for procedural training. The inevitability of approximation does
not decrease the value of this form of learning. Many pilots successfully use
their car, the kitchen, lounge chair or closed-eyes visualization in order to
practise procedures.

From part to whole

Task analysis enables the instructor to reduce complex or serial skills (such
as flying a circuit or a missed approach) to their simpler components. Each
activity can be divided into elements with a discernible break where one
element (crosswind) concludes and another (downwind) starts. As each
simple component passes through the three stages of acquisition and becomes
automatized, it can be removed from the instructional focus. In this task
analysis, there is evidence (Wrightman and Lintem, 1985) that backward­
chaining could be considered. This approach reverses the sequence, dealing
with the last aspect (such as the landing in the circuit) first.
There are cases, however, when breaking the skill down removes the
essential difficulty which has to be mastered. To distinguish between part
and whole approaches, the instructor has to analyse the skill carefully. If the
skill is a single operation, usually completed quickly, it calls for whole
treatment. When the skill is continuous rather than sequential, such as
flying a light aircraft IFR, there are tasks which occur simultaneously. In this
instance, practise of the whole rather the parts should be a faster route to
mastery. Schmidt (1988) gives as an example the instruction of helicopter
control. Rotary wing control is achieved by the following:

229
Aviation Instruction and Training

- cyclic pitch control of roll and pitch;

- collective pitch lever control of the vertical;
- throttle control of power; and
- anti-torque pedal control of yaw.

If these skills were learned separately, the interaction between them (like the
primary and secondary effects of controls in fixed wing aircraft) would not be
appreciated.
Rather than providing blocks of practice, instructors can schedule practice
of particular skills (especially those associated with atypical situations such
as loss of power or communications) more randomly so that the trainee is
called on to deliver the skill as it would happen in reality. Another option is
to have the student practice several skills, not aware of the sequence in which
they are to be required during training.
The whole rather than part approach is supported by instruction which was
applied initially to skiing. That experience suggests that aspects of flight
instruction are better suited to simplification of the complex task, rather than
separation into separate stages. The ski instruction progression goes from
’short skis, smooth terrain' to 'longer skis, smooth terrain' and eventually 'long
skis, complex slopes'. The principle of instructional design is that the whole
task is simplified and presented in a simplified environment. This already
occurs in flying schools when trainees move from training aircraft to more
demanding types. The design is essentially spiral rather than linear (Lane,
1987)

The effect of the observer or audience

At a large North American flying school, there is a window in the corridor

behind a simulator, so that one can observe the cockpit and visual display of
a flight in progress. A number of flying schools carry a second trainee as
observer. What is the effect of an audience on a trainee under instruction in
both these cases?
According to Schmidt (1988), the effect on the student varies according to
the level of expertise. If there is a choice, instructors could restrict audience
presence until later in practice when students are more skilled, less anxious,
and less likely to be distracted about others watching them.

230
 Effective Aviation Instruction

Retaining knowledge and skills

We know more about how to instruct people that we do about how to help
people to retain their learning. Instructors need to appreciate the difference
between 'performance' and 'learning' (Lane, 1987), and between a trainee
successfully demonstrating a skill at the end of the lesson, and being able to
reproduce it later when required.
According to cognitive psychologists, we learn by processing information
and storing it so that we can retrieve it in a useful way when we need it (see
Biggs and Telfer, 1987). Because this input has to be linked with prior
knowledge, instructors aid learning by presenting it in a way that helps it to
be 'filed' appropriately in the long-term memory. This 'chunking' is a feature
of efficient learning and recall. One author suggests three criteria for
instructors: ensuring learning is manageable, understandable and memorable
(Farr, 1987)
The instructor's job is not as simple as ensuring that the syllabus is
presented to the student: it is to devise ways of presenting it so that the
student processes and retains the essential knowledge, values or skills. Some
ways of doing so are summarized in Table 15.5.
When performance in learning is hindered because of distractions in the
environment, such as noise, increased effort is required from learners. It has
been concluded that 'richer' coding of the input takes place, and retention is
enhanced (for details of the research, see Lane, 1987, pp. 95-97). There is
trade-off here which may have value in aviation instruction. If the learning
takes place in a similar context to that in which it is to be applied, such as a
noisy flight deck with unpredictable demands and interruptions, the
instruction will certainly take longer and cost more. On the other hand, the
retention and on-the-job performance will probably be greater.

Conclusion

In order to professionalise flight instruction and begin a systematic refinement

of both methods and materials, the theoretical foundations upon which
industry practice is based needs to be made explicit. The tradition of
continuing what has been done in the past is an inadequate response to the
challenges of the present and, especially, the future.

231
Aviation Instruction and Training

Table 15.5
Helping students retain knowledge and skills

- Design the instruction to resemble as closely as possible, the

on-the-job or real-world operation. Relate it to flying.
- Link new knowledge or skills with those the trainee already
possesses.
- When teaching facts or rote learning, provide rules, principles
or concepts to provide a framework and memory-code for them.
- Help students to recognize and attain the third, autonomous level
of skill acquisition.
- Help the student to process new material in a deep, rather than
a surface way. Strive for understanding and application rather
than simply remembering.
- Heighten expectations for performance by increasing the
criterion for the instructional objective to a level higher than that
required by the syllabus; by reducing the amount of assistance and
aids to learning; and by introducing types of interference or
distractions (such as interruptions, radio calls, diversion or other
simultaneous demands).
- Require the trainee to complete several (not simply one)
successful performances of the task before terminating the lesson.
- For procedures and check lists, unrelated facts or rote learning,
use words or associations (such as acronyms and mnemonics)
to provide links between steps. The more personal and
unusual, the better.
- Provide all trainees with well-structured learning materials, and
allow the high-achievers to work at their own pace.
- Encourage imaginary or simulated rehearsal in which skills and
knowledge are acted out to prevent decay.
- Decay of complex skills can be slowed by practice of one central
aspect in a part-task trainer.
- Procedural skills can be relearned from summary cards without
need to re-execute the full procedure.
- To avoid superfluous re-training, start by testing to ascertain
what is NOT remembered.
- Maintain training records which will assist decisions about typical
progress for each segment; identification of students having
difficulties; determining course and segment duration for both
groups and individuals; and evaluation.

(Adapted from Farr, 1987, and Lane, 1987)

232
 Effective Aviation Instruction

References and further reading

Adams, J.A. (1971), Ά Closed-Loop Theory of motor learning', Journal o f

Motor Behaviour, vol. 3, pp. 11 -149.

Adams, J.A. (1987), 'Historical review and appraisal of research on the

learning, retention and transfer of human motor skills', Psychological
Bulletin, vol. 101, pp. 41-74.

Anderson, J.R. (1982), 'Acquisition of a cognitive skill', Psychological

Review, vol. 89, no. 4, pp. 369-406.

Arends, R.I. (1988), Learning to Teach, Random House, New York.

Bangert-Drowns, R.L., Kulik, C.C., Kulik, J.A. and Morgan, M. (1991), The
instructional effect of feedback on test-like events', Review o f Educational
Research, vol. 61, no. 2, Summer, pp. 213-238.

Biggs, J.B. and Telfer, R.A. (1987), The Process o f Learning (2nd ed.),
Prentice Hall, Sydney.

Brandt, R. (1992), ’On research on teaching', Educational Leadership, vol.

49, no. 7, April, pp. 14-19.

Branyan-Broadbent, B. and Kent Wood, R. (eds.) (1992), Educational Media

and Technology Year Book, vol. 17, Libraries Unlimited, Englewood.

Buchmann, M. (1987), 'Teaching knowledge: The lights teachers live by',

Issue Paper, 87-1, Michigan State University Institute for Research on
Teaching, East Lansing.

Cruickshank, D.R. (1990), Research that Informs Teachers and Teacher

Educators, Phi Delta Kappa, Bloomington.

Dunkin, M.J. (ed.) (1987), The International Encyclopedia o f Teaching and

Teacher Education, Pergamon, Oxford.

Dunkin, M.J. and Biddle, B J. (1974), The Study o f Teaching, Holt, Rinehart
and Winston, New York.

233
Aviation Instruction and Training

Engelmann, S. and Camine, D. (1982), Theory o f Instruction: Principles

and Applications, Irvington, New York.

Farr, M.J. (1987), The Long-Term Retention o f Knowledge and Skills - A

Cognitive and Instructional Perspective, Springer-Verlag, New York,
1987.

Fitts, P. and Posner, M.I. (1967), Human Performance, Brooks/Cole,

Monterey.

Gagne, R.M. (ed.) (1987), Instructional Technology: Foundations, Lawrence

Erlbaum, Hillsdale.

General Accounting Office (1990), Washington. Program Evaluation and

Methodology Division, Report No. GAOIPEMD-91 -4, (ED 326 649),
October.

Holding, D.H. (ed.) (1981), Human Skills, Wiley, Chichester.

Jenson, W.R., Sloane, H.N. and Young, K.R. (1988), Applied Behaviour
Analyses in Education: A Structured Teaching Approach, Prentice Hall,
Englewood Cliffs.

Kember, D. (1991), 'Instructional design for meaningful learning',

Instructional Science, vol. 20, no. 4, pp. 289-310.

Lane, E.L. (1987), Skill Acquisition Rates and Patterns - Issues and Training
Implications, Springer Verlag, New York.

Laurillard, D. (1992), 'Mediating the message: television program design and

students' understanding’, Instructional Science, vol. 20, no. 1, pp. 3-24.

Lesgold, A. (19880, 'Toward a theory of curriculum for use in designing

intelligent instructional systems', in Mandle, H. and Lesgold, A. (eds.),
Learning Issues o f Intelligent Tutoring Systems, Springer-Verlag, New
York, p. 118.

Lintem, G. (1991), 'Instructional strategies', in Morrison, J.E., Training fo r

Performance: Principles o f Applied Human Learning, Wiley, New York,
Chapter 6.

234
 Effective Aviation Instruction

Melton, A.E. (ed.) (1947), Apparatus Tests, US Government Printer,

Washington.

Miller, W.R. and Rose, H.C. (1975), Instructors and their Jobs (3rd ed.),
American Technical Society, Chicago.

Millis, B.J. (1991), 'Enhancing adult learning through co-operative small

groups’, Continuing Higher Education Review, vol. 55, no. 3, Fall, pp.
144-154.

Phillips, D.C. (1980), 'What do the researchers and the practitioners have to
offer each other?’ Educational Researcher, vol. 9, pp. 17-20.

Rasmussen, J. (1979), On The Structure O f Knowledge: A Morphology O f

Mental Models In Man-Machines Systems Context, Report M-2192, Riso
National Laboratory, Denmark.

Reigeluth, C.M. (1983), Instructional Design Theories and Models: An

Overview o f their Current Status, Lawrence Erlbaum, Hillsdale.

Reigeluth, C.M. and Curtis, R.V. (1987), 'Learning situations and

instructional models', in Gagne, R.M. (ed.), Instructional Technology:
Foundations, Lawrence Erlbaum, Hillsdale, pp. 175-206.

Reynolds, A. (1992), 'What is competent beginning teaching? A review of

the literature', Review o f Educational Research, vol. 62, no. 1, Spring, pp.
1-35.

Ryan, R. et. al. (1990), Approaches to Learning Within the Vocational

Education Context, (ED 324 521) Arapa, USSR.

Saettler, P. (1990), The Evolution o f American Technology, Libraries

Unlimited, Englewood.

Schmidt, R.A. (1988), Motor Control and Learning: A Behavioural

Emphasis (2nd ed.), Human Kinetics Publishers, Champaign.

Schon, D.A. (1986), Educating the Reflective Practitioner, Jossey Bass, San
Francisco.

235
Aviation Instruction and Training

Telfer, R.A. and Biggs, J.B. (1988), Psychology and Flight Training, Iowa
State University Press, Ames.

U.S. Department of Education (1987), What Works - Research about

Teaching and Learning (2nd ed.), Washington.

Wang, M.C., Haertal, G.D. and Walberg, H.J. (1990), 'What influences
learning? A content analysis of review literature’, The Journal o f
Educational Research, vol. 84, no. 1, September/October, pp. 30-57.

Waxman, H.C. and Walberg, H.J. (1991), Effective Teaching: Current

Research, McCutchan, Berkely.

Williams, C.J. and Brown, S.W. (1992), Ά review of the research issues in
the use of computer-related technologies for instruction', in Bronyan-
Broadbent, B. and Kent Wood, R. (eds.), Educational Media and
Technology Year Book, Libraries Unlimited, Englewood, no. 17, pp. 26-
46.

Wittrock, M.C. (ed.) (1986), Handbook o f Research on Teaching (3rd ed.),

Macmillan, New York.

Wrightman, D.C. and Lintem,G. (1985), 'Part task training for tracking and
manual control', Human Factors, vol. 27, pp. 267-283.

236
16 Academic training for
the ab initio pilots
B. Blyth

With the rapid expansion of Commercial Aviation in the late 1950s and early
1960s, it became apparent that the traditional sources of pilots (viz. the
military and general aviation) were not capable of forming a sufficient
supply to meet the needs of the industry. This was not a universal problem
in as much as certain countries (e.g. USA and Australia) 'enjoyed' the
benefits of a significant General Aviation sector in relation to the demand of
their Commercial Airlines.
In Europe and in the emergent nations however, the balance of supply and
demand was very much weighted on the side of demand and a method of
supply had to be created.

A b in itio training - the beginning

In the United Kingdom the then Minister of Transport set up a committee

under the Chairmanship of the Duke of Hamilton. The Committee was
tasked to devise a scheme by which accelerated training could be given
which would achieve the following aims:

- Instill flying skills.

- Instill the necessary theoretical knowledge .
- Using full integration, allow the flying skills and theoretical knowledge
to be mutually re-enforcing thus reducing the need for prolonged
exposure to 'training by osmosis'.
Aviation Instruction and Training

All of this had to be achieved in the minimum time and at an affordable

price.
The scheme resultant from this study became known as the 'Ab Initio
Integrated Training Course For Commercial Pilots'. It consisted of 200
hours of flying training and 550 hours of ground training (all of which could
be completed in a period of 13 months) culminating in the issue of a
Commercial Pilot Licence with Instrument Rating. Graduates from such
courses possessed the skills and knowledge to allow them to either:

- command a single crew aircraft for activities such as Air Taxi or

similar;
- act as co-pilot on a two crew general aviation type; or
- act as third pilot (systems operator) on a larger more complex aircraft.

This training system, with various modifications to the syllabus, still

forms the basis of the vast majority of ab initio training schemes around the
world. It is traditional, recognized, economic and (perceived to be) tried
and trusted. It is however 30 years old and, in the intervening period of
time, the airline industry has undergone radical changes in scale, equipment
and technology.
To a greater or lesser extent, changes in the syllabus have kept pace with
the changing technology. However, one major change has occurred which
cannot be answered merely by adapting a syllabus. This is the
disappearance of the Systems Operator, which has been made possible by
the new technology, and the introduction of the two crew flight deck.
Although on the modem 'long-haul' flights three pilots are still carried and
thus there is room for a 'tyro', this is not the case on the short to medium
range aircraft of the 1990s.

The airline ab in itio pilot

It is increasingly apparent that the traditional role of the pilot in an airline

operation is undergoing change which is as radical as the changes in
technology in the environment in which they are employed. The traditional
role of 'operator' has been changed to that of system manager, decision
maker and strategy planner. Ab initio trainees from a 'traditional' scheme are
poorly prepared for this role as their training will have concentrated on the
technologies of the aircraft and of the working of a 'single crew' operation.
Little thought will have been given to 'Human Factors' (other than the

238
 Academic Training fo r the Ab Initio Pilots

minimum required by the authorities), crew cooperation or strategic

planning.
A new, more embracing, approach to training was called for; one which
would achieve the aim of producing a well trained professional whose skills
of operation and management in the multi-crew environment of a modem
aircraft would belie their apparent youth and lack of flying hours
'experience'. Graduates from such a course would have to be capable of
completing a standard conversion course and so assume their rightful
positions as 'route checked’ first officers within 3 months of graduation.
A number of schemes have been raised to 'train the professional' of which
the two acclaimed as most successful are the British Aerospace Hying
College Airline Pilot Course, and the Lufthansa 'Futura' scheme. Both of
these schemes aim at a 'seamless' training philosophy in which the ab initio
training package includes all the statutory requirements for licence issue plus
the progression onto initial preparation for multi-crew and large transport jet
operations.
The training aims for such a course are very clear and are summarized as
follows:

1. SkiUs:
The graduate must demonstrate:
- suitable levels of skill both in a single crew (early training) and a
multi-crew environment;
- a good level of understanding of Cockpit Resource Management;
and
- capability in operating with a high degree of self discipline.
2. Knowledge:
The graduate must demonstrate:
- a good understanding of the basic and advanced theory on the basis
of which the modem aircraft is designed and efficiently operated;
- an ability to apply constructive and diagnostic thought to a problem;
- a knowledge of the needs of people in a team (crew) situation; and
- an understanding of world wide politics and economics and their
resultant effects on the airline industry.
3. Integration:
In order for the trainee to gain maximum benefit from the course both
'skill' and 'knowledge' should be developed in tandem, each being used
to intensify the 'learning experience' of the other.

In these courses, the design and phasing of the practical and academic

239
Aviation Instruction and Training

training packages are critical to the outcome of the course.

The ground training curriculum

In the safe and efficient operation of a modem aircraft the flight crew
requires to be aware, to varying degrees, of the answers to some fairly
fundamental questions such as:

- Where, in a three dimensional world, is the aircraft located at a specific

time?
- What degree of reliance can be placed on this information?
- Is the aircraft behaving as the instruments indicate?
- Is the aircraft being operated in the safest and most efficient manner?
- Is the aircraft being operated within the bounds of national and
international aviation law?
- Can faults be detected and diagnosed quickly and accurately?
- Can an awareness of diagnosed faults be used to decide on immediate
and progressive actions or plans?
- What external factors can effect the operation of the aircraft?
- How are the human components working (both inside and outside the
aircraft)?
- How effectively are 'communications' operating?

In order to effectively deal with these questions, the pilot must have a sound
knowledge of the aircraft, its performance capabilities, its equipment, the
environment in which it operates and the 'people factor'.
As people gain in experience the answers are given - frequently before the
question is asked. For those lacking in experience (the ab initio trainee) a
sound education is necessary in order to allow the questions to be
formulated for, in formulating the questions, the trainee is thinking and so
professionalism is bom.
The trainer should raise such a list of questions (perhaps by the use of a
questionnaire completed by experienced flight crew). This will enable the
trainer to arrive at conclusions regarding the fields of knowledge required
and the depths to which that knowledge should be gleaned.
Take, for example, the first question 'Where in the three dimensional
world, is the aircraft located at this time?' This question, in a more
expanded form asks:

240
 Academic Training fo r the Ab Initio Pilots

- the position in a horizontal plane;

- the position in a vertical plane; and
- the measurement and awareness of time.

Awareness and determination of position in the horizontal plane requires a

knowledge of the following:

- the reference system used;

- the methods of presentation of the reference system;
- methods of determination of position within that reference system;
- accuracy of that position; and the
- limitations of the methods used.

On examination of the 'reference system' it becomes apparent that

knowledge is required of:

- units and datum points;

- the graticule; and
- variances created by irregularities.

All of which, in turn, will require knowledge of:

- angular measure;
- the sphere;
- great circle and small circle; and
- the terrestrial spheroid

Each question can be dealt with in a similar hierarchical manner and each
will reveal a number of 'fields of knowledge'. These may be grouped
together into suitable and convenient sections each of which should have a
common theme or subject area. A sample of such a grouping is shown in
Figure 16.1.

The ground training syllabus

From the curriculum evolves the syllabus. This consists of a description, in

very brief terms, of the individual lessons that constitute a course. The
syllabus design method now most commonly used (and favoured in the
aviation environment) is the objective style. In this, the subject area is

241
Aviation Instruction and Training

Figure 16.1
Navigation

1. Theory
(Form of the earth and aeronautical charts)

1.1 Great circle; small circle; rhumb line; convergency;

conversion angle. Latitude; difference of latitude.
Longitude; difference of longitude. Change of time
with longitude. GMT, LMT, Standard Time and
conversion from one to the other. Date line.
Determination of the times of sunrise, sunset and civil
twilight at sea level using the 'Air Almanac' or
extracts from that publication.
1.2 Units of distance and height used in navigation, viz.
nautical miles, statute miles, kilometres, metres,
yards, feet. Conversion from one unit to another.
1.3 The general properties of the Lambert Conformal
Conic and Cylindrical projections, including the
representation of meridians, parallels, great circles
and rhumb lines. The use of the current aeronautical
charts; plotting positions and bearings; measuring
tracks and distances. Methods of indicating scale
and relief. Conventional signs.

2. Practical

2.1 Track; heading (true, magnetic, compass); wind

velocity; airspeed (IAS, RAS, TAS); groundspeed;
ETA; drift; DR position, air position, ground
position, fix.
2.2 The triangle of velocities; methods of solution for the
determination of heading, ground speed, wind
velocity, track and drift angle. Time and distance
problems. The use of navigational computer.
2.3 Practical operational range and useability of charted
facilities.

242
 Academic Training fo r the Ab Initio Pilots

divided, if necessary, into sub-groups and then into teaching units. Each
teaching unit must have an 'aim' and an 'objective'. The 'aim' is a statement
of the role of that unit within the totality of that subject area. The 'objective'
is the training level which must necessarily be achieved in order to fulfil the
'aim'. A 'teaching unit' may consist of one or more lessons. Examples are
given in Figure 16.2.

Figure 16.2
Example from the Ground Training syllabus

Navigation
Theory Practical

1 2 3
Form Of Earth Charts Time

1 2 3
Earth True Direction
Shape Graticule And
Spherical Distance

Teaching Unit No.l

1. Earth Shape And Spherical Geometry:

Aim: To provide a comprehension of the
fundamental geometric considerations of the
sphere.
Objective: The trainee should,
a. without reference, be able to recall the definitions
of great circle, small circle, axis, angular measure;
and
b. be aware of the spherical shape of the earth and its
effect on are length at the surface.

Having evolved the content of the syllabus, a 'critical path analysis' is

243
Aviation Instruction and Training

carried out in order to establish the integration of the theory being taught
with the practical skills being gained in the flying training. Integration may
be achieved in one of two ways viz. Block Integration or Continuous
Integration.
In Block Integration flying and ground syllabi content are divided into
'skill packages’. Each flying skill package is preceded by a block of
compatible theory training. Such a syllabus may appear in a schematic form
as shown in the Figure 16.3.

Figure 16.3
Schematic form of the syllabus

0 10 20 Phase
Duration

Phase 1 A 280 periods 8 weeks

(PPL) Ground Training
Duration B Basic flying hours 8 weeks

Phase 2 A 525 periods 15 weeks

(CPL) Ground Training
B Intermediate Flying 105 hours 16 weeks
Simulator 20 hours

Phase 3 A 525 periods 15 weeks

Ground Training
B Advanced Flying 45 hours 12 weeks
Advanced Simulator 80 hours

This scheme gives continuity of flying during specific phases but

obviously not throughout. Additionally continuity of ground training is
broken. These will inevitably lead to a need for considerable
review/reffesher at the commencement of each new phase. As a result of
this some duplication of training effort will occur.
The scheme is of use where the various flying phases are completed at

244
 Academic Training fo r the Ab Initio Pilots

different sites and/or where the trainee is sequentially qualified to and issued
with the various licences (PPL/CPL/frozen ATPL').
Continuous Integration is where the course is run on the basis of alternate
days of theory and practical training following an induction period of full
time theory training. The proper use of this method ensures an efficient
learning system with little or no repetition created from lack of continuity of
either theory or practical aspects of training. Figure 16.4 demonstrates the
possible phasing of such a scheme.
Each of these schemes has the proven capability to reduce the transition
from ab initio trainee to first officer on a two crew aircraft and so save the
airlines a significant transition training expenditure.

The lesson plan

In order to provide the trainee with a suitable learning experience it is

necessary that the Ground Instructor should adequately prepare the material
required for each lesson or teaching unit (shown in Figure 16.5):

1. Introduction - this will introduce the aim of the lesson content and, of
necessity, should include some form of motivator i.e. a reason for the
student to apply himself/herself to learning. The experienced instructor
will frequently introduce some humour into this point perhaps in the
form of a cartoon graphic. A brief recap of items previously taught
may be necessary if such information is pertinent to the lesson theme.
2. The Theme - this should be developed as a logical follow on to the
introduction. Suitable view graphs, slides and other aids should be
available, arranged in their order of use and numbered accordingly, as
well as indications of board work.
3. Recap - this involves a quick resume of the material and a carefully
designed question and answer session.

A training period should not exceed 55 minutes in duration and can be

ideally divided into: Introduction (10 minutes); Theme (30 minutes); and
Recap (15 minutes).
On completing a lesson the instructor should carry out a 'self critique'
identifying those points of his/her performance which were felt to be
'uncomfortable' or poorly received. The validity of the graphs/slides and
other aids should be reviewed for accuracy and pertinence to lesson. Timing

245
Aviation Instruction and Training

Figure 16.4
Airline course structure

Phase 1 Basic Ground Training 273 periods

(Duration 8 weeks)

Phase 2 Intermediate Ground Training 455 periods

Basic Flying 55 hours
(Duration 14 weeks)

Phase 3 Advanced Tround Training 455 periods

Intermediate Flying Training 90 hours
Flight Procedures Trainer 15 hours
(Duration 24 weeks)

Phase 4 Advanced Ground Training 140 periods

Intermediate Flying Training 10 hours
(Duration 4 weeks)

ATPL Examination

Phase 5 Advanced Flying Training 45 hours

Flight Simulator 25 hours
(Duration 8 weeks)

Phase 6 LOFT 70 periods

Flight Simulator (JET) 60 hours
(Duration 6 weeks)

246
247

Figure 16.5
Lecture brief
Aviation Instruction and Training

should also be reviewed as it is essential that a training period should be

filled but equally, a particular lesson should not repeatedly be completed in a
time much less than that allocated. When reviewing the lesson be sure to
take into consideration such factors as:

- whether the trainees have undergone a properly balanced selection

system; and
- whether the trainees are being asked to complete a course in a language
other than their 'mother language'.

At all times the lecturer must guard against the assumption (all to easily
fallen into) that the trainees have the same grasp of the basics as the lecturer.
The smile on a cadet's face does not necessarily mean Ί understand', but may
frequently mean 'I've got wind'!

The learning environment

In order to obtain maximum value from the learning process it is absolutely

essential that the learning environment should be given careful
consideration. Environments in which noise and visual distractions abound
are worse than useless but by a similar token, an airless, claustrophobic
environment is of little value since the soporific effects far outweigh the
screening from distractions.
Lecture rooms should be light and airy with windows (double glazed)
which give some form of non-distracting view (e.g. distant hills). Sound
insulation should be effective and, therefore, good air conditioning, with a
humidifier, is essential. Decor should not use restful colours such as fawns,
golds and browns, but should concentrate on that part of the colour spectrum
around the blue/grey area. Lighting should be adequate and fitted with
antiglare devices. Furnishing should be functional rather than aesthetic.
Student work stations must be provided with a comfortable seat and
adequate working surface with storage for books or bag.
Lecture rooms should be sufficient in numbers to accommodate up to 70
per cent of the College capacity. They should be grouped together in one
learning centre ideally co-located with the flight operation, simulation and
aircraft engineering establishments. Discrete 'quiet rooms', seating up to
four, should be provided for periods of intensive study. These should lead
off the library area so the maximum use can be made of the reference
material contained there.

248
 Academic Training fo r the Ab Initio Pilots

It is absolutely essential that the staff offices should be adjacent to the

learning centre thus facilitating the cadets' access to their mentors. Various
other rooms must be arranged and suitably equipped for specific roles, viz.
demonstration rooms with working modules, components, wind tunnel etc;
communications laboratory for use in R/T training and as a language
laboratory.
In addition to attending to the learning needs of the trainees it is very
important to remember that they will need to 'let off steam'. The provision of
adequate sports and recreation facilities must be an integral part of the
training environment and should include at least a gymnasium, swimming
pool and a multi sports area or games hall (depending on the climate).
Properly organized sport can be of immense value not only from the
recreational/psychological point of view but also for the inculcation of
leadership and teamwork skills. One of the finest teamwork/leadership/
recreational exercises is yachting, and any reputable College located near a
suitable body of water should give serious consideration to the ownership of
a boat requiring a crew of six to eight.

Computer based training (CBT)

In recent years a great degree of interest has been generated in the use of
self-paced learning systems which will enable ab initio cadets to reinforce
the traditional training. The use of computers can considerably enhance the
self paced learning concept by exploiting the benefits of:

- clear graphics;
- animation;
- audio commentary;
- touch responsive screens;
- interactive video; and
- video/graphic mixing.

The 'objective' style syllabus readily lends itself to computerization, and

provided the courseware is properly structured and designed, the resultant
lessons will considerably enhance the learning process by:

- improving standardization;
- allowing cadets to learn at the pace best suited to the individual;
- allowing cadets to revise/review lessons without embarrassment;

249
Aviation Instruction and Training

and by
- allowing the instructor to give more time to each individual.

A networked system of courseware delivery stations will also provide the

training manager with a management tool by means of which the order of
access to courseware can be controlled for each individual trainee and
feedback of a trainees learning rate and achievement can be recorded.
The major disadvantage of CBT is the perception frequently given that a
properly constructed course given entirely on computer removes the
necessity for formal and traditional teaching. In the case of ab initio trainees
this is entirely unacceptable for, at this very early stage of their career,
trainees need the personal attention of lecturers who have not only
knowledge but also experience to pass over. It must also be clearly
understood that the education of ab initio trainees includes theories and
concepts which, in a computerized form, cannot be fully dealt with in a
media that excludes the interaction between the student and the lecturer.
In the light of this, many training establishments would find it difficult (if
not impossible) to justify the considerable expense involved in setting up a
CBT system in spite of the training value that derives from such a system.

Testing, examination and evaluation

There is no single method of testing, examining or evaluating that will

provide a 100 per cent guaranteed indication of a person's knowledge or
capability. In broad terms a person's knowledge or capability can be
evaluated either by on going assessment or by formal testing.

Assessment

By this method, the experienced lecturer can observe the trainee on a daily
basis and, by observation and careful questioning, can gain an impression of
the trainee's ability to learn and his/her retention. This method is extremely
subjective however and can easily be distorted by such elements as:

- personality clashes;
- subject material (boring/interesting);
- skill of lecturer (boring/interesting);
- experience of lecturer; and a
- lecturer's resistance to 'marking a trainee down'.

250
 Academic Training fo r the Ab Initio Pilots

Assessments should not be discounted however, as they can provide a

powerful overall view of each trainee and the lecturers themselves.

Examination

This provides the means by which a more detached evaluation of a trainee’s

academic achievements can be established. Much has been written about
the methodology of testing and the reader is recommended to consult the
ICAO publication DOC 7192-ANI857, Part A l, which contains an excellent
resume of these debates and writings.
The writer will restrict himself to comment only on the broader issues
such as:

- Why do we examine trainees?

- Where should examinations for licence issue take place?
- Should licence examinations be taken over a period of months (one
subject or one block at a time) or should they be given as a single diet
of examination?

The answer to the first point is patently obvious and that is that the
trainee must demonstrate his/her level of knowledge of the syllabus material.
In many areas, there is a tendency towards proving the trainee's theoretical
knowledge before any practical experiences given.
This cuts across the fundamental advantage of integrated training and
results in a situation where a trainee can leam (by role) the answers to
questions but will not be able to relate this material to real life for some
time. This frequently presents problems at a later point when, for example,
the newly qualified pilot is required to do the first complex aircraft
conversion.
It is preferable to continue the ground training throughout the ab initio
course and raise the trainee for examination at or near the end of that
training period.
There has been in some areas, the ability to take examinations in
individual subjects or blocks, gradually building sufficient credits for the
issue of a licence. Although this may be fine for a pure ’academic' course it
is of little value to the training of a professional pilot since it encourages, in
the trainee the philosophy of 'cram for exam and forget'. This may make it
easy for the examination candidate the time but will make life difficult for
the training department of the airline to whom they offer their services. In

251
Aviation Instruction and Training

the end trainers and airline sponsors must consider what the aims of ab initio
training are and how they can best be served.
Perhaps the following questions should be considered:

- Is it sufficient just to achieve the minimum standards demanded by the

examination?
- Should the ab initio training period be used as an extension of a
selection system?

Conclusion - the way forward

Theoretical training in aviation has always been treated as 'a necessary evil’
in the training of a pilot. It has taken the role of the support to the necessary
manual skills which pilots have to display in order to prove their
professional competence.
With the radical technical changes which have swept the industry the
pilot’s manual skills, although still important, have been relegated to a
secondary role whilst the pilot's skills in other areas have taken a much
higher priority. To recognize this situation and its continuing development,
the various regulatory authorities must constantly review the role of the pilot
whilst the industry must accept its responsibilities, cooperating with the
authorities and demanding the highest quality of training at all level.

252
17 Preparing better flight
instructors
C. Elshaw

Over the relatively short life of aviation there have been dramatic
developments in the technology associated with air transport. Hand in hand
with developments in vehicle design have been comparable developments in
training systems. However, these have been almost entirely devoted to
training hardware, culminating in the sophisticated multi-million dollar flight
simulators which incorporate state of the art visual systems and six-axis
synergistic motion systems. Unfortunately, there has been little comparable
progress in the practice of flight instruction. An examination of basic flight
training shows that there have been no dramatic improvements in training
practice comparable to the developments in training hardware. Roscoe’s
(1980) comments seems particularly apposite:

Pilot training in the United States has grown like Topsy, with little
guidance in its development. Basic research in human learning has had
virtually imperceptible effect on the configuration of either civilian or
military pilot training programs despite lip service that it continually
receives. The popular term 'training technology' is a euphemism for trial
and error.

Although there has been a growing research literature on instructional

techniques it has had little impact on the practice of flying training. Flight
instructor training, when it is formally taught, has changed very little since
Major Smith-Barry first codified his principles of flight training in 1916 while

The views expressed are those of the author and do not necessarily represent
those of the Ministry of Defence.
Aviation Instruction and Training

serving at the Central Flying School in the UK. He was the first person to
insist on standardization in flight training based on the instructor
demonstrating the flight manoeuvre while at the same time describing what he
was doing - generally known as the instructor’s ‘patter’. This is perhaps
highlighted by a retired air force pilot, who undertook recurrent training in
1982, who said, 'the training has not changed very much over the years. The
exercise, the instructor’s "patter", the technique, are much the same as they
were when I first took my flying training back in England in 1946' (Mold,
1982).
It is generally thought that the main qualification for a flight instructor -
and this is common to many other professions - is that the person needs to be
a good practitioner of the art. This together with an enthusiasm for the task is
seen as sufficient to instil enthusiasm and proficiency in others.
It is not usually recognized that being a good teacher has its own skills and
knowledge which are generalizable and independent of the subject domain.
There persists a belief that the good instructor develops insight into the
mysteries of instruction spontaneously through experience.
This emphasis on the instructor as an outstanding practitioner has produced
a corresponding emphasis on the content of the training syllabus but at the
expense of the process of learning to fly. As Telfer (1983) has said in the
context of Australian flight instruction:

There remains the tendency ... to see teaching, rather than learning to be
pre-eminent. The instructor’s strength is that of personal skill in a
specialist field. This leads to an emphasis on content expertise in flight
instruction (placing a liability on the learner to observe and replicate the
skills) rather than a commensurate weighting being given to the importance
of the process of learning to fly (which would recognise the central role of
the learner, rather than that of the instructor).

This situation has been supported by a prevailing philosophy in flying

instruction. This has been effectively described by Ronden (1969) who
undertook a study into elementary flying training in the Royal Netherlands
Air Force. The study resulted from an analysis of the unreliability of pilot
selection tests. He concluded that the major factor contributing to this was
the unreliability in the training process. As a consequence he set about a
systematic investigation of flying training. He reported that the then existing
educational philosophy on flying training was based on two assumptions.
The first assumption was that only those with a distinct natural aptitude for
flying would succeed and that most instructors could identify these ‘natural

254
 Preparing Better Flight Instructors

pilots’ early in training. The second assumption was that subjecting students
to high degrees of stress helps to identify and weed out those students
without natural aptitude. This resulted in training becoming an unrewarding
experience for students who see every exercise as a potential 'chop ride' and
reduces the instructor’s role to that of aptitude selector.

The changing scene

In the above study by Ronden the research psychologists introduced a

program of training which for instructors included discussions on aptitude,
stress, fear of flying, assessment, learning, motivation, rate of learning etc
together with the introduction of five new principles which were:

1. To give the student-pilot an easy introduction to flying by refraining from

any effort to give instruction dining the first few flying hours, so the
student-pilot could relax and get acquainted with the new situation of
moving about in three dimensions.
2. To leam to amend the stereotyped way of giving flying instruction, and
allow the instructor to give special attention and flying time to the
learning difficulties of each individual student-pilot.
3. To postpone the 12 hours flying test until such a time that 'the slow
starter' had made sufficient progress.
4. To improve human relations between instructors and student-pilots, for
instance by promoting social contact during on and off duty hours.
5. To let the student-pilot assess his own progress by means of grade-slips
and to have this assessment corrected and discussed by his instructor.
Under the old system the student pilot was most of the time unaware of
his progress so that elimination sometimes came as a complete surprise.

The researchers had to persevere against much opposition and ridicule but
nevertheless the program increased success rates from 50 per cent to 80 per
cent. Even then the improvement in the results was attributed to
improvements in the quality of the student intake and it was predicted that
wastage at the next stages of flying training would increase. All the excuses
proved to be foundless and the predictions groundless.
However there are encouraging signs that the situation is now changing and
a new climate of open-mindedness developing. With the increasing
complexity of aircraft, the contribution of psychological factors to the efficient
operation of the man-machine system has now become accepted. The

255
Aviation Instruction and Training

increasing attention being given to training in cockpit resource management

points to a changing situation. This may well lead to greater attention being
paid to the psychological factors associated with basic skill acquisition. The
UK Royal Air Force has taken the lead and includes discussions on
'Psychology of Instruction' in its pilot training instructors’ course. It has also
recently introduced changes to its philosophy of Basic Flying Training. This
experiment known as 'the training initiative' has incorporated ideas similar to
those proposed by Ronden. The results so far have been very encouraging
and are currently being evaluated.

Why students fail

In a study which addressed the topic of training for high workload O’Reilly
and Elshaw (1987) examined the reasons for suspension during Basic Flying
Training in the RAF. The frequency of reported reasons for suspending
students is given in Table 17.1.

Table 17.1
Most frequently quoted reasons for students failing
basic flying training

Factor Frequency

W orkload/capacity 66%
Awareness 48%
Slow thinking 36%
Retention 35%
Performance under stress 17%
Consistency 16%
Decision making 15%
Accuracy 15%
Progress 12%
Over-confidence 11%

(From O’Reilly and Elshaw 1989)

256
 Preparing Better Flight Instructors

The most frequent reason given was an inability to cope with workload and
what was referred to as spare mental capacity, lack of awareness, slow
thinking and poor retention of instruction. This would probably be a typical
list were the views of any other instructors to be taken. It is possible to
explain this situation by saying that the students had reached the limit of their
natural abilities. However, the same reasons tended to be given whether the
student failed early in the course, when the workload was relatively low or
late in the course when it was high. Again this could be said merely to
represent the range of abilities to be found in the student population.
However, an examination of some 'psychological issues' in skill acquisition
might provide a different perspective on the situation. Three issues will be
discussed, the theory of skill acquisition, motivation and stress and anxiety.

Skill acquisition

A very simple overview of the process of skill acquisition was given by Fitts
(1964) who described the development of skill acquisition as passing through
three distinct stages.

The cognitive stage

The first stage is called the cognitive stage. Here students attempt to
understand what it is that they are required to do. The skill is broken down
into discrete meaningful actions which can be verbalized. For very simple
tasks this could be in terms of a single sentence instruction. For complex
tasks, however, there are many rules and procedures to leam and understand.
There are also relationships between sets of mles and the fact that rules are
modified under different circumstances. This cognition takes place in working
memory. Working memory has a very limited capacity: people can only really
attend to one concept at a time. Research has shown that most people can
only hold about seven items of information in this store. The so called magic
number 7 (Miller, 1956). This limitation has important consequences for flying
training. If instructions are very long some of the information may well be lost
before the student has an opportunity to attempt a specific procedure.
Two main processes help to overcome this problem of a limited capacity
store. The first, is rehearsal, whereby items are repeated over and over again
until they are fixed in long term memory. This is usually a long and tedious
process. Second, by a process known as encoding, new information once
understood and assimilated, is coded to form new concepts. New information

257
Aviation Instruction and Training

is easier to assimilate and recall once a link has been established between
what the student already understands and the new information. For example,
and this is a trivial example, take the numbers 8250953417 and
975318642975318642. After a little thought it is evident that the second list
is easier to remember than the first despite containing more digits! This is
because it can easily be coded to give - odd numbers then even numbers
starting with 9 - then repeat. Once you have the code you can create the
numbers. Therefore coding is used to simplify a learning task where the
material to be learned has a structure and an individual has the relevant
background knowledge to use that structure. Because of this process,
meaningful material is more easily learned and is a more enjoyable experience.
Information stored in long-term memoiy goes through a series of
transformations (i.e. it is re-coded) as new material is encountered and more
detail is revealed. The student is always using old knowledge as a bridge to
help absorb the new. The student is looking for similarities between what is
already known and what needs to be learned. That is why it is important that
students are not allowed to acquire bad habits because once encoded, these
are difficult to eradicate. Moreover, wrongly coded behaviour will be
misleading when new material is encountered.

The associative phase or fixative stage

During this second stage students essentially know what is required but need
time to consolidate what they have learned by means of rehearsal. This
process gradually establishes the skill in long term memory. During this
phase, errors are extinguished and correct responses established. Performance
is often inconsistent, with good and bad days common. The length of this
phase varies greatly between individuals and between tasks.

The autonomous stage

The final stage in the acquisition process is known as the autonomous stage,
because once an action is initiated it appears to run off on its own without
apparent conscious thought. Unlike the earlier stages of training where one
action acts as the stimulus for the next - set power - select attitude - trim -
deliberate one to one verbal-motor actions are unnecessary. The process by
which this takes place is not fully understood, but what appears to happen is
that whole motor sequences become encoded into single actions. Since
complex tasks now place little or no load on working memory, once this stage
has been reached, performance is more resistant to stress and to interference

258
 Preparing Better Flight Instructors

from other tasks which can now be performed concurrently. Once skilled
motor programs are established they are very resilient - 'you do not forget how
to ride a bicycle!' However, the fine tuning of events is very easily lost and
timing is often the first casualty when skills that are unpracticed.
In summary complex skills have the following characteristics:

- once initiated they appear to run off rapidly and accurately;

- they require minimal conscious effort;
- they take a long time to leam; and
- they are controlled by a central plan and are not dependent on one to
one verbal motor associations.

Implications for instructors

What are the implications of this developmental explanation of skill

acquisition? Is this description only of theoretical importance or are there
practical consequences?
First, research has shown individual differences in learning rates - hour for
hour, not everyone will make the same progress. Moreover, the rate of
learning in the early stages is not necessarily a good predictor of ultimate
levels of performance. Therefore, within the constraints of a fixed length
course, extra time spent consolidating the basics in the early stages may pay
dividends later.
Second, encoding is crucially important in skill acquisition. Encoding is
only effective when the new material can be placed in an already existing and
appropriate context. However, it is not always true that practice makes
perfect. A student can practice a great deal but if he is not paying attention to
the relevant and appropriate aspects of the tasks there will be no improvement
(Schneider 1988). The richer a student’s prior knowledge, the greater the
ability to place new information into an appropriate context. Even before the
student has flown a 'mental model' will be formed of what will happen. The
student may understand the 'effects of controls' and can imagine what will
happen. In the air, the controls might, for example, be found to be more
sensitive than anticipated. However the prior expectations will be modified by
experience. The richer the detail and the relevance of the models that are
formed the greater the learning process because every event encountered will
not come as a total surprise and can thus be more easily absorbed into an
existing plan. The importance of groundwork, briefing and debriefing is
immediately obvious. It is also important to avoid introducing new material

259
Aviation Instruction and Training

into an exercise when it has not been previously briefed and instructors should
explore the deeper understanding of relevant concepts and not accept
stereotyped replies. They should welcome and encourage questions from their
students!
As already described, in the early stages of learning, analysis takes place in
working memory and individual actions are verbally mediated. While some
students might audibly talk themselves through situations, often this
monologue will be internal. Although a student may appear not to be very
busy, he could be working very hard mentally, trying to make sense of what
he is doing and encoding new information. Moreover, on making an error the
student may wish to talk through his actions. This desire to verbalize the
problem may all too easily be interpreted as a student trying to excuse his
apparent carelessness. This may be true on one level: the student may wish
to indicate to the instructor that the error has been recognized. It is also
possible that the student is using it as an opportunity to mentally rehearse the
action in order to sort out precisely where the error lay. Thus this
verbalization can be part of the encoding process. The student’s
verbalizations can also be a useful source of information for fault diagnosis by
the instructor.
It is obvious that in order to instruct, instructors must give guidance to the
student and point out relevant cues and appropriate responses. It is also
probably obvious that as learning takes place, the frequency of these prompts
should be decreased as it is essential that students internalize the relevant
contextual cues so that they form part of the 'central plan’ which control his
actions. Unless the instructor withdraws from cuing the student, skill
acquisition would fail to develop beyond the stage of one to one verbal
associations, i.e. the automatic stage of skilled performance would not be
achieved (ITRU, 1972).
When a skill becomes automatic and no longer involves conscious effort, it
is very difficult to decompose it into its individual subskills. It is very
important that incorrect patterns of behaviour are not acquired. Errors should
be corrected immediately or as soon as possible after they have occurred. It
follows, therefore, that instructors must be consistent in their behaviour. For
example, because of their different operational backgrounds, different
instructors may vary in their handling of emergencies. While these different
approaches may be operationally valid they may well confuse students who
are expected to react differently depending on the instructor. This is
illustrated by a remark quoted by Costello (1984) in which the views of
students to aircrew training were obtained:

260
 Preparing Better Flight Instructors

On the whole training is excellent but I think there is a need for a more
standardised level of instruction from staff. There are still vague areas
where you will do a certain thing or approach the task from a different
angle depending upon which instructor you have.

Similarly, the difficulty of decomposing a skill into its parts may prove equally
difficult for instructors who, having attained a high level of skill, are expected
to be able to break them down into their components in order to teach them.
Since it is only in the last stage of skill acquisition that skills become
resistant to stress and multi-task performance is possible, it is obvious that as
new skills are being developed old skills may suffer. Accordingly, a student’s
performance, which was previously acceptable, might appear to deteriorate
while new material is encoded and incorporated with the old. As
workload increases performance deteriorates not necessarily because the
student does not possess sufficient 'natural ability' to cope, but as a possible
indication that previous learning has not been fully consolidated, so there is no
spare capacity available in working memory.
Finally, because flying and in particular military flying involves a high
workload it is generally believed that training must be conducted under similar
conditions. From the above description of skill acquisition it follows that
great care must be taken in introducing high workload. Because intrinsic
workload will be sufficiently high during the early cognitive and associative
stages of skill acquisition, additional stress should be avoided. Under no
circumstances should the student be artificially loaded with nontask related
demands. This will only disrupt the learning process by interfering with
encoding. Until the instructor is satisfied that the student has reached the
autonomous stage of skilled performance, workload should always be intrinsic
to the tasks being performed.

Motivation and anxiety

So far, only the cognitive aspects of skill acquisition have been discussed.
However, the affective aspects of learning, that is how the students feel about
instruction are just as important. In a report on Royal Navy Observer
Training, Costello (1984) drew some very revealing comments out of students
undergoing training. Here are a sample of comments made by them:

- Generally I would like to fly with the same instructor for more than one
flight in a row so that one could build up a form of rapport.

261
Aviation Instruction and Training

- Before I came to BOC (the Basic Observer Course) I was looking

forward to flying. Now I am pleased when a sortie is cancelled - 1 don’t
enjoy it. I don’t suppose you are meant to, but how do you work up
enthusiasm for something you don’t enjoy?
- ... I think it quite important that the student should not get the feeling that
every ride is a ‘warning ride’. Provision should be for
solo-consolidation sorties - these are particularly useful in building up
self-confidence, relaxation and crew cooperation.
- If a relaxed, correct atmosphere is created the student is more confident
and a better result is obtained. Being in a different building adds to an
‘us and them’ feeling and few members of staff make the effort to come
down to the crew room to just chat about work informally.

Although these comments were from Observer students, (the Royal Navy’s
equivalent of Air Force Navigators and Weapons Systems Operators) I have
heard similar remarks echoed by pilot students. Like all things involving
people, learning involves the total person. What the student feels and believes
about training are just as vital to success in pilot training as such things as
perceptual motor aptitude. Two aspects of this area will be considered:
student motivation and student stress and anxiety.

Extrinsic and intrinsic motivation

Motivation is often thought of as a fixed characteristic of an individual such as

being an extrovert or having brown hair. That is, when students are selected
they must have enough motivation. The implication is that if they are
sufficiently motivated then they will have enough to sustain them through a
difficult training course no matter what happens to them. Naturally, those who
aren't motivated will fall by the wayside. A little reflection on personal
experience suggests that reality is more complicated. Sometimes we feel
enthusiastic about what we are doing sometimes not, some days at work are
good, some bad. What is it that makes us feel positive about what we do?
Why do students want to fly? Individuals offer many possible explanations
such as, Ί have always liked aeroplanes', Ί want to feel the exhilaration of
travelling fast', or, 'To experience the freedom of flying'. Motivation at this
level is often based on an ideal without the benefit of direct experience and as
a consequence may be short lived.
From the point of view of learning it is useful to distinguish between two
sorts of motivation - extrinsic and intrinsic. With extrinsic motivation,
motivation is driven by factors external to the individual. For example, 'if you

262
 Preparing Better Flight Instructors

graduate from college, I will reward you with a car'. With extrinsic motivation
flying is seen as a means to an end such as material gain. With intrinsic
motivation the pay off is the sheer satisfaction of being able to do the activity
itself. From a training point of view the ideal type to cultivate is intrinsic but
the use of extrinsic factors are necessary until the intrinsic motivation
develops and takes over.

Attributing success and failure

Another important aspect of extrinsic and intrinsic motivation relates to how

students explain their level of success or failure. Using concepts from
attribution theory (Weiner et. al., 1971), success or failure to achieve a task
can be attributed to extrinsic or intrinsic factors. Failure to achieve a goal can
have a depressing affect on most people. However, the real effect it has on
motivation depends on how this lack of success is explained.
For example, any feedback on lack of progress is derived from the
instructor’s comments. This will have a tendency to decrease intrinsic
motivation. However, the effect that this has on motivation depends on how
this lack of success is attributed. If a low level of achievement can be
attributed to something that is unlikely to occur again, or something that can
be prevented from happening, then long-term motivation will not suffer. For
example, failure can be attributed to external factors such as:

- My instructor is not very good. All the other students have gone solo. If
I can just change instructors, then I ’ll be okay.
or,
- I just didn’t spend enough time preparing for the exercise. Next time I’ll
put in more effort.

Alternatively, lack of success can be attributed to internal factors:

- I’m no good at instrument flying I haven’t got what it takes,

or,
- I failed because I was overloaded by having to fly and navigate. After
more experience I will be alright.

Two dimensions underlie these different attributions: intrinsic (my

responsibility) vs extrinsic (not my responsibility); and stable (unlikely to
change) vs unstable (amenable to change). This makes four possible kinds of
attribution and reaction following success or failure:

263
Aviation Instruction and Training

1. Unstable-external. The failure is attributed to bad luck in getting a

particular instructor. Luck can be changed and little shame is attributed
to failure. Future motivation is difficult to predict.
2. Unstable-internal. The failure is attributed to lack of effort. Shame is
attached to the failure but the student can be optimistic about the future if
more effort is put into preparation.
3. Stable-external. The failure is attributed to the fact that instrument flying
is difficult There is little shame but little to provide motivation for the
future. This is unlikely to lead to greater intrinsic motivation.
4. Stable-internal. The failure is attributed to lack of ability. This is
unlikely to change. If the student believes this then the outlook for
future motivation is very poor indeed.

Attributions to stable factors reduce expectations of future success and

leads to very poor future motivation. The most damaging attribution of all is
to lack ability. Instructors should avoid situations that would lead students to
make this attribution. Excessive criticism, sarcasm or comparison with other
students after failure are obviously to be avoided.
Similar attributions follow success:

1. Unstable-external. Success is attributed to good luck. This can change

so failure is possible. There is little pride in achievement and intrinsic
motivation is likely to be low.
2. Unstable-internal. Success is attributed to extra effort but the student is
not sure that this can be maintained. There is more pride than following
an unstable-external attribution but future motivation is unclear.
3. Stable-external. Success is attributed to an easy task, future success is
anticipated but there is little pride and therefore little intrinsic motivation.
4. Stable-internal. Success is attributed to ability and achievement. Pride
is felt and future success is likely. Intrinsic motivation is likely to be very
high. Consequently it is not a good idea to abandon an exercise after
failure. If time permits allow the student to return to an activity that has
been mastered in order that confidence may be regained, thus finishing
on a high point.

An examination of the reactions and attributions should reveal that there is

a very close link between the student’s self-esteem and intrinsic motivation.
That is attributing failure to internal factors will lower the student’s self
esteem.

264
 Preparing Better Flight Instructors

The maintenance of motivation is, therefore, a shared responsibility

between the student and the instructor. It is the instructor’s task to create a
learning environment that promotes the development of intrinsic motivation
and self-esteem in his students.

How to promote intrinsic motivation

A little thought about these issues leads to the conclusion that the most
damaging attribution of all, both from the point of view of developing intrinsic
motivation and destroying the self esteem of students, is to criticize students in
terms of their stupidity and lack of natural ability. Excessive criticism,
sarcasm or comparison with the performance of other students after failure are
obviously to be avoided. Performance should be criticized in terms of how
current performance compares with what still needs to be achieved. Clear
aims and objectives against which the student can measure his or her
performance are necessary. Students are always aware of their position in the
class and being bottom is always dispiriting. However, instructors should
avoid using class position as a means of performance measurement. Even the
student who is low in the class order makes some progress. Comparing the
student’s current performance with performance a week ago will demonstrate
that progress has been made even though the relative position in the class has
remained unchanged.

Conditioning and reinforcement

A student's first experience of flying is very important and it is very easy in the
early stages of training for flying to become associated with something
negative such as lack of confidence or something physical, like nausea. This
can happen accidentally and is an obvious danger with students who are
nervous or underconfident. For example, care must be taken if contemplating
a demonstration of aerobatics on a familiarization flight, even if the students
appear keen. Any adverse feelings may colour future expectations. Once
established, this kind of associative learning or conditioning is very difficult to
dispel.
There are other forms of conditioned learning that are beneficial.
Behaviour that is rewarded is more likely to be repeated. One of the most
powerful forms of reward available to the instructor is praise, or social
reinforcement. Social reinforcement can be spoken, such as saying 'well done'
or non-verbal, such as paying close attention to what the student is saying,
nods of agreement and praising glances. Such responses can be more

265
Aviation Instruction and Training

powerful than verbal reinforcement. However, it is important that both verbal

and non-verbal cues are consistent. Saying 'well done' while casually looking
out of the window is not very convincing.
Social reinforcement depends for its success upon the relationship between
the instructor and the student. If the student neither likes nor respects the
instructor then praise from such a source can seem unpleasant or patronising.
It is easy for instructors to forget to give praise as they are often 'ahead of the
aircraft' planning the next task. It is also natural to comment on mistakes
which are, after all, what the instructor wishes to correct and take actions
performed well for granted.
The timing of reinforcement is also very important. It should follow as
close as possible to the praise-worthy action. There is little point on
complimenting a student on a good landing a week after the event. However,
once a correct response has been established, giving reinforcement
intermittently is most effective (Fester and Skinner, 1957). Once a student
knows that the instructor praises good performance then they expect to receive
it. However, if it is not given the student will strive to repeat a good
performance in order to obtain the reward. Most instructors are incredulous
when this is explained however when one realizes that this is the principle that
makes the fruit machine game so successful, they see the point. The instructor
as a fmit machine is a concept to toy with!
Another important factor in promoting intrinsic motivation is the feeling
that the student has some control of what is happening. Therefore motivated
learning is more likely to occur where some form of individual pacing is
possible. However, when operating under time and cost restraints this is not
always possible. A feeling that you are in control and have responsibility is
important. Flying training is regulated by rules and routines in order to ensure
safety and to operate efficiently. However, too close and obtrusive
supervision leads to external-stable attributions to be formed: 'What’s the
point ... it’s all up to them anyway'. From the study described above
(Costello, 1984) the following comments illustrate the point. In describing
instructors who they rated as 'good', the students said:

- During consolidation sorties (he) listens carefully and allows the student
to do things for himself.
- Does not allow the student to continue to ‘dig a hole for himself’ but
does allow the student to ‘get out of the hole’ on his own.

Of instructors rated as 'bad', the following comments were made:

266
 Preparing Better Flight Instructors

- Would take over from the student during the sortie e.g. fill in the log
card.
- Changes navigation aids to ones which differ from those agreed on at
briefing.

Stress and anxiety

Most people will accept that flying is a stressful activity. The stress is created
by the physical environment, temperature, noise, ‘G ’ forces but also by the
psychological concomitants of fear or workload. For the student pilot there
are additional stressors in addition to those created by the unfamiliar
environment of flying. They are those created by fear of failure, the anxiety
that he or she might not succeed in their chosen career or that they are not
keeping up with the rest of the course.
Stress is important for the instructor because it directly affects learning.
Studies have shown that not only does stress reduce efficiency but it can
produce interfering responses which impair both performance and learning of
tasks (Eysenck, 1976). The mechanism by which stress influences
performance is beyond this chapter, nevertheless it is associated with
physiological arousal. The relationship between arousal and performance can
be described as an inverted 'U'. At low levels of arousal, performance is poor
(at an extreme level we are asleep!). As arousal increases, performance
improves until an optimum point is achieved. As arousal further increases
beyond the optimum point, performance begins to deteriorate until an
individual is so overloaded that performance is impossible. On the upward
curve arousal has an energizing effect and on the descending slope it interferes
with performance. Ideally instructors need to keep students at the optimal
level. However, the optimal level varies depending on individual
differences, the nature of the task and other factors. For example, very simple
exercises which are well rehearsed and practiced are less likely to be affected
by arousal than complex exercises requiring a lot of information processing.
Therefore complex tasks require lower levels of arousal while simpler ones
can withstand higher levels.
Anxiety can have a similar effect on performance as does the high
workload generated by a task. This is partly because there is a cognitive
factor associated with anxiety that can be conceived of as involving thought.
An example is saying to oneself, 'I’m going to fail this because I’m stupid’, or
'My instructor thinks I’m stupid'.
The effect of these task irrelevant thoughts is to take up space in working
memory. One of the functions of working memory is the processing and

267
Aviation Instruction and Training

‘temporary’ storage of information, this means that it will also deal with the
processing of task irrelevant information generated by worry. Therefore
during a complex task there will be less processing capacity available. When
this occurs an individual adopts the voluntary strategy of restricting the
information inputs, i.e. fewer of the incoming sources of information are
sampled. In terms of attention the result is that a student’s attention will
become very narrowly focused.
It is, however, possible for individuals to exert extra effort to increase their
information processing capacity, because there is a component of attention,
which corresponds to effort. The factor which determines the amount of effort
exerted by an individual is the evaluation of how difficult the task is. What
happens is that the decrease in efficiency, arising from task irrelevant
processing, is overcome by the extra effort made by the individual to
compensate for the decrement. Anxious individuals can perform as efficiently
as non anxious individuals but only through investing more effort. Under high
task workloads, the effort required to perform the task is already very high.
Under these circumstances any additional load caused by worry might be
sufficient to have an effect on performance. Experimental evidence shows
that this is particularly true where the possibility of failure is involved.
Studies have shown that 'anxiety adversely affects perception, cognitive
functioning and motor responses and are likely to occur whenever the outcome
of a task in which a person is highly motivated appears to be in doubt - that is,
at times of uncertainty, insecurity or impaired confidence'. (Davis, 1964)
There is, therefore, a possibility that the difficulty some students have in
learning to deal with high workloads is exacerbated by the anxiety induced by
the stress involved in learning to fly.
Not all students will suffer from undue anxiety but there are large numbers
of students who are faced with this problem. Instructors should attempt to
reduce or control stress during learning. There is not a great deal that they can
do about the physical environment except perhaps through their actions
demonstrate their confidence in the aircraft and their ability to recover from
any situation that the student may induce!
However, some research has demonstrated that some instructors, far from
reducing stress, can be responsible for increasing it. Krahenbuhl et al. (1980)
found that 'one of the most salient stress producing agents in pilot training is
the instructor pilot'. They conducted a study to determine the level of stress
imposed by pilot instructors during flying training. They found that 'positive
instructors' (those who relied on encouragement, acceptance, praise and
instructional cues) produced less stress on the student pilot, this resulted in
better performance. On the other hand, 'negative instructors' (those who relied

268
 Preparing Better Flight Instructors

on harsher tones, criticism and scolding) significantly raised stress levels in

students, thus inhibiting performance.
What are the implications for instructors? Clearly uncertainty contributes
to student anxiety. Therefore providing students with precise feedback and
knowledge of results will leave them in no doubt about where they stand.
Avoid threats involving future events, such as threatening to place the student
on review, immediately before an exercise. Ensuring common performance
standards between instructors is also important in reducing uncertainty.
Finally, establishing rapport and a good relationship with students and using a
positive instructional style will contribute to an effective learning climate.

Conclusions

There are many dedicated and effective flying instructors who are already
applying sound psychological principles, however, there are some who do not.
It is possible that by including psychological issues, such as those described
above, into pilot instructor training that training effectiveness will be improved
not least by making it a more rewarding experience not only for the students
but also for the instructors.

References

Costello, M. (1984), Observer Instructors: A Report of the Responses of

Observer Trainees and their Instructors to a Questionnaire on Effective and
Ineffective Instructor Behaviour', Unpublished M.O.D. Report, Ministry
of Defence, London.

Davis, P. (1964), quoted by Henley, I. 'Student pilot stress: an emotional

barrier to efficient flight training', in Proceedings o f the Third Symposium
on Aviation Psychology, (1985), Ohio State University Press, Columbus,
Ohio, pp. 541-549.

Eysenck, M.W. (1976), 'Arousal, learning and memory', Psychological

Bulletin, vol. 83, pp. 389-404.

Fester, B.C. and Skinner, B.F. (1957), Schedules of Reinforcement,

Appleton-Century-Crafts.

269
Aviation Instruction and Training

Fitts, P.M. (1964), 'Perceptual-motor skill learning', in Melton, A.W. (ed.),

Categories o f Human Learning, Academic Press, New York.

ITRU (1972), Industrial Training Research Unit, Report No. 3.

Krahenbuhl, G.S. et al. (1980), Undergraduate Pilot Training: Instructor

Pilot Behaviour and Student Stress and Performance, AFHRL-TR-80-15,
Air Force Human Resources Laboratory, Brooks AFB, Texas.

Miller, G.A. (1956), 'The magic number seven plus or minus two-some limits
on our capacity for processing information', Psychological Review, vol. 63,
pp. 81-97.

Mold, E. (1982), 'Bom again pilot', Air Force, vol. 6, no. 2, June.

O’Reilly, S.J.A. and Elshaw, C.C. (1987), Training for high workload',
Research Note 4/87, Unpublished MOD Report, Research Branch, HQ
Royal Air Force Support Command, RAF Brampton, Huntingdon, UK.

Ronden, J.M. (1969), 'Selection and training of fighter-pilots in the Royal

Netherlands Air Force, in Aviation Psychological Research WEAAP,
Report of the 8th Conference for Aviation, SWISSAIR, Zurich, pp. 54-61.

Roscoe, S.N. (1980), Aviation Psychology, The Iowa State University Press,
Ames, Iowa.

Schneider, W. (1985), Training high-performance skills: Fallacies and

guidelines', in Human Factors, vol. 27, no. 3, pp. 285-300.

Telfer, R. (1983), 'Back to basics: suggested solutions to some flight

instruction problems', in Jensen, R.S. (ed.) Proceedings of the Second
Symposium on Aviation Psychology, Columbus, Ohio.

Weiner, B. (ed.) (1974), Achievement Motivation and Attribution Theory,

General Learning Press, Morristown, New Jersey.

270
18 Instructional design
and curriculum
development in
aviation
H.H. Lehrer

The need to educate individuals in an orderly, deliberate, and sustained

manner is becoming more important in aviation settings every day. This
importance is a result of the complex systems and equipment that we plan,
build, operate, and repair. To this end, there is an increased need for more
effective, efficient, and meaningful educational experiences for all who have
responsibilities in the field.
The primary goal of an educational experience, whether it is an airline
transition course or an avionics laboratory exercise, is to transmit or develop
knowledge, skills, concepts, attitudes, or habits. Such experiences or
perhaps more appropriately, the educational events, take place in many
differing aeronautical settings. These settings can be as varied as a ground
school class, an aircraft simulator lesson, an engine maintenance laboratory
exercise, or a cross-country training flight. While each backdrop is
somewhat different, the methodology for the development of any required
lesson plans, course outlines, or instructional designs is very much the same.
For a check airman, a military instructor pilot, or an airline director of
training, there have been few models available for aeronautical curriculum
development. To assist those that find it necessary to prepare aviation
instructional programs, this chapter contains an overview of the concepts of
curriculum development and instructional design, the development of a
matrix that can be used to prepare for educational events in most aviation
settings, and some examples of applying the course development matrix to
different aeronautical activities.
Aviation Instruction and Training

Instructional design and curriculum foundations

Before it is possible to prepare any instructional model or curriculum, it is

first necessary to think of as many methods, devices, or things that can be
considered as models of curriculum. Possibly the first definition that comes
to mind is thinking of curriculum as an educational plan. Such a plan could
simply be the manner in which a learning experience takes place in a formal
classroom activity or a self-study computer-based training module. Also as
correct a definition are a course of study for a pilot ground school program,
a training syllabus or outline for a flight rating, or simply a lesson plan for
teaching someone to do a compression check on an airplane engine. But just
as appropriate are things identified in somewhat more academic terms like
intended learning outcomes, modes of thought, an organization of
knowledge, or a cognitive experience. There are some educational planners
that refer to curriculum simply as the matter, the manner, and the outcomes.
It is possible then to see that there is not one concrete definition available
because all of the above are just as valid as stating that curriculum is an
educational plan.
Perhaps an initial focus on curriculum development can be gained by
reviewing the work of Ralph Tyler (1949) in Principles of Curriculum and
Instruction. Tyler, considered by many to be the foremost curriculum
theorist of this century, posed four questions over 50 years ago. Tyler's
questions are:

1. What educational purposes should the school seek to attain?

2. What educational experiences can be provided that are likely to attain
these purposes?
3. How can these educational experiences be effectively organized?
4. How can we determine whether these purposes are being attained?

While these questions clearly relate to a school setting, could not the first
question be rephrased to ask, 'What educational purposes should the aviation
maintenance training program seek to attain?' The second question might
now become, 'What educational experiences will provide the best trained B-
747 first officers?' The third question seems to stand on its own while the
last question could stand alone or be logically structured to reflect the
concerns of a government regulatory agency such as 'How can the FAA
determine that the pilot candidate has successfully completed each phase of
the training syllabus?' Although we will spend a little more time with these

272
 Instructional Design and Curriculum Development in Aviation

questions later, there appears to be one component missing in Tyler's model,

the question of assessment of needs.
The issue of needs assessment is one that deals with, as an initial step,
who should receive specific training or education. In addition, needs
assessment as it relates to curriculum might be interested in why training is
important. Hilda Taba (1962), in Curriculum Development: Theory and
Practice, extended Tyler's four questions somewhat in the development of a
seven-step model. The seven steps include:

1. Diagnosis of needs.
2. Formulation of objectives.
3. Selection of content.
4. Organization of content.
5. Selection of learning experiences.
6. Organization of the learning experiences.
7. Determination of what to evaluate and of the ways and means of
doing it.

Probably the most significant difference between the Taba offering above
and the Tyler model is the focus on a diagnosis of needs as the initial step of
any curriculum development. Beyond that difference then, most of Taba's
steps seem to be extensions of Tyler’s model.
Current curriculum development thought seems to lean toward thinking of
the process as one of instructional design. Within the instructional design
arena there is both a process and a plan. The instructional design process
asks the questions:

- For whom is the program being developed?

- What do you want the learners or trainees to be able to do?
- How is the subject content or skill best learned? and
- How do you determine the extent to which learning has been achieved?

The instructional design plan, a very comprehensive approach to

developing a training or instructional package, is a bit more complex and is
composed of ten steps. To utilize this method, the instructional designer
should:

1. Assess learning needs.

2. Select topics or tasks.
3. Examine learner characteristics.

273
Aviation Instruction and Training

4. Identify content and tasks.

5. State learning objectives.
6. Design learning activities.
7. Select resources.
8. Identify support systems.
9. Evaluate learning.
10. Pretest learners.

Kemp's (1985) The Instructional Design Process offers an extensive

review of the instructional design plan above but it should be noted that
Kemp suggests that each person using this plan may want to arrange the
steps in a different order. It may be that different users of such a plan might
be more flexible and add or delete certain steps to account for unique local
variables.

Aviation curriculum design matrix (ACDM)

Now that we have reviewed several models for curriculum and training
development, is there a somewhat universal instructional design
methodology that would work well in most aviation settings? The answer is
probably yes. By combining the most generic components of Tyler, Taba,
and the instructional design process and plan (and considering the work of
others not already mentioned but who have many similar ideas to offer) we
can develop a five-step model that will fit most situations. Let's call it 'that
which forms or determines', or a matrix. The five steps in the Aviation
Curriculum Design Matrix will be:

1. What is the assessment of needs for the educational experience that

will be offered?
2. What is the purpose of the educational experience?
3. What experiences are needed to ensure a meaningful educational
event?
4. How should the experiences that make up the curriculum be
organized and presented?
5. How can the effectiveness of the curriculum be evaluated?

274
 Instructional Design and Curriculum Development in Aviation

What is the assessment o f needs fo r the educational experience that will be

offered?

An assessment of needs is a necessary first step in any aviation curriculum

development in that several closely related questions must be answered.
The first question should be, 'Who is this curriculum intended for and is
there truly a need for such an instructional training program or educational
event?' By asking this question, you are developing a focus for identifying
your potential population. We are not talking numbers yet, but merely
attempting to determine the identity of our target trainee or subject pool.
A logical extension of the needs assessment question most certainly must
include an attempt to find out the size of the population. You might have to
ask yourself, 'Are there a sufficient number of potential students available to
make the class or workshop that I want to offer a success?' If you want to
begin a certain training program at your flight school, you must attempt to
determine whether there is a sufficient number of potential candidates
available and attempt to determine whether they are interested in your
offering. Should the answer be yes, you might decide to proceed; if there is
not a sufficient participant pool, you might not proceed further.
Although the determination of your sample population may be a
marketing problem that is beyond the scope of this chapter, you can decide
at this point to take either an informal approach, or choose to develop a full-
scale target marketing program. If you choose to stay on the informal side,
valuable information can be obtained by simple opinion surveys either sent
to a mailing list available to you already or by purchasing a database mailing
list from a professional aviation organization such as the University Aviation
Association, the Aviation Technical Education Council, or the Professional
Aviation Maintenance Association. If you want to canvas a larger
population base, a marketing firm specializing in such a search might be
your most logical approach to determining a need for your educational
offering.

What is the purpose o f the educational experience?

One way to think of the purpose of an educational event may be focus on

'why' is this experience necessary. In aviation, the primary driving force in
defining why the content of a specific learning event is developed may well
be government regulation. The need for such government definition of
educational requirements for various aviation ratings from air traffic
controller (ATC), pilots, ground instructors, mechanics, and including any

275
Aviation Instruction and Training

other aviation professional is quite evident. We all probably reluctantly

agree that there must be some ultimate controlling agency. But there are
several other entities in the aviation community that influence the need to
define the purpose of an educational event.
Saylor, Alexander and Lewis (1981) in Curriculum Planning fo r Better
Teaching and Learning state that curriculum planning decisions are shaped
by clients, critics, professionals, legislative groups, and finally, the courts.
Although instructors and training entities have the initial responsibility for
defining the purpose of an educational event, the impact of these five groups
can in differing ways effect any final decisions greatly.
The impact of the needs of clients is seen when an analysis is made as to
why someone comes to an aeronautical school for a flight or ground rating.
The individual is there for the appropriate schooling and experiences. A
logical extension of the same client question would indicate that an
experienced pilot enrolling in a type rating course at a company such as
Flight Safety or Simulfite is coming to that organization with the expectation
of receiving the required training.
The critics also have considerable influence on educational and training
matters. This influence can be clearly seen immediately after some
unfortunate aviation accident or incident. The public, many times fired by
the accounts of the press concerning such mishaps, is quick to call for a
complete investigation and possible redefinition of the educational purpose
of aviation schools, ATC training academies, and air carrier training
programs.
The influence of the professionals, particularly the unions, on the purpose
of training curriculum is an important consideration in aviation. The pilots,
mechanics, controllers, technicians, and all others that are working in the
field each day are well positioned to see the need for a redefinition of the
educational purpose for a training curriculum. Many of the ideas that have
come from the professionals include those which have increased safety in
the total aviation environment, enhanced job productivity, and enriched the
work experience. Included among the professionals are the aviation interest
groups as well.
Finally, while the legislative and legal system may not directly affect the
educational process as much, they certainly can influence matters
tangentially. An example of such impact can be seem in the findings and
conclusions of the National Transportation and Safety Board (NTSB). It is
not uncommon for NTSB investigations to provide the catalyst for industry­
wide reform with respect to aviation training.

276
 Instructional Design and Curriculum Development in Aviation

What experiences are needed to ensure a meaningful educational event?

When considering the planning of the experiences of an educational event,

this dimension could be very large and include goals, objectives, tasks,
lessons, units, periods, or any number of similar phrases. So that this
discussion can be delimited somewhat, the experiences that will be
considered in this section will be confined to developing learning objectives,
or, 'what should someone be able to do after a learning session has taken
place?' But prior to determining the shape and form of a learning objective,
we must ascertain which domain of learning represents the focus of the
curriculum. There are three domains of learning that encompass most
aeronautical instructional programs, the cognitive which represents
knowledge, the affective which includes values and feelings, and the
psychomotor which is skills and their development.
Within each domain of learning, as indicated in Table 18.1, there are
several categories of learning development from the simplest to the most
complex. In the cognitive domain, the categories include knowledge,
comprehension, application, analysis, synthesis, and evaluation. The
affective domain has receiving, responding, valuing, organization, and value
complex as major categories while the psychomotor domain categories are
perception, set, guided response, mechanism, complex overt response,
adaption, and origination.

Table 18.1
Taxonomy of domains and categories of learning

Cognitive Affective Psychomotor

Simplest Knowledge Receiving Perception

Comprehension Responding Set
Application Valuing Guided response
Analysis Organization Mechanism
Synthesis Value complex Complex overt
response
Evaluation Adaptation

Most Complex Origination

277
Aviation Instruction and Training

More complete references for the cognitive domain may be found in

Bloom's (1956) Taxonomy o f Educational Objectives: Handbook I,
Cognitive Domain. For the affective domain, refer to the Taxonomy o f
Educational Objectives: Handbook II, Affective Domain by Krathwohl
(1964), and for the psychomotor domain read Simpson's (1972) The
Classification o f Educational Objectives in the Psychomotor Domain.
Within each domain's several categories are illustrative verbs which
portray the specific action necessary to define activity in that category. In
some cases, more than a dozen verbs are present, while in others six or
seven suffice. A complete list of verbs for each domain and category is
contained in the Appendix of Gronlund and Linn (1990), Measurement and
Evaluation in Teaching. To illustrate, in the cognitive domain evaluation
category, the action verbs include: appraises, compares, concludes,
contrasts, criticizes, describes, discriminates, explains, interprets, justifies,
relates, summarizes, and supports. In the psychomotor domain origination
category, the action verbs are: arranges, combines, composes, constructs,
creates, designs, and originates. The verbs that are associated with each
category are very useful in preparing performance objectives for
instructional programs and it should be noted that the preparation such
objectives is a fundamental component of any instructional sequences.
A performance objective (be it called a learning or behavioral objective or
an intended learning outcome) is a statement that specifies the behavior a
learner is to exhibit after a learning episode (lesson or series of lessons). An
objective usually details the conditions under which the learning is to occur
and the level of performance expected. Most performance objectives take
the form of: At the completion of the course (lesson, task, or syllabus), the
trainee should be able to do a specific thing under a certain regime and to
such a level of performance.
To write a performance objective, the first step is to determine in which
domain of learning the activity is centered. The second step is to determine
the major category within each domain that indicates the level of
performance expected for the objective. Once the level has been
determined, the third step is to select the action verb that best exemplifies
the type of activity the learner will pursue. Finally, the action verb is
combined with criteria that can be measured and the condition(s) that relate
to the specific activity (sometimes the condition and criteria can be implied).
There will be examples of preparing performance objectives in the 'ACDM
in Action' section below.

278
 Instructional Design and Curriculum Development in Aviation

How should the experiences that make up the curriculum be organized and
presented?

There will be two major points discussed in this section. The first will be in
what format to present the learning session and the second will be what
audio-visual medium will be used to reinforce as well as enhance the
learning experience.

Different methods o f presentation. When considering a presentation format,

there are several different choices, many of which are probably familiar as
they are contained in the FAA's Aviation Instructor's Handbook (1977).
These different educational presentation methods include the:

- Lecture
- Teaching interview
- Guided discussion
- Case study
- Performance-demonstration
- Computer-based training

The lecture method is probably the most prevalent and accepted form of
presentation for ground school classes. Speaking from a prepared lesson
plan or notes, the instructor delivers the class material with little interaction,
if any, from the students. An interesting variation on the lecture is the
teaching interview. During the teaching interview, a visiting expert, invited
to share his or her experiences and expertise with the class, answers
question posed by the instructor as well as the students. The instructor acts
as the interviewer and the class may get an opportunity to engage in the
dialogue. Educators that have used this technique report excellent results.
The guided discussion is somewhat similar to a combination of the
lecture and the teaching interview. Specific questions are posed by the
instructor, who acts as the expert, but the students supply the answers. To
maintain continuity through out the session, the instructor must be sure to act
as the discussion leader and has a primary responsibility to guide the
dialogue and, in conclusion, summarize the main points.
The case study method, traditionally used in university graduate schools,
could have some interesting applications in aviation settings. A case study
is based on developing a hypothetical situation which can possibly be solved
in several different ways. The students, after being given the details of the
scenario, may be required to work either individually or as small groups and

279
Aviation Instruction and Training

are asked to reach conclusions based on several questions posed by the

instructor. Case studies can be in the form of an in-class project or as
homework assignments and can be either oral or written. The discussion
that completes this presentation technique calls for the instructor to
summarize and critique the points made by the students.
The performance-demonstration has always been a foundation of the
aviation training regimen. This traditional approach to instruction is very
appropriate when teaching something psychomotor in content. The
performance-demonstration centers on the instructor presenting the correct
procedure for a flight maneuver, instrument approach procedure, or
maintenance operation followed by the trainee executing the same followed
by repetitions until a certain level of competence is achieved.
A final method of presentation that is becoming extremely important in all
fields, and aviation is no exception, is the use of computers in instructional
situations. Included among the things that a pilot can now easily do with a
computer are flight planning, weather briefings, test preparation, flight
simulation, as well as maintain a log book. All that is required to use most
of the sophisticated but very user-friendly software that is available from
numerous vendors is a small personal computer. Although most users will
not be doing any computer programming, it is quite easy to see that off-the-
shelf aviation computer-based training packages can be easily integrated into
an exciting instructional format.
This section has outlined only six methods of delivering an instructional
program; there are many other formats available. Although many of the
following methods have not found general acceptance in aviation, you may
want to consider as possible instructional strategies the use of debate, a
panel discussion, a colloquium, readings, listening groups, buzz groups, role
playing, group problem-solving, the socratic method, or games. Other
options include workshops, conferences, as well as a symposium or forum.
For a more complete discussion of these instructional techniques, consult
Knowles'(1980) Modern Practice o f Adult Education - From Pedagogy to
Andragogy.

Media resources. The use of media not only reinforces the presentation
method that you have selected but also adds a component of interest that can
capture a student's attention. It is important to realize also that the
aeronautical audience that you are teaching is becoming quite sophisticated
as to the quality of media that they expect. Twenty or thirty years ago, a
chalkboard could be used for a large proportion of the teaching that was
done but now the more colorful, animated, and active the presentation, the

280
 Instructional Design and Curriculum Development in Aviation

better. There are many kinds of instructional resources and they can be
divided into five distinct groups. These groups include:

1. Real things: guest speakers; objects and devices; and models and
mock-ups.
2. Non-projected material: printed paper copies; chalkboard; diagrams,
charts, and graphs; photographs; and procedures lists.
3. Audio recordings: audio-cassettes and disc recordings.
4. Projected still pictures: slides; filmstrips; overhead transparencies; and
computer programs.
5. Projected motion pictures: films and video recordings.

For years, educators have pondered the question of which type of media is
preferred in a given teaching situation. Unfortunately, there is no conclusive
data indicating that learning takes place any better or any worse with respect
to a specific medium being utilized or not. However, there may be more
utilitarian reasons to select a certain presentation aid. Kemp (1985) has
outlined several questions that may, when answered, be very helpful in
assisting one to make a more informed decision. The media selection
questions are:

- What is readily available?

- What are you most familiar with and comfortable in using?
- What is the most objective reason for selection?
- What are the costs?
- How much time is there to prepare?
- Do the learners prefer a specific type of media?
- Do any scheduling problems result from a specific media choice?
- Is one medium more suitable regardless of any other constraints?

One final note is appropriate to organizing and presenting the curriculum

and that is that a lesson plan is an integral part of the instructional package.
A lesson plan serves as a road map or guide and is a tremendous confidence
builder for the less than secure. It is not so important that your lesson plan
follows any one format but rather that you use an outline of some form.

How can the effectiveness o f the curriculum be evaluated?

The final component of the ACDM is the evaluation of the curriculum or the
learning experience. Even though a very complete overview of evaluation in

281
Aviation Instruction and Training

aviation is contained in The Psychology o f Flight Training by Telfer and

Biggs (1988), it is appropriate to review several main points concerning the
evaluation process. Although most individuals probably tend to believe that
evaluation is used solely to determine a subject's grade, evaluation has other
purposes. Among these purposes are to assist in making course
improvement decisions and to facilitate the administrative regulation of the
instructional program. The evaluation that forms the basis for all these kinds
of decisions fall into three specific categories: an initial evaluation such as
an aptitude or interest test; formative evaluation might be represented by a
quiz, unit test, or stage examination; and finally, a summative evaluation
that could include an FAA written examination or simply a final examination
for a ground school course. No matter which of the three forms of
evaluation is taking place, there are several different types of tests available.
Table 18.2 contains contrasting test types and their characteristics.
Probably the most familiar type in aviation are the supply versus selection
type - essay or multiple choice and true/false respectively. The selection
type of test has been the backbone of the aeronautical testing regime and is
likely to remain so for quite some time. This assumption is based on the fact
that selection type tests are objective, easy to give and grade, provide
readily understood statistical data, and can be very comprehensive. It might
be interesting for aeronautical tests to begin to include more subjective test
measures such as the essay, fill-in-the-blanks, or short answer. Allowing a
test subject to offer a more complete explanation and not be restricted to
merely marking an answer could enhance the evaluation process greatly.
Use of such a testing method will require more instructor effort in preparing,
administering, and grading examinations.
The preparation of test questions and the statistical analysis of exams are
major subjects in themselves and beyond the scope of this chapter.
However, if you want information on these subjects, the FAA Aviation
Instructor's Handbook (1977), Telfer and Biggs (1988), as well as any test
and measurement text book such as Measurement and Evaluation in
Teaching by Gronlund and Linn (1990) will provide a great deal of useful
guidance. In addition, the availability of computer test bank databases,
software containing examination questions based on a specific textbook, is
becoming more available in many academic fields; unfortunately, aviation is
not one of them. If you already do, or decide to prepare your own test
questions, there is software available from many educational vendors. Use
this software to make your own test bank, then you will be able to prepare
multiple test forms for virtually any audience.

282
 Instructional Design and Curriculum Development in Aviation

Table 18.2
Contrasting test types and test characteristics

Test Types Test Characteristics

Informal versus Informal are constructed by individuals whereas

Standardized test specialists construct standardized.

Individual versus Individual are on a one-to-one basis whereas others

Group involve group testing.

Mastery versus Mastery covers a limited set of outcomes whereas

Survey survey measures a broad range.

Supply versus The examinee supplies the answer whereas others

Selection require selection of the correct response from given
choices.

Speed versus Speed indicates the number of items completed in a

Power given time whereas power measures performance
with ample time.

Objective versus Objective indicates that all scorers will obtain the
Subjective same score whereas scorer opinion influences
subjective.

Verbal versus Verbal requires a verbal response whereas

Performance performance requires a motor response.

Concerning the use of evaluation, namely test scores, to facilitate course

improvement or to make any administrative decisions requires a few general
comments. First, it is important to realize that most classes or groups of
learners that appear to be similar are often quite different; in fact, seemingly
insignificant things like the time at which a test is given or temperature of
the room can have a profound effect on the outcome. Considering these
affects, you may have to temper any conclusions that are drawn about the

283
Aviation Instruction and Training

results. Secondly, many of the same considerations about individual

differences must be weighed when making an administrative decision about
the effectiveness of one method of teaching or when comparing the
examination scores by the students of a specific instructor with the scores of
another instructor's students. Finally, examination scores should never be
used in a punitive manner either toward the students or the instructor. What
you really have when considering test data is a measurement of something
that happened at a specific time to a specific group or individual and little
else.

ACDM in action

The following section is an application of the ACDM using two different

aviation settings. One setting is the preparation of a flight and ground
training course and the other is the development of a course that would be
part of an avionics technician repairmen certification program. In each of
these situations, the five questions posed in the ACDM are answered. The
answers to the questions will be brief but should be considered as only
illustrative of the use of the matrix to define the basic requirement of a
curriculum in that specific area. In an actual application of the ACDM, the
answers would probably be very extensive. The focus, with respect to
government regulation, will be toward a United States program so the FAA
will be used as applicable.

A flight and ground training scenario

In this scenario, you want to develop a flight and ground training course to
prepare currently certified flight instructors for the addition of a multi-engine
airplane instructor rating to their certificates. You are the chief flight
instructor and the location of your flight school is in a medium-sized
metropolitan area.

What is the assessment o f needs fo r the educational experience that will be

offered? For an instructor to give flight instruction leading to the issuance of
a multi-engine rating, he or she will have to be appropriately rated. Training
is required, but you must first determine if a sufficient number of current or
potential instructors that wish to add this rating are in your area. You may
have a current of list of possible candidates already or it might be helpful to
contact the local office of the FAA concerning access to a mailing list. Your

284
 Instructional Design and Curriculum Development in Aviation

metropolitan area may be large enough so that a mail survey and flyer or a
newspaper advertisement about your course offering may yield several
potential enrollees. If you have grander aspirations for you course, you
might want to consider advertising in one of the national trade magazines.

What is the purpose o f the educational experience? The graduates of your

training program must be able to perform, in excess of the minimum level,
all the required maneuvers and operations as outlined in the appropriate
flight test standards publication. You must also prepare your graduates to
instruct others that seek a multi-engine rating. Although the potential course
enrollees are already instructors you should review the development of
instructional training courses with an emphasis on multi-engine techniques.

What experiences are needed to ensure a meaningful educational event?

The pertinent flight test guide for this additional flight instructor rating is
very clear on the performance objectives and the completion standards.
However, there are probably several additional performance objectives to
include. You will recall, as stated earlier, performance objectives consist of
an action verb combined with criteria and condition.
Since individuals that graduate from your course will have to evaluate
students, consider using an action verb from the cognitive domain -
evaluation category to write a performance objective to reflect this
responsibility. These verbs include appraises, compares, concludes,
contrasts, criticizes, describes, discriminates, explains, interprets, justifies,
relates, summarizes, and supports. The performance objectives you prepare,
for example, might include one that states: 'At the completion of the course,
the multi-engine instructor candidate will be able to, with no errors, explain
and support with appropriate chalkboard diagrams, the correct methodology
for teaching Vmc to primary multi-engine students.' Notice the use of two
verbs from the category and there is criteria and a condition. Remember
though that the criteria and condition can be implied.
There will be other performance objectives appropriate for the flight
training segment of the course that involve the psychomotor domain.
Consider the verbs from the psychomotor domain set category: begins,
displays, explains, moves, proceeds, reacts, responds, shows, starts, and
volunteers. A performance objective in this domain and category might be:
'At the completion of the course, the multi-engine instructor candidate will
be able to react correctly every time when asked to establish, merely by
touch, all crossfeed configurations appropriate for that airplane.' Again, a
verb combined with condition and criteria.

285
Aviation Instruction and Training

How should the experiences that make up the curriculum be organized and
presented? The traditional way in which flight and ground instruction has
been given, namely the performance-demonstration and the lecture and/or
guided discussion, will probably suffice for this training course. However,
the use of media could enhance the ground training lessons greatly.
Consider combining an evaluation instrument with the use some of the
commercially available videotapes on multi-engine techniques. Such a
strategy could form the basis for students completing a self-paced home
study portion of their ground training, or you may use such videotapes as
part of your ground training lecture.

How can the effectiveness o f the curriculum be evaluated? Evaluation in

the form of objective testing is probably a must for this training course. In
addition, the need for subjective measures, either in the form of short answer
or essay, seems to lend itself nicely to preparing your graduates for the
rigors of the oral/practical portion of their checkride with an FAA examiner.

An AVIONICS laboratory experience

The second application of the ACDM will be preparation of an avionics

laboratory experience. This laboratory segment, an aircraft transponder
testing and troubleshooting lesson, is from the Avionics Repairman Course
(ARC) by Travis (1992) and is part of a proposed curriculum to develop a
new avionics repairman certification course.

What is the assessment o f needs fo r the educational experience that will be

offered? This course, as well as the entire ARC curriculum, is an effort to
better prepare avionics as well as airframe and powerplant mechanics to
respond to the rapid advances in technology with respect to aircraft
electronic equipment repair. In fact, a related issue is that a joint
industry/FAA group would like to have mechanics become aircraft
maintenance technicians, and repairmen would become aircraft repair
specialists. So if new ratings emerge, there is a vast pool of possible
candidates, particularly with respect to recertification, available for
additional schooling.

What is the purpose o f the educational experience? To focus specifically

on the transponder lesson, the goals of this unit require the student to apply
electronics and avionics knowledge in the hands-on troubleshooting and

286
 Instructional Design and Curriculum Development in Aviation

testing of transponders. During the course, the student will also learn to
remove and reinstall the equipment in a safe and professional manner.
Additionally, all attempts will be made to duplicate actual avionics shop and
flight line environment.

What experiences are needed to ensure a meaningful educational event?

Since there will be FAA certification of this course, the preparation of
performance objectives will be mandatory. The domains of the objectives
will include not only the cognitive and psychomotor but the affective as
well. The need to write performance objectives for the affective domain is a
result of the stated requirement within the curriculum for safety and
professionalism.
Writing performance objectives in the affective domain is challenging.
This challenge is because the measurement of the achievement of values is
quite subjective in nature. Considering the affective domain - value
complex, the verbs are: acts, discriminates, displays, influences, listens,
modifies, performs, practices, proposes, qualifies, questions, revises, serves,
solves, uses, and verifies. Consequently, a performance objective related to
this domain and category might be: 'At the completion of the lesson, the
student will display and practice the highest standards of safety in not only
the shop environment but on the flight line as well.' Such behavior, although
highly prized in any situation, is quite difficult to assess.
In the psychomotor domain complex overt response, it is easy to see that
the construction and the measurement of performance is much easier. Verbs
from that category include: assemble, build, calibrate, construct, dismantle,
display, dissect, fasten, fix, grind, heat, manipulate, measure, mend, mix,
organize, and sketch. These verbs lend themselves easily to objective
preparation.

How should the experiences that make up the curriculum be organized and
presented? The use of the lecture method of presentation for academic
theory combined with a hands-on laboratory exercise for practice has always
been an effective means of delivering an instructional package such as this
transponder repair lesson. Travis suggests that the use of several self-paced
instructional modules, either in written or as an interactive computer-based
training form, may be appropriate for the classroom component. The
laboratory experience then could be split into both a formal laboratory
component as well as a repair station environment.

287
Aviation Instruction and Training

How can the effectiveness o f the curriculum be evaluated? There are two
distinct segments of the transponder course. Each segment needs to be
measured in different ways. The evaluation of the classroom theory section
could be measured against a criterion reference using a standard norm of
performance. This kind of measurement uses a specific standard such as a
given numerical score. In the laboratory, the evaluation would tend to be
focused on the performance of the student in troubleshooting and repair as
demonstrated through oral and practical test measures.

Conclusions

There are many ways to go about developing instructional programs and the
ACDM is but one. The ACDM is an initial attempt at constructing a model
to serve as a useful tool for those in aviation that must develop curriculum.
If the ACDM is not right for you, feel free to modify the matrix to fit your
needs. You may decide to organize an instructional sequence quite
differently. If this is the direction in which you would like to proceed, there
are several excellent textbooks contained in the reference list that might
serve as primary resources for you.

References and further reading

Bloom, B.S. (1956), Taxonomy o f Educational Objectives: Handbook I,

Cognitive Domain, McKay, New York.

Bowsher, J. (1989). Educating America: Lessons Learned in the

Nation's Corporations, John Wiley, New York.

Brookfield, S.D. (1986), Understanding and Facilitating Adult Learning,

Jossey-Bass, San Francisco.

Federal Aviation Administration (1977), Aviation Instructor's Handbook

(AC 60-14), US Department of Transportation, Washington.

Gronlund, N. and Linn, R. (1990), Measurement and Evaluation in

Teaching (6th ed.), Macmillan, New York.

288
 Instructional Design and Curriculum Development in Aviation

Hass, G. (1987), Curriculum Planning: A New Approach (5th ed.),

Allyn and Bacon, Boston.

Kemp, J. (1985), The Instructional Design Process, Harper and Row,

New York.

Knowles, M. (1980), Modem Practice o f Adult Education - From

Pedagogy to Andragogy, Cambridge, New York.

Krathwohl, D.R. (1964), Taxonomy o f Educational Objectives:

Handbook II, Affective Domain, McKay, New York.

Marzano, R., Brandt, R., Hughes, C., Jones. B., Presseisen, B., Rankin, S.
and Suhor, C. (1988), Dimensions o f Thinking: A Framework fo r
Curriculum and Instruction, ASCD, Alexandria.

McNeil, J. (1985), Curriculum: A Comprehensive Introduction (3rded.),

Little, Brown and Company, Boston.

Posner, G. and Rudnitsky, A. (1982), Course Design - A Guide to

Curriculum Development for Teachers (2nded.), Longman, New York.

Saylor, J., Alexander, W. and Lewis, A. (1981), Curriculum Planning

fo r Better Teaching and Learning (4th ed.), Holt, Reinhardt and
Winston, New York.

Simpson, E.J. (1972), 'The classification of educational objectives in the

psychomotor domain', The Psychomotor Domain: Vol. 3, Gryphon
House, New York.

Taba, H. (1962), Curriculum Development: Theory and Practice,

Harcourt, Brace and World, New York.

Telfer, R. and Biggs, J. (1988), The Psychology o f Flight Training,

Iowa State, Ames.

Travis, G.W. (1992), Avionics Repairmen Course, Unpublished

manuscript, Embry-Riddle Aeronautical University, Daytona Beach.

289
Aviation Instruction and Training

Tyler, R. (1949), Principles of Curriculum and Instruction, Chicago

Press, Chicago.

290
19 Aviation medicine and
flight instruction
L. Lambeth

Once the decision to learn to fly has been made, the would-be aviator's first
official contact on the road to becoming licensed is with a doctor. The
student is made aware that certain standards of medical fitness must be
achieved even prior to the issue of a student pilot's licence. The student also
finds that not all medical practitioners can issue a medical certificate, rather,
it must be issued by one especially authorized or designated by the licensing
authority. The more astute student will realize that there must be some
special connection between aviation and medicine. The even more astute will
begin to consider the complex inter-relationship between the environment
(air), the pilot (man) and the machine (ship), and will thus have taken the first
tentative step on the road to the pilot's most sought after and desirable quality
- airmanship.
Once the medical certificate is issued, the student will, in the normal
course of events, have little further contact with the doctor save for the
statutory regular review. The aims of the original medical examination and
subsequent reviews are to exclude the presence of any incapacitating medical
condition and to predict the absence of such conditions for the period during
which the licence and medical certification is valid. The philosophy behind
this is the general mission of civil aviation authorities which is, 'to promote
aviation while assuring safety in the air and on the ground' (Norwood and
Jordan, 1982).
The first requirement for the issue of a pilot's licence is medical fitness.
Any pilot - student, private, professional or military, is subject to grounding
for medical reasons. Most pilots have a fear of medical contact and will do
their best to avoid it, sometimes with unwanted or even disastrous results. In
this chapter the focus will be on the role of the doctor in aviation, and the
relevance of aviation medicine to flight training.
Aviation Instruction and Training

The beginnings

Aviation medicine had its beginnings with the early balloonists, many of
whom were medical practitioners. Pilatre De Rozier made the first successful
manned balloon ascent in 1783. He was a surgeon, who later developed a
breathing apparatus which was a forerunner of the modern pilot's oxygen
mask. A French physician, Paul Bert (recognized as the 'Father of Aviation
Medicine') invented the altitude chamber and demonstrated the effects of the
decreased oxygen partial pressure and the beneficial effects of breathing
oxygen at altitude.
It was World War I however, which brought home to the aviation
community in general, the connections between medicine and flying. Early in
the war, the rate of aircraft losses due to human failure attributable to aircrew
was in the order of 90 per cent. All too often, these losses were the result of
hypoxia, or physical defects. This provided a stimulus for major research and
developments in aviation medicine, particularly in determining selection
standards for aircrew, improved oxygen delivery systems, and training for
flight surgeons.
Between the wars, a school for flight surgeons was established in the
United States, by Dr Louis Bauer, who published a textbook on aviation
medicine, and later became the first Medical Director of Aeronautics at the
Aeronautics branch of the Department of Commerce. Civil aviation medicine
was beginning. Medical standards were set for private, industrial and
transport pilots. Aircraft were flying higher, faster and for longer periods of
time. They were far outstripping the capabilities of those who flew them and
flew in them. Aviation medicine had to keep pace.
The pace increased with the advent of World War Π, and many
improvements were made in oxygen delivery, pressurization, the development
of protective clothing, pilot selection and medical training. By war's end, jet
aircraft brought new problems in 'G' tolerance, and soon after, the
development of jet transport aircraft spurred on the development of pressure
systems and oxygen delivery for civil airliners.
Along with this went improvements in the training of flight surgeons for the
military, medical examiners for licensing authorities, and airline medical
officers. Towards the end of the war the Convention of International Civil
Aviation was signed in Chicago. This included several articles which called
for the adoption of international regulations in all fields where uniformity
facilitates and improves air navigation. Standards and Recommended
Practices for Personnel Licensing were first adopted in April, 1948, pursuant
to Article 37 of the Chicago Convention. These SARPS have been adapted

292
 Aviation Medicine and Flight Instruction

by the International Civil Aviation Organization as the minimum standards for

personnel licensing.
The student pilot therefore, is confronted by a system in which medicine
and aviation are inextricably interwoven, and have been dependent on each
other for many advances. On top of this, the student is faced with a complex,
internationally recognized (but not uniformly applied) system of licensing
which is often perceived as being hostile to his or her ambitions to fly. The
validity of this perception requires challenging, and the role of the doctor, or
'flight surgeon' needs defining.

Civil aviation medicine

The flight surgeon's role in civil aviation is a complex one due to three
factors. Firstly, the diversity of operations, ranging from piloting a passenger
carrying jet airliner, to air charter operations, to flight instruction, to sport
aviation, to private flying. Secondly, the variety of flying machines, from
complex supersonic jets to commuter aircraft, to training aircraft to balloons,
and rotary wing aircraft. The third factor is the pilot, coming from differing
backgrounds and bringing diverse sets of training, experience and education.
Medical input may be of two types. The doctor may simply carry out the
statutory medical examinations and reviews, make the appropriate
recommendations, and issue (where applicable) medical certificates. On the
other hand, the doctor may act as a flight surgeon, examining, advising,
educating and counselling aircrew from student pilot level to airlines transport
level. The doctor may be a full-time specialist in aviation medicine, or may
combine aviation medicine with normal clinical practice. Doctors employed
full-time by the airlines usually specialize in the field and carry out a wide
variety of tasks ranging from examination of aircrew to advising on methods
of dealing with passenger incapacitation, to instruction on the prevention of
toxicological and environmental hazards in airline operations. Advice on the
avoidance of, and dealing with, fatigue and stress is a major part of the
doctors’ role at all levels in the aviation industry.
The majority of doctors involved in aviation medicine will have a part time
commitment. Such doctors may have formal academic qualifications
(Masters degree or Diploma in Aviation Medicine), or may have just
sufficient training (sometimes none), to be certified as Designated Medical
Examiners. Many will be active pilots, with flight deck experience being
strongly recommended by ICAO. The interests and experience of these
doctors will vary considerably, and will often be channelled by their patients.

293
Aviation Instruction and Training

The country doctor will need to conduct initial and routine medicals, but may
require an extensive knowledge of toxicology if aerial spraying is conducted
in the area. Many of the chemicals used are toxic and can cause illness or
death. These include organophosphate insecticides, carbonate insecticides,
chlorinated insecticides and herbicides, nitrophenols and paraquat. The
doctor must be aware of the toxic properties of these substances and the
treatment in cases of poisoning. Being able to communicate this knowledge
in a readily understandable way to local operators is an important function for
the doctor involved in such work.
The doctor may be more involved in recreational flying, as an aero club
member, or as the local Designated Medical Examiner. Pilots are expected to
have knowledge of the effects of minor illness on flying, along with the
effects of drugs, alcohol, tobacco and ageing. They must have knowledge of
hypoxia, hyperventilation, visual problems, spatial disorientation, and G
problems. The doctor's role here can be variable and may involve formal
tuition, 'corridor' advice or discussion of the issues at periodical examinations.
Quite often, situations will arise where pilots are misinformed or ignorant of
their limitations or the limitations of their equipment. The doctor, by
providing sound and rational advice may be able to correct the pilot,
preventing problems before they begin.

The military flight surgeon

In the military, the flight surgeon is a doctor who has completed a formal
course of instruction in aviation medicine. This course may be limited and
brief, or may, in some cases, consist of an academic degree or diploma and
residency training. In some countries, the flight surgeon receives limited
flying training, is paid 'flying'pay', and is seen as a fully fledged member of
aircrew. The flight surgeon is, ideally, the squadron doctor, working
alongside patients and living and playing with them. This system enables the
doctor to gain a more intimate knowledge of the duties of aircrew and an
understanding of the stresses and problems experienced by them. The
surgeon is then in an ideal position to act as educator, adviser, supporter and
clinician to the aircrew.

Medical problems in aviation

While this chapter does not purport to provide a textbook of aviation

294
 Aviation Medicine and Flight Instruction

medicine for the pilot, some of the more common medical problems to be
encoumtered in aviation will now be discussed. Many of the medical
problems may be classified as operational problems, which are due to the fact
that people are in an environment for which their bodies were not designed.
These include hypoxia, decompression sickness, acceleration problems and
noise and vibration. There are, however, other problems of a medical nature
which may be faced by the pilot which would come broadly under the
definition of self induced stress. These include alcohol, smoking and self
medication. What follows is a brief introduction to these topics, which it is
hoped will stimulate the reader into making further enquiries, whether it be by
reading a specific aviation medical textbook, or by discussing some of these
matters with a local flight surgeon.

Hypoxia

A Beech Baron, under the command of a senior commercial pilot, was

involved in an incident in which a position report was not made. An alert
was declared. Contact was finally made and the aircraft descended to
below 10,000 ft. The flight was later completed normally, with
investigations showing that the pilot had fallen 'asleep' for 22 minutes. The
pilot had in fact suffered from an episode of hypoxia which was considered
to be allied to fatigue (Aviation Safety Digest).

Right represents man's achievements at escaping the clutches of the earth.

It is often said that 'If God wanted man to fly, he would have given him
wings'. This might be more accurately stated by saying that 'If God meant
man to fly, he would have given him wings and an oxygen mask and a
pressurized cabin'. Man is not physiologically equipped to cope at altitudes in
excess of 10,000 ft. Ascent to altitudes above 10,000 ft will cause everyone
to experience a degree of hypoxia sufficient to cause performance decrement.
The individual with a compromised cardiovascular or respiratory system will
suffer significant degrees of hypoxia below that - as low as 5,000 ft.
Hypoxia means oxygen deficiency in the tissues sufficient to impair
function. The most common type is hypoxic hypoxia which is caused by
insufficient oxygen in the inspired air. This is the major cause of hypoxic
incidents in flight. There are three other types of hypoxia - hypaemic
(insufficient oxygen carrying capacity of the blood), stagnant (abnormal
circulation e.g. in high G stress), and histotoxic (tissues cannot use the
oxygen). Perhaps the major feature of hypoxia is its insidiousnesss. It
literally creeps up on the sufferer, and has effects of which the sufferer may

295
Aviation instruction and Training

be quite unaware. Even up to 10,000 ft performance of normal tasks may be

impaired and the dark adapted eye may be compromised. Between 10,000 ft
and 15,000 ft the ability to perform skilled tasks is impaired with the sufferer
frequently being unaware of impaired performance. Physical work capacity
is markedly reduced and dizziness and headache may occur. Above 15,000 ft
there is a serious and marked decrease in psychomotor performance, slowing
of thought processes and marked emotional changes. Euphoria, disinhibition,
even violence may be seen. Tunnel vision may occur. Physical activity, a
cold environment, intercurrent illness and alcohol may all increase an
individual's susceptibility to hypoxia and modify the pattern of signs and
symptoms produced.
The effects of hypoxia may be summarised as follows:

1. Visual ability decreased.

2. Psychomotor performance decrease.
3. Cognitive function reduced.
4. Personality change.
5. Lack of insight.
6. Lack of judgement.

Much is often made of the 'time of useful consciousness'. The important

word here is 'useful', and the term 'effective performance time' more
effectively refers to the pilot's ability to function. Table 19.1 shows the
effective performance time for healthy individuals at rest, at various altitudes.
There is little point in being conscious if you are unable to recognize hypoxia
and take steps to deal with it. Effective performance time can be
considerably shorter than the time you are awake.
Aviators are usually well protected from hypoxia by pressurized cabins and
the provision of oxygen systems. However, failure of such systems is
relatively commonplace and their provision and use is not common in general
aviation. The pilot's best protection against hypoxia is however, a reasonable
knowledge of high altitude physiology, an appreciation of the insidious
effects of hypoxia, and the good sense to use this information. The treatment
of hypoxia is the administration of supplemental oxygen under pressure and
descent of the aircraft to below 10,000 ft.

Hyperventilation

A 32 year old female student, while being instructed in techniques of stall

recovery, was noted to be responding slowly to the instructor's commands.

296
 Aviation Medicine and Flight Instruction

The instructor noted her hands were excessively sweaty, her concentration
was poor, and she seemed to be staring straight ahead. The student's
breathing was noted to be slow and deep, and on enquiry from the
instructor as to how she was feeling, she complained of 'feeling funny
(tingling) around her lips. The instructor correctly diagnosed anxiety with
hyperventilation, and terminated the lesson so that the student's anxiety
could be dealt with prior to resumption of the exercise.

Table 19.1
Effective performance time at altitude

Altitude (in feet) Effective Performance Time

18,000 20-30 minutes

22,000 10 minutes
25,000 3-5 minutes
28,000 2.5-3 minutes
30,000 1-2 minutes
35,000 0.5-1 minute
40,000 15-20 seconds
43,000 9-12 seconds

Hyperventilation produces symptoms similar to hypoxia. Its most frequent

cause is is anxiety, and it is often seen in flight instruction, instrument flight,
low visibility landings and formation flying. It is a frequent accompaniment
to hypoxia. Hyperventilation is a condition of over breathing, either in rate or
depth of respiration. It causes an excessive loss of carbon dioxide from the
lungs, and a reduction in the normal acid component of the blood.
Hyperventilation causes dizziness, lightheadedness, sweating, tingling of
the lips and fingers, visual disturbances, muscular uncoordination and an
inability to concentrate on flying. It will be accompanied by rapid breathing
or slow deep respiration. Muscle twitching or spasm, and unconsciousness
may supervene. The sitution will not markedly improve if the sufferer is given
oxygen. The fact is that hyperventilation and hypoxia can be difficult to
distinguish, and aircrew should assume that any suspicious signs and

297
Aviation Instruction and Training

symptoms are due to hypoxia whenever the cabin altitude exceeds 10,000 ft.
Treatment of hyperventilation requires a reduction in the rate and depth of
breathing. Breathing into a bag that collects the exhaled carbon dioxide for
re-breathing is an effective treatment.
Because of the difficulty in distinguishing hyperventilation from hypoxia,
the use of a bag and breath-holding should be avoided at altitude. The
following steps should be taken when one or the other condition is suspected:

1. Administer 100 per cent oxygen under pressure.

2. Reduce the rate and depth of breathing.
3. Check the oxygen supply, flow meter, oxygen line and fit of the mask.
4. Descend the aircraft to below 10,000 ft.

Decompression sickness

In February 1992 a student pilot was conducting a solo navigation

exercise. He was suffering from a head cold at the time, and undertook the
flight despite advice to the contrary. The student suffered from severe pain
in the ears following his descent, with the pain making concentration
during the landing phase of flight extremely difficult, and according to the
student's account, dangerous. The student was forced to seek medical
attention, and was delayed in making his return flight for 24 hours. He had
suffered from otic barotrauma, and an obvious failure to exercise good
airmanship.

Ascent to altitude involves a decrease in barometric pressure placing man

at risk of hypoxia, decompression sickness and barotrauma if adequate
protection is lacking. Body gases may bubble out under conditions of
decreased ambient pressure causing decompression sickness, or they may
expand without a means of escape causing barotrauma.
Decompression sickness is unusual below 25,000 ft, but has occurred at
14,000 ft. It is mainly due to the bubbling out of nitrogen dissolved in the
body tissues and fluids, and does not occur if the nitrogen is removed by
breathing 100 per cent oxygen prior to exposure to altitude. The patient
complains of joint and limb pains, respiratory disturbance (a dry hacking
cough is common), with skin rashes and itching. Collapse may be seen.
Barotrauma is caused by the expansion of trapped gases. Intestinal pain
and aerotitis media with air trapped in the middle ear is seen. Air may be
trapped in the sinuses causing severe pain and an inability to concentrate.
Decompression sickness may be increased in severity and incidence by the

298
 Aviation Medicine and Flight Instruction

duration of exposure, re-exposure, exercise, cold and hypoxia. The after

effects of alcohol also increase susceptibility. Treatment includes descent to
ground level, 100 per cent oxygen and warmth. Local pressure applied to an
affected joint may relieve pain. In persistent and severe cases, treatment in a
hyperbaric chamber may be required.

Acceleration

Any aircraft can subject crew members to accelerative forces. However it is

the military pilot, the aerobatic pilot and the crop duster who are most
frequently subjected to positive and negative G stress. It should not be
forgotten that the student pilot is also frequently exposed to unfamiliar and
uncomfortable degrees of G forces that would not worry the more
experienced aviator.
G forces act on three axes of the body. If the axis is back to front, it is
referred to as X, if side to side, it is Y, and if head to foot, it is Z. Positive
and negative G stress is the major form of stress experienced in high speed
military flying, aerobatics and crop dusting.
High positive G mainly affects the cardiovascular system with the blood
becoming 'heavier' with a tendency to drain towards the lower parts of the
body. The heart is displaced downwards, thus increasing the distance this
■heavy' blood has to be pumped to the brain and eyes, and great muscular
effort may be required to raise the limbs and hold the head upright. The
result of all of this is that the eyes and brain may be starved of oxygen
resulting in 'grey out', 'black out', or loss of consciousness.
Fortunately, high positive G can be well tolerated for short periods.
Tolerance is, however, reduced by hypoxia, fatigue, illness, alcohol and age.
Tolerance may be increased by the use of an anti-G suit and the proper
performance of the M-l manoeuvre - exhaling forcibly against a partially
closed glottis, interrupted once every 3-4 seconds by a rapid inspiration
(Glaister, 1988). This may raise the visual threshold by 1-2 G.
Negative G acts from the feet to the head. It may occur in inverted flight,
an outside loop, or some rolls. Its effects are generally quite unpleasant. Its
cardiovascular effects are the opposite to positive G with vascular pressures
increasing above the heart. Pooling of blood in the facial and neck tissues
with engorgement of the lower eyelids occurs. 'Red out' with the pilot being
conscious of a misty redness is a result, and sub-conjunctival haemorrhages
may eventuate. There are no effective counter-measures to be taken against
negative G other than avoidance of the manoevres causing it.

299
Aviation Instruction and Training

Noise and vibration

All aviators are at risk of acoustic trauma. This is particularly so in general

aviation with poor protection from the noise of piston engine aircraft being
the norm rather than the exception. A piston engine noise a few feet away
constitutes 120 decibels, sufficient to cause discomfort in the ear. Sustained
noise levels above 86 decibels can lead to permanent or temporary hearing
loss, while exposure to 120 decibels for several hours a day, for 3 - 6 months
will cause deafness. Intense high frequency noise can cause confusion and
disorientation and must be regarded as a stress factor.
Protection against noise is vital to the long term career prospects of the
pilot with well fitting headsets providing attenuation of up to 27 decibels at
500 Hz, 32 decibels at 1,000 Hz and 37 decibels at 6 - 8,000 Hz. In the
arlines and the military, aviation medical departments provide programs for
hearing conservation. In general aviation the individual pilot bears the
responsibility. It is a wise pilot who takes this seriously and ensures that he
or she has a comfortable, well fitting headset with good noise attenuation.
Vibration is known to cause a reduction in pilot performance. Motion
sickness, fatigue, shortness of breath and abdominal and chest pain may
occur. Much vibration is unavoidable, with aircraft design often being a
compromise between safety, comfort and performance.

Fatigue

In November 1988 a Robinson R22-Beta Helicopter crashed. Mustering

had been commenced by the pilot at approximately 0500 hours. The pilot
flew for three hours, then carried out routine station duties prior to
mustering another paddock. After approximately 20 minutes airborne, he
decided to land, and in the process struck a tree, causing the tail rotor and
gearbox to separate from the helicopter. The pilot was unable to recall
having sighted the tree, with the outside air temperature at the time in
excess of 38 degrees centigrade. At the time of the accident the pilot had
been on duty for in excess of ten hours in temperatures in excess of 38
degrees centigrade. This accident is a stark reminder of the effects of
fatigue and heat on the human body (Bureau of Air Safety Investigation).

The whole process of flying is one which induces stress, both physiological
and psychological, with some degree of fatigue being an inevitable outcome.
The stresses of erratic duty schedules, hypoxic environments, acceleration
forces, noise and vibration and difficult environments all contribute to fatigue.

300
 Aviation Medicine and Flight Instruction

The failure to adhere to assigned altitudes and the wheels up landing at the
end of a long flight are all too common consequences of fatigue. Aviation
safety systems around the world comment frequently on the hazards of
fatigue with most fatigue related incidents occurring during the descent,
approach and landing phases of flight.
Fatigue affects all pilots, with no immunity being granted to airline or
military pilots. It may be acute or chronic. Acute fatigue, which is obvious
to the sufferer and everyone else, operates mainly on the muscles. Those
affected will feel tired, will have slow reaction times, and will have a reduced
ability to work efficiently. The cure is rest and sleep. Chronic fatigue is
insidious, and frequently goes unnoticed. It has adverse effects on all body
systems and may result in the so called 'nervous breakdown'. It is a result of
inadequate rest and recreation, often combined with poor diet and an erratic
lifestyle. High quality sleep and rest is essential if chronic fatigue is to be
avoided, with strict attention paid to dietary and other lifestyle factors.
Fatigue can be prevented or at least minimized by enlightened crew
scheduling, adherence to flight limitation times, dietary regulation, optimizing
sleep potential and good relaxation techniques and exercise. If sleeplessness
is considered by the doctor to be a major contributor to fatigue in specific
cases, its persistence after due attention has been paid to sleep habits and
work schedules may raise the question of the use of hypnotics. If the
sleeplessness is considered to be a temporary problem, with very firm
evidence of disturbed sleep, the use of a hypnotic, may well be warranted.
The hypnotic must preserve the normal sleep pattern, and must be free from
undesirable effects on daytime function such as the 'hangover' effect.
Fortunately, there are hypnotics that meet these requirements, and at present,
Temazepam, 10 - 20 mgs, is considered to be the drug of choice. It has a
relatively short duration of action, and accumulation does not occur. It
therefore gives an adequate margin of safety for aircrew during what may be
a demanding schedule. It should, of course, be tested on the ground prior to
being prescribed during flight schedules, with an interval of 24 hours elapsing
between ingestion and flying, although this may be reduced to 12 hours under
supervision. Temazepam is at present considered safe for use by a number of
civil and military aviation authorities including the United States Air Force
and Royal Australian Air Force. It is necessary to stress, however, that it
should only be used under medical supervision, and the use of alcohol should
be avoided while taking Temazepam. Its use should be limited to short
periods aimed as assisting the pilot to overcome sleeplessness during times of
specific scheduling or operational problems.

301
Aviation Instruction and Training

Alcohol

In November 1983 at about 0200 hours, the owner of a Piper PA24-400

and a passenger boarded the aircraft. The aircraft was heard to take off,
and was seen in the circuit area. It was noted to make a go-round on the
first approach, and landed after another circuit. Shortly afterwards a take­
off was carried out in the opposite direction, and after becoming airborne
the aircraft struck the ground. Neither person was qualified to operate the
aircraft, with post mortem examinations revealing that both persons had
high blood alcohol (Bureau of Air Safety Investigation).

This is perhaps the most widely used drug in the world. Its use by people
of all social classes from a wide variety of educational and occupational
backgrounds constitutes a significant drain on the health dollar. It is a factor
in about 40 per cent of separations and divorces and is a major factor in
violent crime, accidental death and suicide. It is a significant factor in fatal
light aircraft accidents with one UK study giving an incidence of 11.6 per
cent in which alcohol played a part (Underwood Ground, 1975).
Alcohol not only has an effect on the pilot flying under its influence, it also
has a detrimental effect on the pilot flying with a hangover. Alcohol causes a
progressive fall off in alertness, judgement and performance. Its effects are
most marked where multiple tasks are being performed simultaneously. Its
effects are in fact, very similar to hypoxia, and are intensified by it. A blood
alcohol concentration which would be of little significance at sea level, might
well be disastrous at 10,000 ft in an unpressurized aircraft. To make matters
worse, the metabolism of alcohol at low oxygen concentrations is reduced.
Alcohol reaches its peak blood level about 60 - 90 minutes after a single
dose. Consumption of 3 standard drinks (10 oz. beer, 120 ml. wine, 1 nip of
spirits) in one hour by an average sized male will give a blood alcohol of 0.04
mg. per cent, with 4 standard drinks giving a blood alcohol of 0.06 mg. per
cent. When one considers that a blood alcohol concentration of as little as
0.03 mg. per cent (2.5 standard drinks) will affect psychomotor performance
so adversely as to be incompatible with flight safety, the importance of
careful and restricted drinking will be appreciated.
Alcohol is eliminated at a constant rate of 10 gm. per hour. However, one
needs an awareness of the hangover effects of alcohol. During a hangover,
there is lowered resistance to disorientation problems such as the Coriolis
reaction. The symptoms of intoxication during a hangover are not readily
appreciable, and the impairment of mental and physical function may be
unrecognized until disaster occurs. Hangover effects have been shown to

302
 Aviation Medicine and Flight Instruction

decrease pilot performance with respect to maintaining heading, glideslope

and air speed by greater than a factor of 5.
Alcohol does not mix with flying. It should be regarded by pilots as a
major threat to air safety, and the practice of the pilot's art and science.

Smoking

Smoking is by now well recognized as a hazard to health. It links with

cancer, coronary artery disease, peripheral vascular disease and chronic lung
disease have been convincingly established. Smoking is now seen as a threat
to community health, but is it a threat to safety in the air - is there something
special about aviation which increases the risks of smoking?
Smoking has not been shown to be a causal factor in aircraft accidents.
However, theoretical considerations force one to the acknowledgement of
smoking as a potential hazard to aviators. Carbon monoxide is a by product
of cigarette smoke. It has an affinity for haemoglobin approximately 210
times that of oxygen and will displace available oxygen from the haemoglobin
molecule. Carboxyhaemoglobin is formed and some degree of hypoxia is the
result, even at sea level. The smoker is already hypoxic, even if not
symptomatic at sea level. With increased altitude, the smoker becomes more
hypoxic, with many smokers having a higher physiological altitude than
actual altitude. Thus, as sea level, a smoker with a 5 per cent
carboxyhaemoglobin level (not uncommon) has a physiological altitude of
8.000 ft, and at 5,000 ft, the same smoker has a physiological altitude of
10.000 ft. There can be little doubt that anti-smoking programs should be
offered to pilots.

Medication and pilot performance

A 61 year old pilot was undertaking a short solo instrument flight for
proficiency. At a height of 4,800 ft. six miles north of the airport, and in
cloud the pilot was noted to be off course. When asked to turn back on
course by Air Traffic Contro, the pilot began a left 180 degree turn.
Control of the aircraft was lost, with eyewitnesses reporting that the
aircraft emerged from clouds in a spiral dive, lost part of a wing, and
impacted the ground. Therapeutic levels of Diazepam (Valium) were
found in the pilot's liver, which may have contributed to spatial

303
Aviation Instruction and Training

disorientation, considered to be the most likely cause of the accident

(Garrison, 1992).

All medication has the potential to do harm. Even seemingly innocuous

medications such as aspirin, when taken in sufficient quantities, can cause
gastric bleeding, and other serious problems. Before flight is undertaken, the
pilot who practises good airmanship will ask and answer several questions.
Firstly, the pilot will ask if he or she is physically fit to fly. Secondly, the pilot
will ask whether he or she really needs to be taking any medication, and if
medication is being taken, will have ensured that it has been given a personal
trial on the ground for at least 24 hours prior to flight, to ensure that it will not
have any adverse effects on flying ability. Illnesses which on the ground,
may cause not more than minor problems or inconvenience, may assume
much greater importance in the air. Medications which may give great benefit
to those taking them on the ground, may become potentially dangerous once
in the air. It must be remembered that although a particular illness may not of
itself prevent one from flying, the treatment for the condition may in fact do
so. Because of the potential threat to flying safety brought about by taking
medication, those physicians specializing in or having a major interest in
aviation medicine, are quite aware of their side effects. It is unfortunate that
many physicians, who have little interest or knowledge of aviation, often
prescribe drugs without due consideration being given to their effect in the
cockpit environment.
Medications have both desirable and undesirable effects on the human
body. Undesirable effects may in fact be quite predictable, and consist of the
unwanted effects likely to be encountered when normal doses of medication
are taken. A good example of this is the blurred vision and dry mouth which
occurs when anti-cholinergic drugs are taken for indigestion. There may also
be effects from overdosage of medication in the particular individual. These
effects are of course unwanted, and represent an exaggeration of the normal
action of the drug. A good example of this is over-sedation from taking
sleeping pills. There may be allergic reactions, particularly common with
antibiotic, and there may be idiosyncratic effects of drugs. Such idiosyncratic
effects may occur in anyone taking normal amounts of medication.
Unfortunately, these effects are quite unpredictable, and it is vital that
medical advice should be sought if they are experienced.
It is beyond the scope of this chapter to provide an exhaustive list of
medications and their side effects, but a few examples might well serve to
illustrate the potential dangerousness of using medications without proper
medical advice. Generally speaking, any drug which can be classified as a

304
 Aviation Medicine and Flight Instruction

hypnotic or sedative should not be taken while flying. These drugs cloud the
mind, slow reaction time, reduce awareness, and cause mental confusion.
Many of them have quite severe degrees of 'hangover' effects, with
performance on the day following their ingestion being quite markedly
adversely affected. Tranquillizers, anti-anxiety drugs, and anti-depressants
are other drugs which should be treated very carefully. The rule is - DO
NOT FLY. The real danger here is self medication, and taking one’s
mother's, cousin's or friend's pills to relax or calm oneself, is treading in very
dangerous waters. Unwanted sedation, reduced awareness and a state of
euphoric optimism may occur with the effects lasting for many days after
cessation of the drug. If the drug has, in fact, been prescribed, then it is very
likely that the condition for which it is prescribed would be sufficient to
prevent flying. A commonly taken form of medication is the anti-histamine
group. Anti-histamines are to be found in a wide variety of compounds used
to treat an equally wide variety of illnesses, many of a minor nature. Fever,
exzema, dermatitis, and the common cold may be treated with some form of
anti-histamine. Side effects include drowsiness and sedation, with a dry
mouth and stomach upset being quite commonly experienced. If a pilot needs
to take a cold cure, a check as to whether it contains an anti-histamine should
be made, and a doctor consulted before running into trouble in the air.
Weight reducing drugs often contain amphetamines. The so-called 'pep pills'
are also amphetamine based. These drugs are stimulants, and are addictive.
They may cause sleeplesness, euphoria, over-confidence, disorientation, and
paranoia. There is little justification for their prescription, and they should be
avoided by all flight crew. Even the inadvertent taking of such medication
should be a bar to flying.
Caffeine is a social drug which is not always recognized as such. It is
mainly found in drinks, particularly in coffee, although it is present in tea,
cocoa, and cola based soft drinks. It is also present in chocolate. Caffeine
may also be found in over the counter stimulant medications. A cup of
coffee, of the instant variety, contains 60 mg. caffeine, with a percolated
coffee containing 100 mg. caffeine and drip coffee containing up to 150 mg
caffeine. A cup of tea contains approximately 40 mg caffeine and a can of
soft drink contains approximately 40 mg caffeine. In people who are not
regular users of caffeine, doses of less than 500 mg. may cause nervousness
and anxiety. In such people, doses above 500 mg. may be toxic. Certainly,
caffeine users who increase their intake suddenly by more than 500 mg. per
day frequently experience toxicity. The toxic symptoms include restlessness,
nervousness, excitement, insomnia, a flushed face, gastrointestinal problems,
facial twitching, cardiac arrhythmias, agitation and rambling thought and

305
Aviation Instruction and Training

speech. Withdrawal effects of caffeine include increasing lethargy,

drowsiness, headaches and irritability, with the caffeine withdrawal syndrome
beginning approximately twelve to sixteen hours after the last caffeine
ingestion. Many air crews use caffeine, though most do so in moderation.
Caffeine consumption enables one to accomplish motor tasks more quickly,
especially if affected by boredom or fatigue. Clearly, the use of small doses
of caffeine can be useful where a high degree of vigilance is demanded. The
fact is that a little caffeine may be good, whereas if used to excess, caffeine
may cause serious toxicity.
There are many other drugs that may be taken by pilots quite safely, and
without serious effects. There are also many others which should be avoided
at all costs. The general rule is that one should not fly while under the
influence of medication, and that even if medication is considered to be safe
for flying, it has been tested for at least 48 hours prior to flight. A healthy
respect for the effects of drugs should be maintained. Pilots should remember
that following local and general dental and other anaesthetics, a period of at
least 48 hours should be spent on the ground before resumption of flying.
It need hardly be pointed out that illicit drugs, including marihuana, cocaine,
heroin and LSD are absolutely contraindicated for use at any time, by any
member of flight crew. The taking or use of these drugs is of such gravity
that immediate suspension of a pilot's licence should be considered whenever
any of these drugs has been used.

Conclusion

In this chapter, the role of the medical practitioner in aviation has been
discussed. It is to be hoped that from the very beginning of the student's
career as a pilot, contacts made with the doctor or flight surgeon will be
positive, enhancing the pilot's enjoyment and practise of the profession and
improving aviation safety.
If this is to be achieved, the involvement of the doctor, from the time of ab
initio instruction, with appropriate input throughout the pilot's career should
be aimed at. Once a pilot gets to know the doctor as a fellow aviator and
preferably pilot, an atmosphere of mutual trust and respect can be built up.
The doctor can then treat, educate and counsel aircrew in an open and
friendly environment, which is more likely to allow the preventative aspects
of medicine to be practised, dealing with problems before they arise, and
more often avoiding the necessity of the dreaded 'grounding' of aircrew.

306
 Aviation Medicine and Flight Instruction

Doctors, of course, do have many other roles in aviation, including

involvement in the investigation of aircraft accidents, and research into
methods of making flight safer and consequently more enjoyable for all those
concerned. It is to be hoped that the student reading this chapter will have
gained a new perspective on the role of the doctor, and feel more comfortable
in using the doctor's knowledge and expertise as one of the tools in increasing
the student’s professionalism. This should apply not only to the class of
'professional' pilots, but also to those engaged in flying for sport and pleasure.
To assume that flying can be approached in anything other than a professional
manner is to make an assumption inconsistent with good airmanship.

References and further reading

ASD (1990), Aviation Safety Digest, no. 145, Winter.

BASI (1991), Bureau o f Air Safety Journal, no. 8, June.

BASI (1989), Bureau o f Air Safety Journal, no. 5.

de Hart, R.L. (1985), Fundamental o f Aerospace Medicine, Lea and Febiger,

Philadelphia.

Emsting J. and King, P. (1988), Aviation Medicine (2nd ed.), Butterworths.

Harding, R.M. and Mills F.J. (1988), Aviation Medicine (2nd ed.), British
Medical Association, London.

Garrison, P. (1991), 'No Go, "Aftermath"', Flying Magazine, July.

Glaister, D.H. (1988), 'Protection against long duration acceleration', in

Emsting, J. and King (eds.), Aviation Medicine (2nd ed.), Butterworths.

ICAO (1985), Manual o f Civil Aviation Medicine (2nd ed.), International

Civil Aviation Organization, Montreal.

Norwood, G. and Jordan, J.L. (1982), Regulatory Aviation Medicine, its

Philosophies and Limitations (FAA-AM-82-14), Federal Aviation
Administration.

307
Aviation Instruction and Training

Underwood Ground, K.E. (1975), 'Alcohol associated with fatal light aircraft
accidents, United Kingdom, 1964-73', Aviation, Space and Environmental
Medicine, no. 46.

308
20 Educating the aviation
professional
W.E. Pippin

The aviation industry in the US has undergone a significant transformation

during the era of deregulation. Now, more than ever, an Aviation
Professional needs a keen awareness and understanding of the political,
economic, and social, as well as the technological dimensions of the industry
in order to understand the trends and portents of the profession and be
prepared for the future. That future is very much effected by the implications
of the newly-coined word 'globalization', a term which is much in vogue as
we look to the next century of aviation progress.
For over a half century, the Department of Aviation at The Ohio State
University has offered an academic program to prepare students in the design,
management, and operation of aviation systems, present and future. The
Department Faculty monitor very closely the actions and events which affect
the aviation industry in order to maintain relevance and currency in the
aviation academic curriculum. Our projections for the industry are based
principally on a shared interpretation of events occurring in various sectors of
the industry, and the government policies and actions which affect them.
These assessments of the significance and import of events are, of course,
coloured by our own professional values, so these conjectures are subject to
the usual caveats. However, we believe that our propositions are logically
compelling and that they will be validated empirically over time.
The implications of aviation globalization, as we see them, have prompted
us to make significant change in our aviation baccalaureate degree curriculum
which reflect our consensus on the future of commercial aviation and the
career demands that can be expected to confront aspiring aviation
professionals in the near and far term. In this paper, the new curriculum
emphasis and the rationale for the associated changes will be discussed.
Aviation Instruction and Training

A new focus

The essence of the curriculum change is in the integration of the subject

matter; that is, introducing a holistic view of aviation systems and the
environment in which they operate, together with an emphasis on the context
in which special skills and knowledge are employed. Our premise is that
'Aviation', in sum, has become far more than a collection of job skills,
vocational trades, and disciplinary specialties. Fledgling aviation practitioners
must bring to their jobs an integrative and synergistic mode of thinking to use
effectively the profusion of technological inventions available to aviation
today and increasingly so in the future. Above all, they must be adaptable if
they are to contribute effectively in entry-level positions and to expect career
positions of increasing responsibility. Further, aviation leaders should be able
to create the additional social inventions which enable organizations and
people to master new developments in technology rather than be threatened
by them.
While these assertions may seem to imply simply a traditional 'liberal'
education, the intent is a good deal more. To clarify what is meant by 'more',
we should recognize that aviation education at the college and university level
is relatively new in the US. In part because of newness, collegiate aviation
programs seem to suffer from an 'identity problem'. There is not a general
understanding among prospective students or employers of what an aviation
education entails. A few educators who are comfortable only in the realm of
the traditional disciplines commonly associated with higher education, would
even question whether aviation is an appropriate or legitimate field of study
for a baccalaureate degree.
Why offer a baccalaureate degree in Aviation? What is so demanding or
otherwise special about this particular service to the public that justifies a
special program of study when most other areas of commercial activity are
served by education in the traditional academic disciplines or by skill training
in vocational trade schools?
The short answer is that Aviation has become critically important to almost
every economic endeavour, and aviation systems are becoming increasingly
complex and consequently less able to accommodate on-the-job learning or
tolerate trial-and-error practices. The social consequences of operational
failure in Aviation, for whatever reason, are sufficiently grave to demand the
level and type of professionalism that can be afforded by higher education.
As we move to clarify this answer more fully, a brief historical summary of
Ohio State's aviation program will provide a useful departure point for the
reader.

310
 Educating the Aviation Professional

Aviation at Ohio State

This year, 1992, aviation education at The Ohio State University (OSU)
marks its 50th year of continuous evolution. Since the introduction of
aviation education on the OSU campus in WWII, the program purpose and
objectives have progressively expanded from the initial offering of flight
instruction to augment wartime military training, to a much broader concept
commensurate with the growth and critical economic role assumed by
commercial aviation and its demand for educated professional practitioners.
From its beginnings as an appendage to other degree programs, aviation
studies evolved first as a major specialization in the College of Business and,
several years later, gained additional status as a major curriculum offered in
the College of Engineering and the Colleges of Arts and Sciences. Over 500
students are enrolled in Aviation degree programs.
The Aviation curriculum in each of these three colleges is currently
comprised of three components:

1. a selection of basic courses called the General Education Curriculum, or

GEC, which provides a foundation in mathematics, the physical and
social sciences, and the humanities;
2. a set of Aviation core courses and their introductory prerequisites; and
3. a group of aviation specialization courses.

The GEC is a University-wide baccalaureate requirement, but each college

faculty has a degree of latitude in selecting the mix of specific courses. The
Aviation Major adds to the GEC a common core of aviation-specific courses
in communication, law, management, and human factors, the intent of which
is to relate the subject matter in these generic knowledge domains to its
application in aviation. These core courses are preceded by introductory
prerequisites which describe the evolution of aviation and its various
organized activities, together with an introduction to flight phenomena and
the regulation of aviation systems. The introductory courses, then, survey
and describe the activities of aviation while the core courses provide insights
into the operational context for the activities and the functional work done by
aviation firms.
Certain categories of functional study have been offered in the past as
aviation specializations. For example, an Aeronautical Science option
provided a basic aeronautical engineering approach to aircraft design and
performance although it was necessarily much abbreviated in depth and scope
from a full aeronautical engineering program; an Aviation Communications

311
Aviation Instruction and Training

option focused on effective communication in a variety of media; an

Electronic Systems option emphasized basic electrical engineering courses,
but also was abbreviated from a full electrical engineering program; a
Computer Science option introduced the use of computers in aviation
systems; and an Atmospheric Science option dealt with aviation
considerations of weather phenomena.
Each of the specializations listed above had two common and salient
characteristics. First, each was operationally-oriented at the necessary
expense of theoretical depth; that is, the focus was on application, technique,
and utility of an area of functional knowledge as applied to aviation activities
and tasks. Second, these areas are also the province of traditional degree
programs which produce functional specialists with in-depth theoretical
knowledge which is applicable to aviation operations without much
adaptation. Because of these considerations, these options were phased out
of the curriculum resulting in a sharper program focus.
The new focus is on the three other options offered, which do not have
counterparts in other established degree programs and which further differ in
their orientation: Aircraft Systems, (the professional pilot course sequence);
Aviation Human Factors, study of the human-machine interface in the
aviation environment; and Aviation Management, which would appear to
compete with other degree programs but does not because of its focus on the
unique aspects of the political, social, economic, and technological
environments of aviation. These three options are the key determinants of the
next stage of curriculum evolution, as the program which we have briefly
traced transitions to a new stage of development.

The need to move forward

'Evolution' is an important condition for any educational program, but a

continuing evolutionary state is particularly appropriate for studies which
prepare students for a career in aviation, a profession which has been
exceptionally dynamic and continues to undergo dramatic and increasingly
rapid change. We should recognize, however, that aviation has a colourful
history which fosters a strong sense of tradition among many of its
practitioners. In any profession where there is much respect for the deeds of
the pioneers and some attraction to regaining simplicity by recreating the
past, change does not come easily from within. It is more often imposed from
without by those who are dependent on the profession and are not being well-
served.

312
 Educating the Aviation Professional

There is yet another restraining link to the past. As with other complex
industries which employ a number of occupational specialties, aviation firms
have acquired managers and executives from a variety of traditional degree
programs in the sciences and the business and engineering schools or,
alternatively, have 'grown their own' managers through internal executive
development programs and promotions from the technician ranks. In both
cases, general knowledge about aviation is acquired not through formally-
structured study but informally, on-the-job, and over an extended time. Such
managers may not recognize or appreciate the value of formalizing and
teaching the lessons they learned the hard way and may trust (and hire) only
those who have learned as they did.
These are among the challenges to aviation education which prompt our
faculty to review and refine our curriculum's traditional focus and clarify its
modern context.

The traditional aviation focus

It is quite natural that the general public associates aviation primarily with
aircraft and the skills needed to fly them. Airplanes are an exciting and highly
visible part of aviation. If one looks beyond the obvious needs of the cockpit,
however, the aviation profession encompasses an enormous range of
complementary activities which draw on virtually all of the traditional
knowledge domains recognized in higher education curricula. Moreover,
cockpit personnel themselves have become highly dependent on critical
support systems beyond their physical reach. It is hardly remarkable that
aviation practitioners on the flight deck as well as those in non-flying (but
equally essential) roles come from a wide variety of educational backgrounds
ranging from highly-specialized vocational training to traditional higher-
education degree disciplines. Formal study of Aviation as a profession,
however, is a relatively new addition to baccalaureate degree programs, and
in many such programs there remains a significant emphasis on task skills and
job training.
Only recently, and not in all quarters has the flying machine itself been
questioned as a primary professional focus. Notwithstanding the continuing
importance of aircraft technology, per se, and the human piloting skills
needed to operate increasingly-sophisticated flying machines, these elements
are becoming less significant as limiting factors in future aviation
developments. Total system considerations must be added to the study of the

313
Aviation Instruction and Training

component parts if aviation practitioners are to be truly educated for the

future.
Aviation systems take many forms, but let us for a moment move beyond
the vehicle and its support systems and consider the implications of recent
changes in other systems.

The social and political context

High-speed travel by air, with the apparent effect of distance compression, is

an important part of a larger revolution in human communication. While the
application of new information technologies is revolutionising intellectual
contacts over long distances, advances in aviation technologies are facilitating
communication between people and cultures in a physical as well as an
intellectual sense.
Although aviation is changing patterns of human activity, the reverse is
also true. The industry is being transformed by world political events as well
as by new technologies. Along with the accelerating technological evolution,
the aviation industry is also undergoing radical changes in organization
brought about by shifting political priorities and economic dependencies on a
global scale. These are coupled with the dramatic realignments in corporate
structures which are accompanying domestic deregulation. The resultant
pressures are transforming the aviation industry, even as aviation contributes
to transforming society. Despite national agendas to the contrary, aviation
systems are acquiring global characteristics.
Does globalization require a different approach to career preparation in the
aviation system? We believe that it does, and that an evolutionary change in
aviation education must accompany the evolution of a global aviation system.

Education for the future

The prospect of a more-globalized aviation infrastructure together with an

increasing degree of cooperation and interdependency among the user air
carrier corporations not only offers significant challenges and opportunities to
prospective aviation professionals but also places new demands on their
education and career preparation.
Even now, successful aviation practitioners must recognize, define, and
solve problems which cross the boundaries of social cultures and traditional
knowledge domains. Even more so in the future will these professionals need

314
 Educating the Aviation Professional

to integrate the analytical skills and perspectives of many functional

specialties with the operator skills normally associated with aviation
activities. We expect that the emerging structure of the industry and the new
political and economic dimensions of its operating environment will,
therefore, place the narrowly-focused functional specialist increasingly at a
disadvantage as a manager and decision-maker.
The pace of change in aviation system structures, organizational
relationships, and system technology has become so rapid that instruction
which teaches only specific examples of today's system requirements is likely
to be obsolete before the student graduates. Moreover, successful aviation
leaders are demonstrating that an educational perspective that extends beyond
the details of a particular job or specialized area of work is increasingly
important to understanding the changes that are taking place in the systems.
We believe that these considerations strongly favour a broad, conceptual
study of the aviation environment from a multi-disciplinary perspective.
There will continue to be roles for functional specialists having in-depth
knowledge in narrowly-defined areas, many of whom must study at the
graduate level to rise above the technician ranks. However, a baccalaureate
curriculum in aviation which prepares a person for a long-term career, one
which focuses on decision-making rather than simply task performance, must
emphasize subject matter which will enable analysis and interpretation of
future conditions in the industry as well as a better understanding of the
current total environment and its complex of tasks. In general, these decision-
making/problem-solving skills require a strong foundation in the basic
mathematical, physical, social, and behavioural sciences and well-developed
communication skills with which to articulate problem definitions and
solutions.
We will examine these propositions further as we discuss our concept of a
baccalaureate curriculum oriented to the aviation career needs of the future.
Let us begin by establishing a context and meaning for a 'globalized aviation
system'.

Realities of the present

A 'globalized' aviation system conjures images of a complex aggregation of

national and international firms and institutions providing world-wide
transportation in an apparently-seamless web of interconnections in which
political and geographical boundaries are both transparent and effortlessly
permeable.

315
Aviation Instruction and Training

The realities of the global aviation system at the time of this writing,
however, are far from such an idealised concept. The web of services is
neither seamless nor are the interconnections devoid of national politics and
social constraints as well as economic and technological limitations. The
global system has not yet evidenced an internal cohesive attraction among its
components sufficient to satisfy the wide variety of national agendas which
each country's aviation system serves. Yet the global system, as a rather
loose and shifting framework of inter-government agreements and corporate
alliances, seems driven toward a more efficacious integration by economic
considerations as well as by customer demand. Moreover, technology
advancements and political realignments are removing many of the old
impediments to system integration. The move toward globalization of
transportation systems would now seem to be inexorable.
In its present form, the global system is an eclectic aggregation of
organizing concepts for its operational resources and supporting
infrastructure. Though global in scope, it is 'globalized' only in the sense of
being connected by a maze of political, social, and economic ties. Yet it is
within this holistic framework that aviation components will either thrive or
perish. Individual national firms cannot remain insular in a world with far
fewer communication barriers. All of these considerations suggest to us that
professionalism and competence in a globalized aviation system must take on
new meanings.

Competence in a global system

As a general statement of belief regarding what should be considered college-

level aviation education, most professionals would agree that a baccalaureate
degree in aviation should prepare a person for an initial contributing role in a
particular area of aviation enterprise, and it should lay a foundation and
framework for developing career opportunities of wider scope. This sense of
broad purpose is reflected in the mission statement of The OSU Department
of Aviation which states in part that:

The Aviation academic program will prepare a student with the foundation
upon which to develop and function as a professional responsible for the
design, management, and operation of tomorrow's safe and efficient
national and international aviation systems ...

316
 Educating the Aviation Professional

While a statement such as this seems to provide a definitive purpose, there

can be wide differences in interpretation of the operative words and phrases.
There is much in this statement (and in similar ones in other aviation
programs) that begs further definition and clarification in order to convey a
clear program identity to a prospective student or employer or serve as a
curriculum guide. On review, we concluded that the statement raises more
questions than it answers.

The two aspects of professionalism

Objective Competence. Consider for example the phrase, 'function as a

professional'. This seems to imply a higher level of expected contribution
than simply performing a set of tasks skilfully or carrying out some given
instructions with appropriate precision. Indeed, 'Professionals' should be
known for their ability to create, organize, and modify tasks, which suggests
a level of understanding requiring an extensive knowledge of the physical and
social world. Should not this implied foundation be more explicit?
Further, and as discussed earlier, the term 'national and international
aviation systems’ is a deceptively-simple name for a very complex concept,
one which encompasses a multitude of separate but interdependent, private-
sector firms interacting economically and politically with several public-
sector service and regulatory institutions. A Professional should have
knowledge of the separate components and their functions but understand the
context of the 'system' as an integrated whole. Does offering narrow
specializations imply a contradiction?

Subjective Competence. Although the supposedly august status of

'profession' suffers currently from lower-case usage and casual application,
'Professions' have been (and perhaps still should be) known by their guiding
principles and laudable social purposes as well as by their arcane knowledge.
Thus, it should be understood that 'Professionals’ adhere to high standards
and obligations based on a shared set of values, while also possessing
knowledge and skills which are not easily understood or evaluated by the
general public.
One might ask whether 'Professionalism' be explicitly defined in the
curriculum in terms of ethical values, decision models, and representative
applications, or should it be left implicit for later definition in the workplace?
If included in the curriculum, should it 'colour' or 'flavour' the curriculum
content and perhaps bias the student toward or away from certain

317
Aviation Instruction and Training

conclusions, particularly conclusions concerning 'appropriate' (and

inappropriate) professional behaviour?
Recognizing that either technical incompetence or irresponsible behaviour
can have equally disastrous consequences in an aviation environment, we
conclude that 'Professionalism' does require 'conscience' as well as
'competence', no less so than the professions of medicine or law, and that our
aviation curriculum should deal explicitly with a professional ethos. We
consider that both competence and conscience are integral to a professional
demeanour and as inseparable as the two faces of a coin.

Interpretation, analysis, and decision-making

The new program orientation and curriculum revision is intended to increase

the aviation student's competence to deal with problems that are essentially
interdisciplinary and to increase his or her adaptability to change. The desired
end is that students learn not only the skills and techniques employed in
aviation activities, but more importantly, be able to understand them in
context, translate and modify them for other contexts, and recognize problems
caused by changing environmental circumstances.
This approach emphasises the reasoning methods and decision criteria
which underlie and precede actions rather than emphasizing the actions
themselves. Why' becomes more important than 'What', 'When', and 'How' in
sequencing the learning process. We intend that the student develop the
attitudes and techniques of critical thinking and that this mind-set be carried
over into the workplace.
Some critics may infer that this order of priority demeans the importance of
skill perfection and the high levels of task performance that have traditionally
been the pride of aviators. We dispute that interpretation in the strongest
terms, but we also recognize its basis. In teaching aviation functions and
activities which involve task skills, one must often choose between polishing
skills through repetition and practice or spending the time on academic
preparation, analysis, and a better understanding of the skills. In a curriculum
of finite length, trade-offs must be made and a balance struck between
education and training. Let us look more closely at that distinction.

Training or education

In academic vernacular, the so-called 'inquiring mode of thought' may be used

as a distinguishing characteristic between 'education' and 'training'. In other

318
 Educating the Aviation Professional

words, while the same factual knowledge and skill-levels may be associated
with both an educated person and a trained person, they will comprehend the
subjects differently because of the different manner in which each has
assimilated the knowledge or acquired the skills.
Whereas the educated person will, by this definition, understand why as
well as how knowledge and skills are combined to accomplish desired ends,
the person who has been merely trained is less likely to know or ask why a
task is done. Through repetition and practice, a trained person can appear to
be exceptionally competent as measured on performance scales. However, a
trained person may not recognize unique conditions which require subtle
variations on a task learned by rote or, in fact, which may make the task
totally inappropriate for the given circumstances. Therefore, in an area of
applied knowledge, such as aviation, the prevalence of inquiry and in-depth,
scientific explanation may be a useful measure of program teaching level and
its professional quality.
For purpose of clarifying program 'identity', we believe that an emphasis on
a questioning mind-set, together with analysis, extrapolation, generalization,
and contextual understanding, define the major differences between an
'aviation education program' focused on cognitive skills and problem-solving,
and an 'aviation training program' focused primarily or exclusively on skills
associated with a particular job or area of specialized work.
For practical reasons, however, an education-oriented aviation program
may not exclude rote learning entirely. For example, the process of acquiring
some basic flying skills initially requires the memorizing of (and, for safety
and survival, absolute obedience to) certain rules and procedures which may
not be fully understood until later stages of learning. This fact does not
preclude continual encouragement of a questioning attitude on the student's
part which may allow training to convert to education at an earlier stage.

Focusing the curriculum

Considering the range of activities and roles in aviation and the different
nature and demands of the tasks involved, we have elected to focus on three
specializations (or roles) mentioned earlier, Aircraft Systems (Professional
Piloting), Aviation Human Factors, and Aviation Management. In our view,
these areas provide an unlimited scope of responsibility, will become more
critical in a global system, and also benefit most from the great richness and
diversity of instruction available in a major university. We will not describe
the courses and content of these specializations here, but we will offer a

319
Aviation Instruction and Training

classification scheme which we have found useful in analysing the balance of

the curriculum.

Program content

A program's place in the education/training spectrum and its 'character' is

largely determined by the type of subject matter in the curriculum together
with the level at which the material is taught. 'Teaching level' is based on the
strength of the basic knowledge foundation which enables accurate and
sophisticated explanations of subject matter to be understood by the student.
A classification of subject matter is useful for content comparisons. We offer
this curriculum taxonomy for analysing the composition of an undergraduate
degree program:

- General (or Basic) Education:

Subject matter of wide scope and general application. The General
Education Curriculum (GEC) provides acquaintance with the core of
human knowledge in the arts, sciences, and humanities which a faculty
believes is the necessary minimum for all 'educated' persons. Such
knowledge provides the basis for interpreting and understanding the
physical and social environments.
- Discipline-related Education:
Analytical and interpretive subject matter which may have limited
application or be derived specifically for certain knowledge domains or
disciplines (e.g. engineering mechanics, concepts of systems analysis,
micro-economics, etc.).
- Professional Education:
'Technical' or otherwise-arcane subject matter which pertains to a
specific area of knowledge application. Included is this category are
organized studies of the applications of basic science and technology
(e.g. 'principles o f ...' - type courses such as aerodynamics, electronics,
etc.).
- Professional Socialization:
Material which pertains to the values, standards, customs, traditions, and
other elements of a particular professional ethos (sometimes referred to
as the 'corporate culture'). Not included in this category is the generic
study of social value systems, which we consider a part of general
education.

320
 Educating the Aviation Professional

- Professional Training:
Skills, techniques, methods, and other task-related learning which is
narrowly-oriented or tied to a specific application or not easily
generalized, (e.g. aviation regulations (though not the study of their
origins), rules of accounting (though not the study of their principles),
some computer-programming languages, flying techniques, etc.).
- Professional Practice:
Perfecting already-learned skills or applying them in a new context.

In an undergraduate curriculum, effective teaching usually involves relating

theory to a variety of applications in order to facilitate understanding and
generalization. It is likely, therefore, that a particular course of instruction
may be characterized by more than one of these orientation categories. Partial
weights may be appropriate in classifying the principal orientation of a
particular series of lessons.
With these classifications, one may analyse an existing program curriculum
and make a rough determination of where it fits in the education/training
continuum. That is, we may infer from classifying the curricular material
whether the program has a job-skills orientation or whether it leans more
toward interpretation, analysis, and problem-solving.
In that regard, this classification scheme is a useful aid to students in
retrospectively analysing their education and integrating an experience
fragmented by modular instruction. It is even more valuable if the analysis is
continually made and integration takes place during the learning experience.
In either case, integration is a necessary concomitant to a student's
educational self-awareness.
Choosing subject matter which truly associates with the defined objectives
in both kind and degree is more problematic, of course, than simply listing
desirable subject titles and course descriptions. The questions of 'How much
is enough?' and, 'What is the appropriate emphasis?' are governing
considerations in any program of specified length. Program balance is not a
trivial task, and we would argue that program balance, together with teaching
level/quality and student performance standards, determine the 'strength' and
'character' of an aviation education program

Summary and conclusion

We have not attempted to offer or suggest a recipe for educating all aviation
professionals, but rather to share with the reader this faculty's choice of

321
Aviation Instruction and Training

program focus and orientation, and our philosophical basis for that choice.
The gist of this essay may be summarized in the words 'competence' and
'conscience'. The public should expect and demand no less from aviation
professionals. We have offered our approach to imbuing those qualities in
aspiring aviation students; other programs may have equally effective means
to their professional ends. However, we believe strongly that prospective
students and prospective employers must become better aware of the value of
educated aviation practitioners as well as the distinctions in purpose among
aviation programs and what significance these distinctions hold.
To the two complementary attributes of competence and conscience which
we would intend for aspiring aviation professionals, we may add a synergistic
third, 'confidence'. We suggest that confidence comes from understanding
and being comfortable with what professionalism means in aviation, from
being able to integrate a fragmented series of education and training courses
into a cohesive, sensible whole, and from an articulate self-awareness of what
one knows (and doesn't yet know) which creates a sense of trust in a
prospective aviation employer. These three C's will clearly convey the value
of a baccalaureate degree in aviation.

322
21 Non-technical training
and education of ab
initio student pilots for
KLM Royal Dutch
Airlines
A. Droog

The KLM Flight Academy is the main Dutch flight academy, situated at
Eelde Airport near Groningen in the north of the Netherlands. It has
belonged to KLM since 1991 after having been a government school since
1946. The school has a capacity of 50 ab initio pilots a year. Moreover
there are a number of conversion courses for flight engineers of KLM,
helicopter pilots of KLM ERA Helicopters and air force pilots, preparing
them for an airline pilot career with KLM. Flight Instructor's Courses are
given for the Royal Netherlands Navy, which has its basic flight training
with the KLM Flight Academy.
The fleet of the academy includes the Slingsby T67C Firefly (9), the
Beechcraft A36 Bonanza (9) and the Cessna Citation I (4). One of the
Citations is operated for beacon-calibration. The three Citations for training
are also used for passenger transport purposes such as flights with the royal
family and government leaders. Eurotransplant has a contract with the KLM
Flight Academy for the execution of medical flights.

A b initio training program

The ab initio training course takes 24 months. Flight training starts on the
Slingsby Firefly, including some aerobatics to make the trainees acquainted
with the control of unusual attitudes and manoeuvres. The advanced and IF
A v i a t i o n I n s t r u c t i o n a n d T r a in in g

The fleet of the KLM Flight Academy:

the Slingsby Firefly and the Bonanza (top); the Citation (below)
(Photos: KLM)

324
 Non-Technical Training and Education ofAb Initio Student Pilots fo r KLM

training are done on the Beechcraft Bonanza. The crew coordination and
crew cooperation training also start in the advanced phase. Trainees
perform mutual flights, the Pilot Not Flying (PNF) being the safety pilot.
After this phase they are on the CPL + Instrument Rating level.
The third part of the training includes the Type Rating Citation and the
KLM Flight Academy Final Examination. After the type-rating exam the
trainees perform 20 hours of route training all over Europe on the Citation,
giving them important experience with navigation and R/T on the one hand
and flight management and captaincy on the other. Having their Type
Rating, the trainees are in front as captain and co-pilot; the instructor is in
the instructor's seat. At the end of their training their total number of flying
hours amounts up to 175 of which 110 hours are as Pilot in Command.

Table 21.1
The flight training syllabus: number of flying hours

A/CType A/C hours Simulator hours

Firefly 65 29 (General Aviation

Trainer)
Bonanza 70 29

Citation 40 + 20 PNF 38 PF + 38 PNF

As shown in Table 21.1, all flying training is supported by simulator

training: since 1988 the school has a full-motion C-500 Citation simulator
and since 1992 a new Bonanza A36 simulator with color visual screen.
The following table shows the groundschool program. Trainees leave the
school having passed the ATPL Theoretical Examination. They start their
career with KLM as First Officer Boeing 737 or Second Officer Boeing 747-
300 or -400, sometimes at the age of twenty. During their course, the KLM
trainees live at the boarding school on the terrain. There is a restaurant and
there are facilities for sports and recreation.

325
A v i a t i o n I n s t r u c t i o n a n d T r a in in g

The Bonanza (top) and Citation simulators (below)

(Photo: KLM)

326
 Non-Technical Training and Education ofAb Initio Student Pilots fo r KLM

Table 21.2
Groundschool - subjects and hours

Aircraft Performance 120

Engines 95
Instrumentation 90
Navigation 130
Avionics 130
Meteorology 95
Radio Telephony 75
Law and Regulations 60
Psychology and Physiology
of Aviation 25
Sports 120
English 60

Psychology and Physiology of Aviation will be replaced by a Human

Performance and Limitations Syllabus according to European requirements
and standards, although the existing program already meets most of the
requirements of the Joint Aviation Authorities.

Selection of trainees

The 50 ab initio trainees have been selected out of a group of about 1000
applicants. Basic requirements are that they have at least finished secondary
school with grades in mathematics, physics, and english. The school
stimulates boys and girls with higher education to apply, as the level of the
theoretical and practical training is rather demanding. The trainees are
between 18 and 25 years of age and the selection process is intensive.
Fields for selection are capacities and personality: intellectual abilities
(verbal, spatial, technical, arithmetic, logical), time-sharing abilities, respon­
sibility and discipline, exactness, teamspirit, emotional stability, self-con­
fidence, work atitude and motivation. Flying aptitude is assessed by means
of a simulator grading.

327
Aviation Instruction and Training

Non-technical training

Both within KLM and the KLM Flight Academy, much attention is given to
non-technical training. Human Factors Training and Crew Resource
Management have been an integral part of KLM flight training since the
seventies, when Frank Hawkins introduced the KLM Human Factors
Program (KHUFAC). KLM by now is in a process of restructuring all crew
management and instructors' courses to achieve full integration of technical
and non-technical training. A feedback and appraisal system has recently
been developed to be able to give crews feedback on their non-technical
performance.
The KLM Flight Academy also has a human factors and non-technical
training program. This program is meant to support the crew-cooperation
training, to enhance trainees safety awareness and to stimulate personality
development. The ultimate goal of high quality pilot training is a pilot who
is a skilled manager of the flight and a teambuilder. The young airline pilots
must gradually develop themselves as team leaders. When receiving their
first command the pilot will be the leader of the crew, motivating them to
optimal performance and taking decisions that lead to a safe, economic,
efficient, comfortable and satisfying flight within the framework of the
airline policy. Figure 21.1 illustrates this. It shows the different roles of the
captain when executing his flight. The captain is the 'nerve center' of the
flight (Mintzberg, 1973), like a manager is of a department. The Captain's
'inputs' are formal duties and responsibilities, the flight-plan and the
company policy the Captain has to adhere to, the external and internal
resources and processes which are monitored (weather, status of the aircraft,
its passengers and cargo, ATC demands, the crew). The 'outputs' are the
Captain's decisions and actions as pilot (aircraft handler), resource manager,
motivator of the crew and ambassador of the company. Decision making,
which is the use and allocation of technical and the human resources, is in
fact the Captain's core-activity.
Flight training, theoretical knowledge and non-technical training and
education are aimed at the preparation of the modem pilot for the above
shown set of roles and tasks. The non-technical training program for ab
initio students as shown in Figure 21.2 has been developed to lay a firm
base of awareness and elementary skills in the human factors and non­
technical field. The program parallels the development of flying skills. To
be effective, a non-technical training program should be integrated with

328
 Non-Technical Training and Education ofAb Initio Student Pilots fo r KLM

flying training as much as possible. Non-technical themes should be

introduced to the trainee when relevant to the technical training.

Figure 21.1
Flight and crew management are team leadership

Initially, learning to fly an aircraft is an individual learning process. In the

basic training the objectives of training are: learning to control and
manoeuvre the aircraft, learning the basic procedures for communication and
navigation, learning the regulations and the assessment of weather
conditions. In the non-technical field, the development of self-confidence
and safety awareness are of highest priority during this stage. Cooperation
with the instructor, assertion and basic decision making also come into
focus.
During the advanced training, the crew-coordination procedures and
crew-cooperation are trained. The student pilots will work together during
mutual flights. Working in a crew-concept necessitates teamwork qualities
and team behaviour.

329
Aviation Instruction and Training

Figure 21.2
Modules of non-technical training

Firefly Bonanza Citation

Outward Bound Crew Concept Crew Coordination

Training Briefing and Cooperation

Aviation Psychology
and Physiology
Decision Making Captaincy
Safety and
Leadership
Communication and Cross-cultural
Teamwork Basics Communication

Sports and Fitness

Stress-coping techniques
Anti-airsickness Training Program

In the final training, the problems of flight management and captaincy on

a twin-engine jet aircraft must be mastered. Anticipation, quality of decision
making, setting of priorities, crew-concept and captaincy are key issues
when operating the Cessna Citation to destinations all over Europe.

Non-technical training modules during basic flight training

1. Outward Bound Training

A 'take off training of one week in an unknown, non-flying
environment. At the very beginning of the course.
Themes: learning to know each other, mutual trust, self-confidence,
physical endurance, communication, team problem-solving, leadership.
2. Psychology and Physiology of Aviation
Themes: effects of altitude, disorientation, G-forces, human
information processing, stress, fatigue, relaxation.

330
 Non-Technical Training and Education of Ab Initio Student Pilots fo r KLM

3. Safety
Themes: factors inducing human error, hazardous thought patterns,
risk, rules, the I AM SAFE checklist.
4. Communication and Teamwork: Basic Elements
Themes: 'transmitter and receiver', active listening, communication
blockings, ambiguous messages, verbal and non-verbal communication,
group processes, influencing others, assertiveness, competition,
dominance, working together.

Non-technical training modules during advanced training

5. Crew-Concept Briefing
Themes: division of tasks in the cockpit, need for coordination,
responsibilities of the Pilot Flying and Pilot Not-Flying, intentions,
cross-checking, speaking up, crew-briefing and mutual agreement,
checklists, calls, acknowledgements, trans-cockpit authority gradient,
video demonstration of crew-concept in action.
6. Decision Making and Leadership
Themes: the decision making process, factors influencing decision
making, choice and conflict, decision making and stress, team decision
making processes, group think.
Themes: leadership, how to motivate, team building, delegation.

Non-technical training modules during final training

7. Crew-Concept
Line Oriented Flight Training sessions registered on video, feed-back
and reflection upon a trainee's performance.
8. Cross-Cultural Communication
Theme: understanding other cultures, habits, values, prejudice,
communication with people of other cultures, exchange of experience
with KLM-crewmembers.

Sports and fitness are paid attention to throughout the whole of the two
years' course by practice and information. An Anti Air Sickness Training
program (desensitization and progressive relaxation) is possible for students
with persistent air sickness problems, which seldom occurs.

331
Aviation Instruction and Training

The total number of hours in which attention is given to non-technical

matters and human factors amounts up to 120.

Mental practice, remedial training and counselling

A training program and its modules are not the only instruments to help in
developing the technical and non-technical skills of a pilot-trainee. Trainees
are made familiar with the principles and techniques of mental practice to
support their learning from the start. They mentally rehearse and imagine
flight manoeuvres and situations, in their study and with the aid of mock-up
cockpits, to be better prepared for actual flight lessons.
When problems arise during the learning of certain skills, instructors,
trainee and psychologist work intensively together to find the nature of the
problem and the solution to it by a remedial program with clear training
objectives and extra hours. In basic flight training individual trainees may
be too stressed by fear of failure or may lack in assertivity. These problems
are dealt with on an individual basis by counselling by the psychologist.
The psychologist in the ab initio flight-training environment will,
therefore, have multiple roles: trainer, supplier of human factors
information, counsellor etc. Being an 'interventionist' when promoting or
installing a human factors program, he should also have eye for the
organizational and cultural aspects of the organization he works with and, in
order to gain acceptance and committment, adopt an interactional approach
as advocated, for example, by Argyris (1973).

Future developments

Human factors training and the development of non-technical skills will be

standard elements of pilot training in all recognized flight academies in the
near future. The European Joint Aviation Authorities are preparing
examinations for Type Rating and Multi-Crew Courses in which not only the
flying skills are assessed but also the non-technical skills. The Western
European Association of Aviation Psychologists is developing standards for
methods used in the training of Cockpit Resource Management skills and for
the evaluation of these training methods. The safety of civil aviation will
benefit from pilots entering the airline with skills, attitudes and motivation
which fit into the human system the airline and the aviation world is.

332
 Non-Technical Training and Education ofAb Initio Student Pilots fo r KLM

References

Argyris, C. (1973), Intervention Theory and Method, Addison-Wesley

Publishing Company, Massachusetts.

Droog, I.A. (1986), De Verkeersvlieger als Manager, Utrecht, The

Netherlands.

Droog, I.A. (1990), Becoming a Captain, paper presented at the First ICAO
Flight Safety and Human Factors Seminar, Leningrad, April.

Mintzberg, H. (1973), The Nature o f Managerial Work, Harper and Row,

New York.

333
Section 4

In-Service Pilot Training

(Photo: Qantas)
22 Introduction to section
4: in-service pilot
training
[Link]

An occupational norm for those involved in aviation is pervasive change.

Air navigation, aircraft types, security methods, baggage handling, duty
hours and rosters, ticketing ... all are subject to ongoing evaluation,
modification, research and development. Such change brings the need for
updating and retraining staff. Inservice pilot training is thus a time-
consuming and expensive operation for the aviation industry. As aspects of
such training, the seven chapters in this section focus on aircrew resource
management training (usually known as Cockpit Resource Management, but
operating under several different titles); airline transition and conversion
courses; training to adjust to changes in technology, and a timely review of
where we stand in preparing pilots for better decision-making.
The section commences with a basic and logical approach to instruction.
Captain Roy Somerville, former Red Arrow and now Senior Training Pilot
with Cathay Pacific Airways, emphasizes the importance of a simple
message in pilot training. With the clarity that normally accompanies
experience and intuitiveness, he traces a chronology of aviation training
from Lilienthal's flight in 1891. He substantiates the view that adaptation to
change is continuous. Such an adaptation required by conventionally-
trained pilots is the need to make optimal use of the flexibility offered by
modem aircraft systems. Somerville's conceptualization of the pilot as an
'Energy Manager' exemplifies the art of the experienced instructor in taking
a complex matter and reducing it to clear fundamentals on which the learner
can build by elaboration.
Trainee-centred instruction characterizes the approach Captain Bob
Sinclair (Air New Zealand Manager, Instructional Systems) has introduced
Aviation Instruction and Training

at Auckland. He has established innovatory instructional rooms which make

effective use of part-task trainers and audio-visual instruction adapted to
individual needs. It is this adaptation which is the central aspect of Sinclair's
work, a reaction against traditional, rigidly structured, ground training
phases of aircraft conversion courses. Student-centred, adult-leaming-based
instruction with formative evaluation characterizes the 747 conversion
course. His message is also seductively simple: the best judges of our
training are the trainees.
The first of the contributors to discuss Cockpit Resource Management
training, Captain Allan Mellor, left Australian Airlines to join Cathay Pacific
in 1988. Following his interest in Australian's (then Trans Australia
Airlines) early entry into CRM, he played a key role in the design and
development of Cathay Pacific's Flightdeck Team Management Program.
That role, his background of teaching and psychology, and his insight into
successive generations of CRM, has made Mellor a popular source of advice
for airlines commencing CRM training. His stage by stage analysis; advice
on the production of training videos; seating layout; and ideas on the
question of whether or not to involve cabin crew constitute a thoughtful and
helpful chapter.
Consultant Tony Wilson is in the highly credible position of having an
active interest in cockpit resource management (by one name or another)
since its inception. When he retired from airline flying in 1989 he was
managing the Aircrew Team Management Program at Australian Airlines
(which was Trans-Australia Airlines when he joined them in 1957). This
experience has enabled Wilson to make an adaptation of crew management
training to the requirements of a supplementary airline, including flight
attendants. His chapter provides the design and presentation of this
prototype, conducted with Kendell Airlines in Australia.
Should an airline buy or build a CRM training program? Captain Dick
Wilkinson, former Qantas Manager, Base Training, provides his viewpoint
as he discusses the need for CRM, building the program, training trainers,
and establishing LOFT. He emphasizes the need for planning so that
evaluation of the program is structured and based on objective data.
Another topical focus is that of cultural difference, which can be
distinguished by a technique Wilkinson provides. The core of CRM is
essentially the same cycle as that of modem flight training, he argues. It
starts with the classroom, then proceeds to part-task trainer, simulator,
aircraft base training and aircraft en route training. CRM has a place in the
larger training environment, being part of new hire training, promotional
training, endorsement training and cyclic training and checking.
The creation and use of a generic non-normal checklist by Aer Lingus is
described by 737 Captain, Neil Johnston, a member of the Aerospace

338
 Introduction to In-Service Pilot Training

Psychology Research group at Trinity College Dublin. Use of this checklist

offers two advantages: it helps to build multi-crew cockpit management
skills and has resulted in an innovatory initial transition syllabus. Because
of its success, this type of course will probably be applied to command
upgrades. The Johnston approach has much to offer: it maximizes training
transfers to a range of aircraft types in a fleet; it conforms to a stringent set
of seven design criteria, and it contributes to problem-solving on the
flightdeck. The use of integrated Line Operational Simulations (LOS) as
well as the traditional Manoeuvre Oriented Flying Training (MOFT) in a set
of eight simulator sessions is an innovatory and successful instructional
strategy.
The distinction between judgement, airmanship, decision-making and
problem-solving is a controversy which has caused confusion, verbosity,
circumlocution and considerable pain (Telfer, 1989). One point of
agreement, however, is that each is vital for aviation safety and pilot
training. The final chapter in this section provides a means of training pilots
in decision-making, as applicable to ab initio training as to airlines.
Judgement and decision-making are central to CRM courses, making this
chapter required reading for those working in that area. Wiggins and O'Hare
provide a carefully-argued and rigorously researched chapter describing a
skills-based approach to training aeronautical decision-maldng. While they
see weather-related decisions as a key area for the majority of pilots
authorized to fly exclusively in visual conditions, the application of the
model is certainly not restricted to that group. Instructors need to provide
practical experience in a variety of relevant decision-making situations.
Only sustained opportunity for actual judgements will assist the
development and maintenance of skills in decision-making.
Despite the excellence of instructional methods and the dedication of
instructors, inevitably some trainees will experience difficulties. What then?
The priority of Somerville and Sinclair would be the trainee. What can we
do to help?
Learning difficulties can be translated as instructional problems. This is in
keeping with a trainee-centred approach in which the instructor's role is to
act as a catylist between the student and the skills, knowledge or attitudes
which are to be acquired. The responsibility for task-attainment is shared by
the trainee and the trainer. It is insufficient for an instructor to cover the
syllabus and consider the job done. Individual differences demand that the
material be adapted for varying abilities and aptitudes.
At first inspection, the problem is difficult. For example, we are told:
'There is no clear method for differentiating how to present material to
remediate a problem discovered after a lesson has been taught from how it
should be presented when taught initially' (Lesgold, 1988, p.l 18). This is as

339
Aviation Instruction and Training

much help as being told that effective training is far simpler than re-training
after bad habits have developed.
A more positive approach is available. '... If the learner has a learning
problem, the problem can be described. If it is described, the remedy is
implied' (Englemann and Camine, 1982, p.310). The basic problems
encountered in instruction can usually be identified and solved by systematic
elimination of possibilities. These possibilities occur at any of the stages of
the instructional cycle (Telfer and Biggs, 1988), such as the preparation of
students for instruction; gaining and maintaining attention; communicating
the lesson; interpreting the trainee responses; providing feedback and
reinforcement; facilitating retention and transfer; or aiding understanding
(Phillips, 1980).
A common reason for trainee failures is their inability to meet the required
standard on an assessment. A first step for the instructor is to decide
whether it is a matter of ability or commitment. The latter calls for
counselling, and the former leads to a check of the test itself to ensure that it
is valid and that the reason for failure can be isolated.
Another valuable instructor response to trainee error can be a check of
communication. The student may not understand what is required. Further,
the rate of learning may be slower than expected. The student could be on a
learning plateau, needing only more practice trials.
The instructor's response to trainee error will depend on whether the error
is isolated or repeated. If isolated, the error is remediated by the instructor
explaining or modelling the correct response and retesting so that the
original assessment is covered. The repeated error required more elaborate
remediation based on whether the error is occurring only in the tested
sequence. If it appears to be a general error, the skill needs to be isolated,
re-taught, then gradually combined into activity clusters of increasing
complexity. If the error occurs only in the tested sequence, similar (but not
identical) sequences can be taught. These should require a similar demand
from a different sequence of activities.
If the fault is the trainee's inability to transfer learning from one situation
to another (say, simulator to aircraft; or navigation chart to cross-country
flight), then the remedy could be for the instructor to demonstrate
similarities in the two transfer situations (simulator and aircraft; and from
map to ground when airborne). If the student is unable to differentiate
between situations then the instructoir has the task of analysing the
situations to show the differences even though they may appear to be quite
obvious. It is the trainee's perception, not that of the trainer, which is the
focus. A student may follow some instructions and perform as required, but
not those in the current exercise. In this instance, the trainer has die task of
teaching the transformation.

340
 Introduction to In-Service Pilot Training

When a student produces a response in some contexts, but needs

prompting in the current exercise, the instructor could undertake the
following three stage strategy:

1. review the various responses needed;

2. rehearse what will happen in the task or exercise with the instructor
monitoring trainee performance and reinforcing appropriate behaviours;
then,
3. gradually reduce the monitoring and expand the application.

It is a valuable, if onerous, exercise to incorporate remedial instruction as

part of the general training cycle. A systematic approach to do so is the
flow-chart model developed by Besco (1992) who provides a five­
dimensional professional performance analysis (PPA) for analyzing and
precribing remedies to reduce pilot error and improve operations. Such an
approach would also lead to a reflection of the issues and options discussed
by the contributors to this section, and the possibility of a refinement of
training methods for greater effectiveness.

References

Besco, R.O. (1992), 'Analysing knowledge deficiencies in pilot

performance', The International Journal o f Aviation Psychology, vol. 2,
no. 1, pp. 53-74.

Engelmann, S. and Camine, D. (1982), Theory o f Instruction: Principles

and Applications, Irvington, New York.

Lesgold, A. (1988), 'Toward a theory of curriculum for use in designing

intelligent instructional systems’, in Mandle, H. and Lesgold, A. (eds.)
Learning Issues o f Intelligent Tutoring Systems, Springer-Verlay, New
York.

Phillips, D.C. (1980), 'What do the researchers and the practitioners have to
offer each other?', Educational Researcher, vol. 9, pp. 17-20.

Telfer, R. (1989), 'Pilot decision making and judgement', in Jenson, R. (ed.),

Aviation Psychology, Gower Technical, London.

341
Aviation Instruction and Training

Telfer, R. and Biggs, J. (1988), Psychology and Flight Training, Iowa State
University Press, Ames.

342
23 Aviation training and
instruction
R.E. Somerville

Lichterfelde, Germany, 1891. The weather was perfect, a light breeze

blowing up the gentle slope which had been specially built. There was an
atmosphere of hushed expectation amongst the small group of onlookers and
assistants. The designer and builder of the strange looking craft was, himself,
checking the wires and fabric before he attached himself to the wooden
structure. Everything was ready and it looked good. It should be, after all he
had been designing and building these craft for nearly twenty five years. This
one was different though. It was the first he had fitted with a flight
management computer which he had laboriously programmed over the
preceding years. He was happy with the basic software. With a brief wave
to the group of friends and servants watching, he ran a few steps down the
slight rise. The fabric taughtened and took his weight. Otto Lilienthal's feet
left the ground.
Soon the butlers and servants of the landed gentry were being coerced into
similar contraptions, encouraged to supplement their meagre earnings by
testing the machines built by their employers. The countryside of Europe was
resounding to the Teutonic and Anglo Saxon words of encouragement and
shotgun assisted launches of countless numbers of subsequent flights.

A little later. There was a short period from the late 1890s to the early 1900s
when aircraft designs attempted to mimic our feathered friends in stability
and control. Although warping or bending of wings and tail structures
produced some effective form of control, it was very difficult to both build
and use precisely. The results were, more often than not, hilarious rather than
practical. This method was therefore relegated to being strictly 'for the birds'.
Inspiration was however, alive and well. It rapidly became apparent that
separate, movable, control surfaces were far more effective and easier to use.
Aviation Instruction and Training

The internal combustion engine was teamed up in this rather eccentric

combination of elegance and noise to produce what would still be recognized
as a basic flying machine.
How did the pilots who flew these machines leam to fly? By accumulated
personal experience if you were lucky enough to survive. Instrumentation
within the cockpit was minimal and not really required as flying these
machines was a purely visual experience, plus of course, the necessary
musical ability to recognize the subtle nuances of tone of the wind in the
rigging wires. With the advent of two seat aircraft it soon became possible to
pass this knowledge on by demonstration and physical experience. Flying
was with us to stay and the art of flying instruction was bom. The rest is
history. Here we are 100 years further along the road of technical
development. Aviation is now an integral, accepted part of every day life.
The cost has been enormous.
This is a good point at which to introduce a question relevant to our look at
aviation training, as it was put to me.

What are our training requirements for the 1990s?

I would like to rewrite that question and ask which nineties? The 1890s,
1990s or the 2090s? I would suggest they will all have a lot in common.
Alternatively we could ask simply, where do we go from here?
To look forward effectively it is imperative that we look back, analytically
and constructively. What lessons have we leamt? Is there a need for change?
Over the past one hundred years, our flying machines have developed
probably faster than any other human achievement, to the extent that their
present complexity demands very specialized training of those people who
will operate them. It is fair to say that as far as basic flying training is
concerned a machine like the Tiger Moth is as an effective trainer today as it
was sixty years ago? Its simplicity is the key! This is a point worth bearing
in mind when trying to teach a new skill, however complex it will
subsequently become. But back to looking over our shoulders for a moment
to see what we have learned. Let us first look at the machines and the
demands they placed on their pilots and crews.

The machines

There are many books written about the evolution of the aeroplane. It is not

344
 Aviation Training and Instruction

my intention to cover this aspect in depth, but merely to draw our attention to
the major changes and their subsequent effects on the requirements of
operating them.
The early aircraft designs were constrained by the materials available, and
relied heavily on wood and fabric: materials which lent themselves very well
to the lower end of the aerodynamic performance scale. Aerodynamics as a
science was still in its infancy. The theories of people like Bernoulli and
Lanchester were still being integrated slowly into the overall concept. The
scientists, engineers and builders worked separately, but in increasingly
overlapping areas. The pilot was the only person to be given the whole
concept to manage as a complete package. His initial appraisal was
frequently that if it looked right, it would fly well! The major changes were
happening in engine design. Power output increased rapidly and soon the
wood and fabric was changed to lightweight metal structures to withstand the
greater stresses placed on the aircraft by this increased power and speed.
The pilot stayed the same. Whereas in the early machines pilots really could
get by with a few well practised simple control movements, pilots now
needed to learn some quite demanding skills to deal with, for example, engine
torque. The pilot also needed to take a lot more care of the engine by
monitoring more and more instruments, as well as the pure flying instruments.
The culmination of these skills are probably best realized in the beautiful
series of coordinated control inputs required to get a large, powerful single
engined, tail wheeled aircraft into the air.
By the early 1950s we had reached the culmination of these powerful,
extremely complex piston engined aircraft. The Lockheed Super
Constellation, powered by four compound 3500hp, Wright radial engines was
among the ultimate manifestations of 'the complicated aeroplane'. The vast
array of instruments, requiring constant monitoring needed a crew of four to
perform the task!
The jet engine and the implementation of quantum theory in the shape of
computors and microprocessors was about to change all that! The jet engine
was an inspired design of great simplicity. It's capability for development has
met the power requirements of the most demanding airframes to date. It still
appears to have more to offer, however we are reaching the limits, presently
imposed by existing materials technology. The basic engine itself still retains
most of its simplicity. Efficiency and power have been matched to equally
impressive developments in airframe design resulting in the production of
both civil and military aircraft of amazing performance and economy.

345
Aviation Instruction and Training

The pilot

I will deliberately avoid asking the question, 'What is a pilot?' to avoid the
dozens of definitions that readily spring to the minds of all those who are not!
We do need, however, to define quite clearly what the pilot's role is and how
he or she integrates into the total concept.
My own simple concept of an aircraft pilot would apply to anyone
controlling a machine with freedom of movement in three dimensions, that of
an 'Energy Manager'.

Figure 23.1
The pilot as an 'Energy M anager'

Our pilot sits, if you like, in the middle of the Thrust/Drag and Lift/Weight
vectors, allocating within the range of control he or she has, quantity to these
vectors in order to achieve the required task. The two vector quantities that

346
 Aviation Training and Instruction

the pilot has most control over, once airborne, are Lift and Thrust. Pilots
have limited control over Weight and Drag although, as we know, the
majority of undesirable Drag is Lift induced.
In addition to 'managing' energy vectors, the pilot has to navigate this
collection of associated forces from one point to another, in a safe, efficient
manner to achieve a viable commercial result. We are starting to ask quite a
lot of our 'Energy Manager'.
In the early days of aviation our pilot was able to carry out most of these
coordinated tasks alone, due to the simplicity of the machine/engine
combination. The pilot has always needed, and will continue to need, a
thorough understanding of the basic relationship between the energy vectors.
This was, and still is, achieved by the use of simple, single engined machines.
The recipe for successful 'Basic Hying Training’ has evolved over the years.
It remains as simple and valid as it always has done, namely, a thorough
understanding of what is required, followed by a demonstration and then
repetitive practice until competence and confidence are achieved. At this
task I feel the industry as a whole is very successful. No changes need to be
made. This produces our basic pilot or 'Energy Manager’. Let us now put the
pilot's acquired talents to use!
At this point I would like to digress a little. A fact of human life is that we
are not very good at doing more than one thing at a time, no matter how
simple. For example if we are required to read and understand information,
simultaneously performing a calculation, we have a problem. For those who
doubt the validity of this statement, please read this page aloud, to a
colleague, whilst writing down the 2 times table. No doubts left I hope! We
can of course play the piano whilst watching and understanding a movie for
example. A fact that was put to good use during the silent film era. Both
these examples are, on their own, quite simple tasks. We are very good at
our 2 times table. We may be very good at playing the piano. Doing both
simultaneously is a virtual impossibility. It is somewhat divorced from our
subject but is important that we understand quite clearly, the limitations of
our brain if we are to train people to use it efficiently.
Back to our aircraft, the complicated, long range piston engined aeroplane
like the Lockheed Constellation. The pilot needed assistance with his energy
management.

The crew

To keep the hordes of pistons and rotating parts working harmoniously and

347
Aviation Instruction and Training

efficiently for hours at a time, our pilot or Captain was given a 'Thrust
Manager' who came in the shape of the Flight Engineer. The complexity of
the engines and the associated systems demanded the sole attention of this
crew member. He was a welcome addition for his detailed knowledge of all
things mechanical.
To calculate and continuously update the position of the aircraft, the
Captain was given a 'Position Manager' in the shape of a Navigator. A
compex task requiring very specialized knowledge and calculating skills.
Any spare capacity was put to good use to calculate fuel requirements and
diversion possibilities, for example. The Navigator was also a very welcome
addition to the crew.
To provide both physical and ergonomic assistance in controlling the
aircraft, a co-pilot or deputy Captain was carried. Routine position reporting
and radio usage was a job often shared amongst the crew. As we can
imagine a lot of coordinated effort was required by the crew to operate these
machines. A side bonus was the 'spare brain power' available for problem
solving.
All this has changed! The development of an accurate inertial gyroscope
coupled with a computor to do the calculations, rapidly led to the demise of
the Navigator as a crew member. The pilot now shouldered the task of
managing the navigation systems.
The advent of the jet engine and its simplicity marked the start of the
demise of our Thrust Manager, the Flight Engineer. Although the engines
and the controls themselves were now far simpler, the power they produced
was far greater, as were the associated performance options such as altitude
and speed, that came with the 'Thrust' package. The pilot also took on this
task.
For safety's sake and system redundancy, two pilots were kept in all large
commercial jet transports. The information processing and system
management was now an enormous task for our two, lonely pilots. Thanks to
the computer and its relative the microprocessor, artificial intelligence was at
hand to assist in our 'Energy Management'. This came in the shape of the
Flight Management Computer (FMC). The early 1980s marked the beginning
of the technological explosion in the use of computers and digital
instrumentation in aircraft. A new era in aviation and begun. Or had it?
In the opening paragraph I lightheartedly mentioned the FMC fitted to Otto
Lilienthal's aircraft - the first manned aeroplane. I was of course alluding to
the FMC he had fitted between his ears. We are now up to date with
development and control of our modem aircraft. We also have had a brief
look at past methods of dealing with system management. We now have

348
 Aviation Training and Instruction

artificial intelligence to assist us. This has proven to be a difficult change for
most pilots to accept. Why? Let us rationalize the problem and try to identify
a solution.

What has changed?

Like most things that evolve, the changes that have taken place in the cockpit
have been gradual rather than revolutionary. Taken in isolation our 'Energy
Manager' has been able to cope very well with these changes. Digital
instrumentation has, after all, been with us for many years. Flight
Management Systems have also been slowly evolving and were fitted to
many aircraft of the 1960s. To cope with these developments extra training
was 'tacked on' to an already heavily loaded training syllabus. It was a bit
like building a bridge across an ever widening gap by sticking bricks
together, end on. It was easy to see how many bricks were needed but
without a central support something was going to give!
The physical properties of our aircraft changed less, the quality of the
hardware certainly improved and became, from a technical standpoint, more
reliable. The radical changes, although more subtle in their appearance, were
in the access and control of the hardware. Traditional training methods were
certainly well proven and had produced very competent, safe operations for
many years. It was around ten years ago that the Flight Management
Computer started to be widely fitted to new models of existing aircraft and
somehow we reached a watershed in aviation development.
The pace of development was fast, coupled with a reduction to just two
pilots on nearly all models of aircraft. This completely integrated, computer
assisted cockpit had not really been properly trained for and a rash of rather
serious incidents, happening to perfectly serviceable aircraft took place on
both sides of the Atlantic. There was, and to some extent still is, a
communication problem between man and machine. Somewhere in our
headlong rush into technical inovation we had forgotten the human being who
was going to use it, our 'Energy Manager'. The 'Energy Manager', who is the
common link with Otto Lilienthal hasn't changed, and not enough time had
been spent re-programming the 'Energy Manager’s' software!
The stark simplicity of the modem cockpit, here illustrated in Figure 23.2
by the B747-400, belies the amount of information it is offering the pilot
This fact, coupled with either a complete distrust of, or a belief in the total
infallibility of the computer, can lead to an air of controlled confusion in the
cockpit, unless we carefully train people how to use this artificial intelligence.

349
350

Figure 23.2
B747-400 glass cockpit
 Aviation Training and Instruction

When used correctly, it is a great improvement on previous cockpit

technology, however a greater knowledge of its capabilities and limitations is
needed. By using an autopilot in conjunction with the properly used FMC we
are in effect, back to a three man crew with better equipment. Simplicity of
operation is the key to this happy but elusive state. KEEP IT SIMPLE!

Human factors

The airline industry is driven by the twin factors of commercial viability on

the one hand and safety on the other. Profitability is a must for without it,
there is no useful industry. Safety is a must for without it, there is no useful
industry. Please forgive the play on words, however here we have the
elements of an equation. Instead of being in conflict these qualities must be
in perfect balance. Profit is wonderful, especially for those participating at
harvest time. It has a tangible quality. Flight safety on the other hand is
intangible. It is only when there is no safety that it makes it's presence most
felt! It then becomes, certainly in this industry, sensational. Because of this
quest to track down this invisible but essential quality, it has been put under
the microscope.
It is now an accepted fact that the vast majority of incidents, accidents or
failures, call them what you will, have at their roots human failure. In the
past it was all too easy to point the finger at the pilot, he after all was in
charge of the overall concept. Just like our friend Otto Lilienthal, the pilot
was most visible. Heaven forbid that Herr Lilienthal's demise was caused by
woodworm in the structure supplied by a carpenter trying to maximize his
profit!
Today the responsibility is collective even though the cause may be
isolated. To come to this enlightened realization, much inspired research was
earned out, identifying frequently, a lack of ability to communicate across the
various disciplines or even more significantly, a failure to use our existing
human resources, to their best advantage. Fortunately research into the
functioning of our brain and thought processes was also fast evolving. A new
discipline with the title of 'Human Factors' was created. Fed by the enormous
amount of research into human behaviour, it became very apparent that we
had neglected this aspect of our training for too long. Fortunately the trend is
now reversed and many specific areas of awareness are now being addressed
with titles such as Crew Resource Management, Flight Training
Management, Line Orientated Flight Training, and so on.
All this should really have been nothing new, after all we are talking about

351
Aviation Instruction and Training

the ability to both communicate and utilize human resources to their best
advantage, both in and out of the cockpit. It somehow had escaped our
attention for too long and needed re-emphasizing in a more formal way.
Human Factors as a training tool is now firmly back where it belongs, as one
of our most potent aids to effective training.
A word of caution however at this stage. From my own perception there
appears to be an industry wide, over-emphasis on the curative powers of this
rediscovered quality, a little like the 'flu vaccine'; it is very useful but only
when used in conjunction with other curative training procedures. If we are
not careful we may find ourselves snowballed due to the mass of apparently
relevant information, to the exclusion of something less evident. Somewhere
in here I feel, we should find a place for Cybernetics, which my dictionary
defines as 'the comparitive study of automatic communication and control in
functions of living bodies and in mechanical electronic systems (such as in
computers)'. As it is a twentieth century dictionary, I leave the reader to form
his own opinions.

Effective training

The above term is one I like very much. It implies a useful end product, the
only rationale for any form of training. How do we achieve this? Most
people can be trained to do most things providing the training is pertinent and
clearly understood. To be effective it must be appropriate to the skill being
taught. We have at our disposal a host of modem teaching aids and we have
also identified clearly any areas that need enhanced emphasis. The basic
structure of most airline training courses would be very similar, something
along the following fines, in order of progression:

1. Human Factors awareness course and training goals.

2. Technical training of specific aircraft systems with the emphasis on
how and why they work.
3. Cockpit procedure trainers and sub system, hands on training aids.
FMC's and Autopilot/Flight director systems for example.
4. Flight Simulators. The modem generation of these machines are
virtually indistinguishable from the aircraft, from the pilots viewpoint. It
is here where the tried and true methods of teaching such skills as
engine failures on take off, engine out landings and missed
approaches can be practised and taught until confidence and competency
are achieved.

352
 Aviation Training and Instruction

5. Line Orientated Flight Training (LOFT). Using the simulator to achieve

an overall, integrated level of competence, such as would be required
on a normal line flight.
6. Aircraft Line training under the supervision of a Training Captain.

All of the above will only be as good as the level of competence of the
training staff doing the teaching! This is where the real change of emphasis
needs to be placed in modem training terms. I do not wish to imply that there
has been a change in standard but merely of emphasis. In the past a great
number of facets of the operation tended to be taught in isolation. This led to
a rather dogmatic, tick in the box, method of training. It worked to a point
but has now been superseded by the integration of so many functions, by the
FMC - back to our 'Energy Manager'. The FMC, by its very nature integrates
so many functions that it now allows many different options in how we wish
to allocate quantity to our energy vectors. In other words it has provided the
pilot with greater flexibility of operation. Flexibility is something that pilots
need to be taught. With increased options comes the need to understand how
they all work and consequently a greater depth of knowledge. Dogma was
fine when the options were not available. As the manufacturer's pilot said to
me on my own line training, 'virtually the only limit on the operation is your
own imagination'. A nice statement but hard to grasp for an old
conventionally trained pilot like myself. It came but slowly. Until we have a
generation of pilots who have learned to fly, from ab initio to airline
command on computer controlled aircraft, we are going to have to allocate
more time during training to teaching and most importantly, accepting, the
flexibility our modem systems offer. Once again the key is KEEP ΓΓ
SIMPLE!

Looking ahead

There is no doubt that many very significant changes have occured in recent
years to the aviation industry. I would suggest most of them have been
evolutionary rather than radical. The problems that have arisen have been
more of a failure to communicate the nature of change to those using it, than
an inability on behalf of the users. Due to the physical nature of aircraft their
basic format will not change. The 'Energy Manager' of the future, be it
human or computer, will still be doing nothing other than allocating quantity
to our four basic forces involved. The significant changes will be, and have
been, in the access and control of them.

353
Aviation Instruction and Training

I feel that for the forseeable future there will be a place in the cockpit for
two pilots. Complete automation is totally feasible even with our present
technology. If we can send a craft to the moon and retrieve it then certainly
London/New York should not be a problem! There are, however, more
decisions to be made, to do with options and choices, between London and
New York than there are between here and the moon. If our automated moon
traffic were to increase to trans-atlantic proportions with the added features
of en route weather, hijackers, inflight bomb warnings, passengers suffering
heart attacks and the like, I would suggest the first scenario would be to put a
human back in charge.
As it would appear we are in the cockpit to stay, it is here that the most
important decisions affecting the man/machine interface will have to be made.
Human beings by nature are a participating species. Even though we build
the most sophisticated automatic control devices, we actually still want to be
in control of the big picture. I feel the greatest challenges in the future will be
to make our access to control more user friendly. Our brains and senses
definitely react more favourably to tactile feedback than a digital readout of
the result. A movable thrust lever, although a little more cumbersome is far
more effective in providing us with a feel for rate of change than a simple
multi function switch. I will not presume to have any expertise other than as
an end user in this aspect of design but I do feel that if we are to remain in the
cockpit we have to be actively involved rather than clinically remote. It must
also be accepted that an error free environment is almost impossible to
achieve and maybe, not even desirable. Error is a form of feedback which
when recognized as such heightens reaction to subsequent events. It also
gives us an indicator of our own and the machine's operating limits. Once
these are known we will hopefully not allow ourselves, or the machine to
overstep them.
This is all very philosophical. I do find however that the longer I spend
flying, it is more about an attitude of mind than physical control handling.
Broadly speaking, as pilots we all share roughly the same degree of physical
flying skill. How we tend to apply it is however totally dependent on the
individual attitude. It is not so much a question of what type of aircraft we
flew in the past but what did we learn from what we flew!
Regulatory authorities and commerical departments of airlines both have a
long way to go in deciding how much exposure is required for retention of
competency and how much that will cost.
As pilots we need to adapt to the inevitable changes and use them to our
advantage. We must learn to use the increased options and flexibility that
this technology gives us but not allow it to dominate our common sense. We

354
 Aviation Training and Instruction

can all still enjoy that same thrill of participation as when we first flew, after
all that's what motivated us in the first place. To keep that clarity of
perspective maybe we need to do what Otto Lilienthal might have said,
'KEEP IT SIMPLE’.

355
24 Conversion courses
and students' learning
needs: a new approach
for Air New Zealand

B. Sinclair

In many airlines, the ground training phase of aircraft conversion courses is

rigidly structured and does not readily lend itself to the accommodation of
differing student learning needs. My airline recently reviewed this phase of
its Boeing 747-200 Conversion Course and effected changes to the content,
structure and facilities. Throughout the review, the need to address the
students' individual needs was prominent in the minds of those people
charged with the revision. The package which emerged has now been tested
on five courses and subjected to exhaustive critiques. As a result, it has
been modified extensively to a form which has allowed a significant
reduction of the time - and cost - of the overall Conversion Course, and it
appears that the graduate competency level is considerably higher. This
result seems to indicate that the differing needs of individual students have
been met to a greater degree. In the hope that the reader may find something
of interest, this article describes the project and its outcomes.

Previous ground training

On the first day of the course, the Ground Instructor gave a brief
administrative outline of the Systems phase and then launched into the first
of a series of audio-visual modules, which were controlled from a position at
the front of the classroom. The modules consisted of slides with
 Conversion Courses and Students' Learning Needs

accompanying taped narration. At audible, tape-signal prompts, the

instructor changed the slides. If required, the narration could be stopped to
allow for elaboration, discussion, or the answering of programmed
questions. Students answered the multi-choice questions by pressing a key
on individual desk pads. The instructor recorded results and used them as
an indication of each student's progress, and of the overall success of the
program.
Life-size, backlit drawings of the flight deck panels were mounted on the
walls of the classroom. These enhanced student appreciation of the
appearance and relative positions of System components. Use was also
made of several, large, free-standing, training panels which each allowed
partial operation of a major System. Brief sessions with these provided
opportunities for some hands-on familiarization with the systems. Near the
end of the course, there were two four-hour periods in a fixed-base
simulator, during which the systems were seen to operate. Once again,
students had some hands-on experience. Finally, just prior to the Systems
examination, the Ground Instructor conducted a two-hour tour of a real
B747-200. The course allowed for twelve students but, however many there
were on the course, the number and duration of these practical activities did
not change.
Learning was augmented by home study of the Systems chapters in the
Flight Manuals and, after fourteen working days, a three-hour books-open
examination ended the Systems phase of the Conversion Course. Aircraft
performance management training, emergency equipment and procedures
training, inertial navigation system training and flight planning followed.
These took a further twelve days. Then, students entered the full-flight
Simulator stage of training. In this phase, there were thirteen four-hour
sessions conducted by simulator or flight instructors. The students were
ready for the single, circuit training flight eleven weeks from commencement
of the conversion course.

Shortcomings of the original course

Widely expressed views were that the competency levels of the graduates
were too varied; the duration of the course seemed to be excessive, and the
costs exorbitant. Furthermore, conversion courses for the company's other
aircraft types made use of a more sophisticated audio-visual system and
management required the B747-200 course to do the same. The process of
meeting this requirement was expected to be protracted and provided the

357
Aviation Instruction and Training

Flight Operations Manager - Standards, with the opportunity for a review of

the overall course. As a result of initial exploratory meetings of interested
parties, the writer, in an instructional design and development role, worked
with a management group and teams of subject matter experts, and
flightdeck crew, with the objectives of analysing the problems, prescribing
changes, and effecting the necessary modifications. Investigations, into
alternative or additional components in the course, included a review of
training processes in other airlines, industry and academe. As a result,
several shortcomings were identified:

1. The first problem was that the course components did not flow into one
another. Not only were the different phases carried out by different
instructors, but these people worked in different sections of the
company. For example, the Ground Instructor was employed by the
company's engineering department and the instructors in the simulator
phase were full-time Simulator Instructors or Flying Instructors. There
was a gap between terminal knowledge and skills at the end of ground
training and instructor expectations at the beginning of simulator
training.
2. A similar problem appeared to exist in the transition from Conversion
Course to route training. Furthermore, the sequence of components did
not follow a logical pattern for student development. There was no
person carrying the responsibility for coordinating the overall course
structure and components. One effect was that, as the students
progressed from one phase to the next, a change of instructor involved
the process of their getting to know each other.
3. Another problem was that, instructors had little knowledge of what
came before and what happened after the phase with which they were
personally involved. There was a need for coordination of components;
instructor involvement with more than one phase of the course; a
reduction in the number of instructors, if possible; and the
incorporation of 'advance organizers', or helpful transitions, to orientate
students through each course component change. It seemed logical to
assume that, if these needs were met, a reduction in course duration
was possible.
4. The major problem area, however, was the inadequacy of attendance to
individual, or differing, students’ learning needs and viewpoints.
Students' perspectives needed greater consideration. It became
apparent that, there was insufficient student participation in the learning
process; in reinforcement of learning; and in assessment processes

358
 Conversion Courses and Students' Learning Needs

which would identify learning problems. This was partly because there
was no evaluation from this standpoint, and partly because of the lack
of a student critique of the course.

Changes and additions to the course

A number of course components were altered or added. These will be

described individually and, later, the changes to the overall pattern will be
outlined. Many of these were developments of examples seen in other parts
of the company, or in other organizations, or airlines, notably the Hong
Kong based, Cathay Pacific Airlines.

The 'Spiritus' audio-visual system (AVT)

The basis of the more sophisticated audio-visual system, 'Spiritus', is that it

is a Self Paced Individual Response Instructional Tutorial System. My
company uses it for the two or three people who make up a crew. In the
case of the B747-200, the ideal situation would be for the crew to be made
up of a student captain, a student first officer, and a student flight engineer.
However, the make-up of the crew will depend on the roles of the students
who are to be trained. The crew may be two captains; two first officers; or
any other combination, according to demand. The important thing is that it
is a very small group. There is no requirement for an instructor to be
present. On the contrary, it is considered desirable that the students are free
from having an instructor looking over their shoulders. However, it is
important that there is access to an instructor, so that learning problems can
be rectified. The crew sits in front of a pair of screens and the tape is
mounted in a computer unit. At hand is a panel which allows student control
of the flow of the lesson. On the tape is a second track which has a coded
signal that provides a programmed sequence of slide selection in accordance
with the narration.
Student interaction is such that the students can stop the tape or cause it
to proceed. They can move the tape forward and back and can see earlier or
following slides. Multi-choice questions are included in each module and
the control panel is the medium by which the students answer questions.
Students can wear headsets which enables a number of cubicles to be set up
in the same room. A problem with this is that ready communication between
members of the same crew is inhibited.

359
Aviation Instruction and Training

Introduction o f the carrel

The B747-200 Conversion Course is now based on the principle that each
crew has a separate, sound-proof room, or carrel. The ancient term 'carrel'
applies to an enclosed, small-group, learning environment. This obviates the
need for wearing headsets and the crew members are encouraged to
communicate: sharing problems or knowledge and entering into discussion
at will - even over the questions. The underlying philosophy is that adult
students control their own learning and are free to do so in their own way.
Because the crew has its own room for viewing the fifty-one modules, the
environment is enhanced by being constructed as an expanded, mock flight
deck. The backlit panels have been set up in the centre of the room, and the
students take up flight deck seats relative to these panels. The 'Spiritus'
screens are mounted where the windscreen is normally found, and the
control panel is at the rear of the central pedestal. Each student has a small
desk on which flight manuals, and the like, can be placed. Air-conditioning
and controllable lighting make the windowless carrel comfortable.
There are two identical carrels and each is self-contained, having a
complete set of modules. When a crew is in session, this is indicated by an
external, illuminated 'In Use' panel, to minimize disturbance. Beside each
student is a whiteboard and markers. If any learning problems, or faults
with the modules or equipment are encountered, comments are recorded for
future reference. The carrel is the crew’s base for the duration of the
Systems training phase.

Student copies o f scripts

Each crew member has the scripts of the narrations for all modules. The
scripts are designed to have a segment of narration associated with each
slide. The student scripts are annotated with both the slide numbers and the
tape count at the beginning of the appropriate segments of narration. During
normal operation, the student is absorbed by the narration and the slide
pictures. However, to resolve any confusion, to check on points, or simply
to review material, any part of the program can be identified and the
pertinent slide, or section of the tape, can be called up. At the end of each
day, the answers to the questions are issued for addition to the scripts, which
are permanently retained by the students.

360
 Conversion Courses and Students' Learning Needs
Instructional tutorial sessions (ITS)

Each day, the students view four or five modules and mark up comments or
questions on the whiteboards. At a programmed time, the crew's instructor
joins them in the carrel. Forty-five minutes is allocated for a tutorial,
interaction session. The instructor will go through the items noted on the
whiteboard and question the members of the crew to establish knowledge
levels on the material which has been covered in the modules. In particular,
the instructor will determine whether there are any members of the crew
who are not comfortably abreast of the program. Finally, the crew will be
prepared for a two-hour, hands-on session in the fixed-base simulator.

Fixed-base simulator sessions (FBS)

Each day, following viewing of a number of modules, and an instructional

tutorial, the crew spends two hours in the fixed-base simulator. This session
is a hands-on conclusion to the day's coverage of aircraft systems and
includes starting up, operating, and dealing with non-normal functioning of
these systems. This practical session, together with a thirty minute debrief,
brings the day's learning activities into the operating environment.

Home study

At the end of the day, the remaining activities make up a period of

consolidation. Firstly, the scripts are reviewed. Then, the appropriate
chapters in the flight manual are read. Finally, a study guide, made up of
questions answers and references, is thoroughly addressed. This prepares
the students for the first exercise on the following morning.

The daily quiz

Each day begins with a books-closed quiz on the previous day's learning. In
fact, the quiz is made up of ten questions from the relevant section of the
study guide. Besides offering an opportunity for feedback on the success of
the learning, the questions are selected to provide an indication of those
things which should be well learned. A training assistant presents the
written questions and collects the answers for marking by the crew's
instructor. Then, at the beginning of the tutorial session, the answers are
discussed. The quizzes are simply an aid to learning and the results are not
recorded. They also provide instructors with indications of student progress.

361
Aviation Instruction and Training

Aircraft systems panels

The aircraft systems panels have been placed in a dedicated B747-200

classroom where they stand around the walls. Increased use of the fixed-
based simulator has reduced the use of these panels. However, the electrical
system panel operates very well and is used on the first audio-visual day, in
lieu of the simulator. Furthermore, the panels are available for free use by
the students who can operate them at will. Another use for these panels is
for general discussion of specific learning problems; for briefings and
debriefings; and for reference during the Systems exam, which is held in
this room.

Instructor utilization

The simulator instructors, who have all been B747-200 flying instructors,
have become AVT/FBS qualified. This means that they can supervise the
audio-visual sessions; conduct the instructional tutorial sessions; and
facilitate the panel and fixed-base simulator sessions. They have also
become qualified Aircraft Performance Management and Inertial
Navigational System Instructors.
The Flight Engineer Instructors deserve special mention. Few pilots,
today, have experience of crews which include a flight engineer. It has been
found that experienced pilots have some difficulty in adapting to the
incorporation of these specialists in the crew complement. To adapt pilots
to the flight engineer concept; to make use of these specialized instructors
for specific aspects of the learning process; and to have a flight engineer to
assist the simulator instructor with presentations and act as crewman; Flight
Engineer Instructors are a part of the overall team right from the beginning
of the course. These instructors are a fount of extended technical knowledge
for answering or researching difficult questions; they conduct visits to the
engine shop, and to the aircraft for an extensive 'journey of exploration'; and
facilitate learning of particular tasks, such as defining and reporting technical
faults.
Coincidental with the review, a restructuring of the airline had one result
which was of benefit to the Conversion Course. The Ground Instructors
involved in crew training were moved to Flight Operations. The B747-200
Ground Instructor was then given the responsibility for overall coordination
of the ground training for the course.

362
 Conversion Courses and Students' Learning Needs

Student numbers on courses

Training management agreed to a policy decision to restrict the number of

students to two crews per course. This allowed comfortable insertion of
conversion training into the simulator utilization programs. Also, feeding the
fleet with the student output from courses of this size avoided bottlenecks in
route training. Even in periods of high demand for new crews, frequent
running of small courses provided a better flow of qualified personnel into
the line operation. Therefore, the expense of having partly trained and
highly paid crew sitting about waiting for further training was avoided.
Significantly, the experience of other airlines confirmed the opinion that
more effective and appealing training would result from small, more
personal, courses.

The course critique

The company has been fortunate in securing the services of a university

graduate who has chosen to use the development of a course critique as the
thesis for a Master's degree. Comprehensive and valuable critiques have
been separately developed for each of the conversion courses for the
different aircraft types. These critiques are being applied, following
completion of the courses, and are valuable for summative evaluation.
Significant contributions for improvement are emanating from the critiques.

The new course

The B747-200 Conversion Course is now completed in seven weeks. Two

crews, each consisting of two pilots, or two pilots and a flight engineer, are
presented for circuit training which can be completed in one aircraft detail.
Route training normally follows without a major break in all students' duty
rosters. The fleet training captains are receiving trainees who are adequately
prepared, and are of a reasonably consistent standard. So, let's look at the
course as it is now constructed.

The program

On the first day, students are addressed by the B747-200 Fleet Manager
who gives them a picture of their new role and life pattern. This is followed
by a run-down on the Conversion Course. It is explained that the ground

363
Aviation Instruction and Training

training begins with sixteen days of AVT/FBS on Aircraft Systems

Management, Aircraft Performance Management, and technical aspects of
the Navigational System; that this phase is followed by ten, four-hour, Full-
Flight Simulator (FFS) details; that, during this phase, the focus is on basic
aircraft operation; and that the course finishes with three days in which the
focus is on the use of the aircraft as an airliner - that is, with emergency
equipment and procedures training; flight planning; and the operational use
of the navigation system. Then comes a detailed explanation of the course
program. Students are also informed that each crew will have its own
simulator, AVT/FBS instructor, and that the two crews share a Flight
Engineer Instructor. If possible, these instructors will remain with the same
crews for the entire course, apart from the emergency procedures and flight
planning components. Furthermore, except for visits to the engine shop and
the aircraft, virtually all training will be completed in the same building.
Still on the first day, students are introduced to the carrels and 'Spiritus'.
Viewing of two modules of Aircraft and Systems introduction provides an
overview of the Systems phase of the Ground Training. Then follows an
overview of Aircraft Performance Management. This is oriented towards
giving the 'why' to the technical training, and facilitating appreciation of
procedures.
The overall emphasis on this first day is on overview; on providing an
advance organizer; on presenting a skeleton to be fleshed out in the
subsequent training; on settling the students down in the training
environment; and on orienting their attitudes to the task ahead.
The following morning, the students begin operating as crews. The
timetable is built around the daily allocation of a four-hour period in the
simulator. One crew will use the first two hours for FBS training, and the
second crew will use the remaining two hours. Thus, the first crew starts
work two hours before the second crew. On this day, the electrical panel, in
the classroom, will be used rather than the simulator, but, on the following
ten days, the simulator will be the venue for the hands-on training. A
detailed lesson plan for the FBS periods has been developed to ensure that
all requirements are met in a smooth-flowing sequence. To the same end,
the viewing of AVT modules has been programmed to ensure that students
are fully prepared for each FBS period so that no items in the AVT and FBS
sessions are out of sequence. Each day consists of viewing four, or five
AVT modules, a tutorial, an FBS session, and a final debriefing. Home
study gets student noses into books for reviewing the scripts; reading the
Systems chapters in the flight manual; and working through the study guide
questions, answers and references. A ten-minute quiz, on the day's Systems,

364
 Conversion Courses and Students' Learning Needs

will be held at the beginning of the following day. During the first few days,
the technical, or Systems, training is focused on developing a knowledge of
the Systems and their normal operation. Later the emphasis shifts towards
the non-normal operations, as these sometimes involve more than one
system, with which the student will then be familiar.
The engine shop visit and the technical aspects of the Inertial
Navigational System are also programmed into these first days. Following
the AVT/FBS days, students spend a full day with the Flight Engineer
instructor. On this day, detailed attention is given to the ways in which the
flight manual is used in flight: how information is found; how problem
solving is carried out; and how technical faults are described and reported,
and to whom. A discussion is held with a Station Engineer, a specialist of
the type that the crews will communicate with at the various airfields. Then
comes a visit to Maintenance Watch which coordinates fault reporting and
problem solving. Finally, the group visits the aircraft and locates the System
components, and sees their relative disposition. They will see nooks and
crannies for the first and, very likely, the last time.
On the fourteenth day, comes the Systems examination. This is a three-
hour exam in the classroom. Students are required to work individually, but
may use the flight manual and the aircraft training panels around the walls.
The principle is that these things are available to them in their working
environment There are too many of the multi-choice questions, however,
for all to be found in the books in the available time. Thus, a good technical
knowledge, and ready familiarity with the flight manual, are needed to get a
satisfactory pass mark.
Later in the same day, the Aircraft Performance Management instructor,
who will probably be Simulator Instructor for one of the crews, will cover
the training associated with the disposition and security of the aircraft's
payload. Then, on the following two day's, the remainder of the
Performance program will be completed. This includes a Performance
examination, based on the Performance Manual.
The Full-Flight Simulator phase of the course is now embarked on. By
now, the students will be well acquainted with the simulator, their own crew
stations, and the manner in which they use these positions. They will be
familiar with the check lists and will be able to prepare for flight, including
starting engines and setting up the Systems; they will be familiar with
normal, and non-normal operation of the Systems; and they will know how
to shut down the Systems and engines at the end of the flight. A further
important feature of this new training is that, they and their instructors will
know each other and will also be fully aware of what training has gone

365
Aviation Instruction and Training

before. For these reasons, they are able to make full use of the first session.
Furthermore, the syllabus will be completed in ten FFS sessions rather than
the thirteen which were previously required. Because of these factors, a
concurrent review of the FFS profiles has adapted the syllabus to this
reduced number of details.
Early in the seventh week, the Emergency Equipment and Procedures
training is held. This, also, makes use of 'Spiritus' AVT modules. After the
two modules are viewed, a hands-on session with the equipment is held.
Home study follows and, on the following morning, an examination is held.
Right Planning is covered, with a Company Navigation instructor, in the
remainder of this day and on the following morning. The final part of the
ground training takes up the afternoon. This is the operation of the Inertial
Navigation System and, once again, this will normally be done with one of
the Simulator Instructors who have attended the students throughout.

What remains to be done?

A comprehensive, computer-based, technical question bank is presently

being checked by expert teams with the objective of establishing the
reliability and validity of the questions as criterion competency indicators.
This means that, a predetermined pass mark, of a test set from a selection of
questions, will establish the competency of crew members in the area of
technical knowledge of the aircraft. It is intended that all students, line crew
and standards personnel, will have a printed copy of all of these questions,
their answers and references within the Flight Deck Manuals. The book will
be the source of all critical written and oral tests of technical knowledge.
The next step will be to ensure that the training, and the manuals, fully
address the required knowledge areas. This will lead to a revision of the
study guide and the daily quizzes. It will also bring about a revision of the
Systems examination papers. When all of this is finished, a detailed lesson
plan for the tutorials will be produced and this should complete the steps
now being taken to ensure that the students' needs are being properly
addressed.

Conclusion

The revision of Air New Zealand's Boeing 747-200 Conversion Course has
been prompted by the need to update the audio-visual system. The direction

366
 Conversion Courses and Students' Learning Needs

and energy for this has come from the Flight Operations Manager -
Standards and from the B747-200 Fleet Manager. While questions of
course duration and cost have been prominent factors in the review, the
opportunity has been taken to look at the review from the students'
perspective. By adding components and altering others, a relatively rigid
program has catered for different learning styles. This has been achieved by
placing emphasis on ensuring extensive student interaction in the various
phases of the course; by presenting the material in a variety of media -
audio-visual, tutorial, hands-on, manuals and study guides; by assessing
progress at frequent intervals; by ensuring that testing is appropriate; by
co-ordinating the sequencing and flow of components in the training; and by
improving instructor familiarity with students and with overall course
content. Whether student needs are being appropriately addressed, now and
in the future, will be revealed by the critiques. Management is dedicated to
continuing the process of meeting changing needs.
The B747-200 is no longer a new aircraft. Later aircraft have added
technological features which allow for dispensation with the flight engineer.
Then, too, computer based training is replacing the audio-visual medium.
But, despite these changes, the same principles apply. Adult learning is the
focus of this exercise and it will continue to be. The more we can do to
meet the students' needs, the more successful our training will be. The
important thing is that it is those same students who will tell us what we
must do to effect more efficient, cost-effective and appealing training. Are
we listening?

367
25 Design, development
and implementation of
a CRM program
A. Mellor

Crew Resource Management (CRM) has been the most significant

development in aircrew training during the 1980s. Many operators fail to
exploit the opportunities offered by this powerful training tool by hastily
implementing a CRM program before carefully considering their needs.
Programs established elsewhere may not fulfill the needs of your
organization.
CRM training focusses on personal and social skills. It presupposes
adequate technical training and Standard Operating Procedures (SOPs)
already exist and cannot be used to cover deficiencies in technical skills. It
should be used to enhance a crew's technical performance.
This chapter looks at the practicalities of introducing a CRM program by
first considering some of the theory behind CRM training and then using this
to build a step by step method for the introduction of a CRM program
compatible with the training goals of your organization. Some of the pitfalls
that are usual in implementing a program of this type will be discussed with
possible ways to avoid them.

Theory

Training and education can be defined as '... the systematic acquisition of

skills, rules, concepts, or attitudes that results in improved performance in
another environment' (Goldstein, 1974). CRM training attempts to overcome
the shortcomings of conventional aircrew training (skills and rules) by
concentrating on concepts and attitudes. The need for this is evidenced by
the high incidence of human factor accidents in aviation (70-80 per cent).
 Design, Development and Implementation o f a CRM Program

There are three basic determinants of aircrew performance (Foushee and

Helmreich, 1988). These are ability, personality and attitude. Ability is a
factor of inherent qualities, ab initio and recurrent training methods. CRM
assumes technical ability and focusses on social skills that may interfere
with the optimal use of technical skills.
Personality is usually well established by the late teens and short of a
major life event or trauma, does not change radically during an adult's life.
CRM cannot and should not try to change an individual's personality. Many
aircrew believe that CRM is some kind of 'headshrinking' exercise that will
try to change their personality. They will need reassurance that this will not
happen. Similarly, CRM is unlikely to change the behaviour of 'Attila the
Hun' on the flightdeck, but it can give the other crew members some positive
tools to help them cope with him or to work effectively around him.
Attitude is the determinant of aircrew performance where CRM courses
can be most effective. A positive, lasting attitude change amongst an
organization's aircrew is an achievable and realistic goal. It is unrealistic to
expect mature aviators to change their behaviour or personality on the flight
deck following short term exposure to CRM principles. However, a lasting
attitude change may bring about gradual (albeit minor) behavioural changes
over a period of time.
So what is it about an aviator's attitude that needs to be changed?
Research shows (Ruffell-Smith, 1979) that aircrew who perform badly
usually don't work together as a team, with poor communication being a
significant factor in their poor teamwork. Other research (Sloan and
Cooper, 1986) has shown that aviators have a poor understanding of how
excess stress can affect their ability to perform well. Generally speaking,
the things that business management researchers discovered in the 1950s
and 1960s are broadly transferrable to aviation. Also, most aviators have a
very good understanding of their aircraft's operating envelope but a very
poor understanding of their own operating envelope. In other words, they
just don't understand their own limitations or the limitations of their fellow
crew members.
A review of the accident literature will convince sceptics of the basic need
for specialist training in the areas consistently being shown as the cause of
aircraft accidents. This broadly should be the aim of your CRM course.
The challenge for those developing CRM courses is to convince all
personnel of the need for the principles espoused by CRM to be practised
throughout the company and not to be restricted to the aircrew.
Research shows there is a group of people that CRM training does not
help (Helmreich et. al., 1989). This group actually has a worse attitude after

369
Aviation Instruction and Training

CRM training than before. We do not yet know whether these people retain
this negative attitude or gradually return to a more neutral position at some
stage post training.

Stage 1

This is termed the conceptual stage. Why do management feel the company
needs a CRM course and what do they hope to get out of it? Is it perhaps:

- a requirement from the regulating authority?

- as an adjunct to existing training?
- as an integral part of overall training requirements?
- as a solution or part solution to an existing problem within the flight
operations department?
- or a combination of the above?

Any program imposed on the flight crew by management without any

rationale or without their agreement will most likely fail to achieve its
objectives. The most successful courses in the Human Factors field are
those designed and run with line crew input and participation. Usually this
is most easily done through existing aircrew representatives or unions. Also,
open forums to which line crew are invited are an effective way of both
canvassing ideas and building interest and enthusiasm.
This input should lead to a clear statement of goals and guidelines from
management before proceeding to the next stage (Goldstein, 1974; Hersey
and Blanchard, 1988). In my experience with numerous CRM programs,
this is one of the most difficult obstacles to overcome. All senior
management, those that will be involved with the day to day running and
administration of the course, and those responsible for the organization's
check and training standards, must have a clear committment to the program
and be fully aware of the implications of their decisions and actions
(Robbins, 1988). It is to be expected that amongst some of these managers,
there will be those that are sceptical or even downright hostile. These people
must be either convinced of the need for the program or counselled so that
their negative attitude does not hinder the success of the program. I have
seen more programs founder or fail to achieve their full potential because of
this problem than any other reason.
A word of caution! Because of a lack of expertise within your
organization, you may decide to employ the services of an outside

370
 Design, Development and Implementation o f a CRM Program

consultant. This can be of tremendous benefit and ease the process

enormously, however, the consultant may try to impose on the company a
course that doesn't fit the style of your organization. The consultant's
preconceived ideas may be a disaster for you. If you have a clear
understanding of what you want to achieve, you can use the consultant's
expertise to your mutual advantage and remain confident your aircrew will
be comfortable with the results.
Not only do all organizations have their own culture, many have a multi­
cultural aircrew population. Don't assume that because your aircrew have the
same colour skin and speak the same language that they see things the same
way - this may not be the case and often isn’t (different country, different
culture). Your course must be designed with this in mind and be sensitive to
the differences (Johnston, 1992). Large differences in cultural background
and native language require skill and diplomacy in the design, development
and implementation of the course.
If your organization is an airline or charter organization, you may want to
consider exposing your cabin crew to some parts of the CRM training. This
will require careful planning to be a success. It is unlikely that the whole
course will be suitable for them and their presence may hinder your
objectives and annoy the technical crew. However, there are some areas
where their participation may enhance the course and help improve
communications between the cockpit and cabin. The safest way to achieve
this integration may be to have a one day addendum to the course where
cockpit and cabin crew discuss effective communication, team building and
stress management, for example, focussing on each group's particular needs
whilst on duty. Alternatively, separate courses for the cockpit and cabin
crew with combined refresher training may be an acceptable solution.
Finally at this stage, you will need a title for your course. Preferably, it
will be one that is distinctive and can be identified with your organization. I
have always admired the name Alaskan Airlines gave to their CRM course -
ICE (Integrated Crew Experience).

Stage 2

This stage involves the design and the administrative resourcing of the
course. Depending on the resources on hand, the number of personnel
available to work on the project and any other pertinent considerations, a
target date should be arrived at for implementation. Generally, most courses
take two years to become operational. Outside consultants working full time

371
Aviation Instruction and Training

may be able to do it in less, but then they suffer from not being able to judge
the mood of the organization and the course may suffer as a consequence.
The rule of thumb is to double the time you think you need.
Next, gather the resource material and write the content for the first phase
of the course. To ensure a lasting attitude change, the course should consist
of three phases, the Awareness Phase; Line Oriented Simulation (LOS);
and Recurrent Training.

The awareness phase

The Awareness Phase should be an interactive course designed to introduce

your aircrew to CRM concepts and teach them basic skills that they will be
able to use in the following two phases. There are a number of areas that
have been identified as integral to successful CRM training. These are:

1. Effective communication skills.

2. Problem solving, decision making and judgement training.
3. Team building and team management.
4. The effects of stress and stress management.

Other topics that could be considered for inclusion are:

5. Human information processing

6. Situational awareness.
7. Personality types.
8. Leadership skills.
9. The trans-cockpit authority gradient.

More information is available in any of the proceedings from the

International Symposia Cockpit Resource Management Training -
Proceedings o f the NASA/MAC Workshop on Aviation Psychology held at
Columbus, Ohio.
Having decided on topics for inclusion, you should next decide the format
for the awareness course. Is the course to be live-in or by day only? The
advantages of a live-in course are: more subject material can be covered in
a shorter period; no distractions; and it assists in building a team
atmosphere. The only real disadvantage is that a live-in course is much
more expensive. If the course is to be a daytime course, it must be
consecutive days. The training advantages of a live-in course far outweigh
the disadvantage of extra cost.

372
 Design, Development and Implementation o f a CRM Program

The course should be designed from the outset to be interactive. Aircrew

will be much more responsive if they feel their own experience and
knowledge is valued and listened to. Any lecturing should be kept to a
minimum - topics should be raised and the group used to bring out points
that need to be discussed.

LOS

The Line Oriented Simulation (LOS) is the phase of the course where the
real benefits of CRM training become apparent, both to the aircrew and the
instructor/check airman. The sessions should take the form of a 'normal' line
flight where problems are encountered with no obvious right solutions.
Checklist reading should be kept to a minimum consistent with normal line
operations. CRM precepts must be used to arrive at an acceptable solution.
Aircraft mechanical deficiencies should be kept to a minimum and if it is
decided to use them, they should fall outside the scope of the Minimum
Equipment List and the Emergency/Abnormal checklists. Problems such as
passenger illness, complex weather problems or sudden unavailability of
destination airport and a resulting fuel problem will force the crew to use
skills learnt during the Awareness Phase. By structuring the sessions in this
way, it should be possible to see the degree to which the crew have accepted
CRM training to help solve problems, use effective communications and
display good teamwork and decision making.
The simulator should be equipped with a video camera that clearly shows
all crew positions, and extra microphones to pick up all conversation. The
instructor/check airman should be able to put event markers on the videotape
to assist during the debrief. Ideally the crew should debrief themselves with
the instructor/check airman drawing their attention to points of interest using
the event marker. Items that are handled well as well as those that aren't
should be discussed. As this is training and not checking, the instructor
should be non-judgemental. If the instructor feels the crew handled
something badly, it should be obvious to them on viewing the replay and
they should be asked how they could have handled it differently. Most
aircrew are their own harshest critics and their self admonition will be more
effective than any imposed criticism from an instructor/check airman.
At the end of the session, the tape should be erased in front of the crew.
This will remove any anxiety the crew may have of anyone else seeing the
tape. In my experience, these sessions are very well received by crews and
provide individuals with a lot of useful information about their performance
as a crewmember and how they can improve.

373
Aviation Instruction and Training

Refresher training

Refresher training should take place at regular intervals to reinforce key

points from the awareness course and the recurrent LOS. This training
should be at least yearly, possibly as part of the licence renewal process. It
can be combined with other refresher training, such as emergency
procedures (i.e. evacuation), one taking place in the morning and the other in
the afternoon. Recent aircraft accidents and their human factor aspects
make good discussion items at this time.
Having decided on the format the course will take, there are several
practical implications which must be considered. Things such as :

1. How many crews can be spared at any one time? As a guide, courses
larger than 20 become unwieldy and smaller than 10 reduces
interaction and cost effectiveness. The number should be a multiple of
crew type attending i.e. 2 or 3 person crew, to facilitate role playing
exercises.
2. Venue - ideally away from work and family distractions but not too far
as to waste valuable time travelling.
3. Cost - work out a provisional budget (if one hasn’t been imposed from
above). It's important to keep the 'beancounters' informed, if not happy.
4. Time - set up an agenda of things to be done and an approximate
completion date for each item. This will give a better idea of how
quickly the course can be implemented. You will have to liaise with
other sections within your organization to see what their constraints are
and how these constraints may affect your plans. Rostering will have
to know when crews will be required and in what numbers. Courses
should be scheduled so that all aircrew go through the awareness phase
within 2 years to allow the orderly introduction of the LOS.
5. Resources:
a. Human - are you going to use facilitators from aircrew ranks or from
within or outside the company? Using ordinary line crew gives the
course much greater acceptability. Facilitators who understand the
environment crews operate in will have better credibility than
outsiders who may be viewed with some suspicion.
Choosing the facilitators should be done with great care. These
people can make or break the course. They need to be personable,
non-threatening and non-judgemental. A good sense of humour is a
great asset. If you decide to use line crews, they should be popular

374
 Design, Development and Implementation o f a CRM Program

amongst their peers. Aim to get an even mix of captains, first

officers and flight engineers. Avoid using check airmen as this may
give the course the wrong tone and preclude more junior participants
from entering open discussion. Minimum training for line crews
used as facilitators should be a presentation skills course. These are
readily available from companies specializing in staff training.
Check and training airmen need to be given specialist training to
ensure they understand the underlying motives behind CRM and to
help them adjust their methods, if necessary, to be consistent with
what the course is espousing. This is especially important if they are
to run LOS sessions. Other crew will use them to justify their own
behaviour (role models), so they must set a good example. If this is
not given careful attention, everything your organization is trying to
achieve with CRM may founder.
b. Training Aids - a 2 or 3 minute video can illustrate a point more
effectively than 30 minutes of 'chalk and talk'. There are many
commercially available videos but they are usually expensive and do
not necessarily make the point you want to make. Check with other
departments within your organization to see if they have any suitable
training videos. A safety training department is usually a good place
to start.
Another option is to make your own video training films. There
are two ways to go about this. One way is to use a professional
video production company. This is a very expensive option but
usually ensures a high quality result. The alternative is to produce
your own videos. Your organization may have its own video
production unit that you can use, or you may decide to be totally
independent. Very high quality Super VHS/Hi 8 cameras are
relatively inexpensive when equated with the program's overall cost
and the quality of the 2nd generation reproductions are as good as
the original. Two of these cameras will provide you with all the
flexibility you need. To edit the video, a compatible video cassette
recorder (VCR) and appropriate editor will enable you to produce a
good training film. For less than the cost of a professionally
produced video, you have equipment you can use again and again.
Your next decision will be whether to use professional actors or
aircrew in the videos. I have seen it done both ways and there is no
doubt that aircrew add realism that professional actors can never
achieve. There are however a few provisos:

375
Aviation Instruction and Training

- Use brainstorming techniques where your aircrew/actors come up

with scenarios to illustrate a point you wish to make. Do not use a
tight script - they are not professional actors and the pressure of
having to remember lines will remove much of the spontaneity. By
devising a scenario, you will be able to utilize their experience to
fill in the dialogue and add authenticity. As they have devised the
scene themselves, they will be more enthusiastic and natural in
their roles.
- Take time to make sure the technical considerations are sorted out
before filming begins. Amateurs will be unable to reproduce a
scene many times over if required due to technical problems with
the cameras, sound or lighting. Make sure the lighting is correct,
the cameras correctly positioned, loaded and powered and
auxiliary microphones and equipment operational.

Figure 25.1
Suggested camera positions

- Try to devise the scenarios so that the crew are in an environment

in which they feel comfortable, such as a cockpit or bar. Fixed

376
 Design, Development and Implementation of a CRM Program

base simulators or cockpit procedural trainers are excellent for this

purpose.

- Don't waste your time by using standard VHS video cameras.

Because video tape degrades with repeated use, you will need to
have master tapes from which copies can be taken. It is not
possible to make good copies from standard VHS tapes.

A cam era set up outside a B747 fixed based simulator

(Photo: A. Mellor)

Use training aids to stimulate and maintain interest in the

presentation. Slide presentations can be effective and cheap to produce
as can overhead transparencies. Use them to summarize key points in
the CRM manual. They have the benefit of keeping heads up and
ready to interact rather than down and withdrawn. A very effective aid
is to have a slide presentation of a Cockpit Voice Recorder (CVR)
transcript from a suitable accident with an accompanying audio i.e.
actors speaking the words. Most aircrew find these fascinating and
when used to illustrate a point, it makes a lasting impression.

377
Aviation Instruction and Training

Finally, there should be a course manual that covers all the material
presented, giving more detail when necessary. This is not so much to
be used during the course as for reference at a later date when the
crewmember's memory needs refreshing. It is acceptable to refer to
small sections of the manual during the course as long as the manual
does not become a focal point. The manual should be clearly written,
without 'psychobabble', and pitched at an appropiate reading age.
There are a number of suitable computer programs that do this in
conjunction with a word processing program. Once the manual has
been issued, resist the temptation to raise amendments - the bane of
every airman's working life.
An important adjunct to the manual is a pre-course workbook. This
should contain some background information about CRM, summaries
of aircraft accidents for discussion during the course and other material
that will get the participants thinking about the course. Do not make it
too tedious or long - 2 hours reading is sufficient. Any longer and you
run the risk of it not being read prior to the course. To assist the
facilitators, produce a tutorial book containing presentation tips and
points to emphasize, with a suggested time frame for each section.

During this stage there should be a number of information circulars to

aircrew informing them of the planned implementation of a CRM course and
how it will affect them. This will assist the acceptance of the course once it
is finally under way. Calling for volunteers to assist in the work can be a
very rewarding experience for all involved. Amongst aircrew there is a high
percentage of very talented and motivated people, often unknown or
unacknowledged by the organisation's management, who are more than
willing to give you the benefit of their expertise. Spreading the load will
result in a better course with less effort on your part and a feeling within the
organization that the course belongs to the aircrew at large.

Stage 3

This stage involves the writing of the course content, arranging the training
of facilitators, the organization of training aids, the filming of videos (if any),
the setting up of the venue and finally, the printing of the manuals. Senior
management not involved with the day to day activities should be kept
informed of progress and any requirements. This gives them the feeling of
involvement and fosters a protective interest that may be useful at a later

378
 Design, Development and Implementation o f a CRM Program

stage when people who feel threatened by the course start to cause
problems. This may seem paranoid, but has happened wherever CRM has
been introduced in my experience.
To write the course does require some basic skills. Tertiary qualifications
and experience in related areas is a bonus. This does not preclude bright
people with no qualifications who are prepared to work hard to research and
write the material. There is an enormous amount of material available in the
public domain. Unfortunately, the majority of this is American and needs
some adjustment for non-USA aircrew to bridge the cultural gap. This
material is nevertheless of very high quality.
Try to keep the manual aviation oriented, using real life illustrations, and
if possible, at times humorous. One airline had an aircrew member who was
a talented cartoonist and used his talents to illustrate various topics in the
manual. Design the course and the manual so that each chapter is a logical
progression from the previous one, each section building on and reinforcing
the previous one.
The venue deserves some special attention. Attractive surroundings make
the participants more receptive to the course and provides a motivating
environment. CRM courses are enormously expensive when you consider
lost productivity and running costs, so why scimp on the venue when a
successful course may save your organization millions as well as promoting
a morale boosting atmosphere within the Flight Operations Department? An
attractive conference room with smaller breakout rooms for team exercises
is desirable. If the course is live-in, these should be in the hotel you are
using. Exercise/recreation facilities are also desirable with an exclusive area
set aside where participants can socialize after a long day. Never let an
opportunity pass to foster the team atmosphere. As much can be learnt
outside of the formal setting as within. Lavish attention on the participants,
give everyone an opportunity to have their say. Provide some momento of
the course (e.g. a pen, folder or similar). The positive effect of this is well
documented (see Hersey and Blanchard, 1988, The Hawthorne Studies).
Towards the end of this stage, the facilitators should be trained, the videos
in their final form, other training aids finished, the venue arranged and the
manual in the final draft form. It is now time to hold a 'Debugging Course' in
front of the facilitators to test their reaction to the content and presentation,
and to test the facilities at the venue. It should be explained to them that
they will have to present the course, so if there is anything that they are
unhappy with they should say so before the manuals get printed and the
course is in final presentation form. You should now have a fairly accurate
timescale for implementation of the course and final arrangements with

379
Aviation Instruction and Training

Figure 25.2
Suggested seating layout at a venue

rostering section and other interested parties settled.

Now, there are a few other administrative matters to consider.

1. Payment for the facilitators. The work is exhausting emotionally and

physically and around 40 hours preparation will go into each course
initially, reducing only after they have run several courses and are
comfortable with the material. This assumes of course that line crew
are being used as facilitators. Make the payment for this extra duty
commensurate with the effort they put in.
2. Because of the effort required, there is a high burn-out rate amongst the
facilitators. Don't be dismayed by this but accept it as normal and
make plans to cover this contingency as it arises.
3. Participants should also receive allowances on a par with what they
would expect for a normal tour of duty. There are a number of ways to
do this:
- Accommodation provided but the crewmember is responsible for all
other expenses. This assumes a live-in course and may not be
possible at the chosen venue.
- Provide everything and pay no allowance.
- A combination of the above.

380
 Design, Development and Implementation o f a CRM Program

Allowances should be worked out in consultation with the union or

aircrew representatives. Don’t start a course that is supposed to foster
good communications and teamwork with a dispute. Remember, if the
crews feel that the course is special and they also are made to feel
important, they will be more receptive to what is happening and behave
accordingly (the Hawthorne Effect again). This enhances the learning
experience and will result in a stronger attitude shift. If the line crews
feel that management is serious about the principles espoused by CRM,
they'll be serious about it too!
4. How are you going to monitor the effectiveness of the course? This
can be done in-house using a questionnaire or through an outside
agency. Most countries have companies specializing in this service.
The most comprehensive service is provided by an on-going research
project conducted by NASA and the University of Texas. With a large
database of aircrew worldwide, they use a questionnaire to analyze
your course and provide a comparison between your course and that of
other operators with complete confidentiality. This will give you a
good indication if your course is doing what it is designed to do.
Consider getting your Flight Safety Department to set up a data base of
human factor accidents/incidents within your company prior to the
introduction of CRM training to compare with the accident/incident
rate after CRM training. This may also indicate whether there is some
aspect of your operation that warrants special attention during CRM
training.
5. Participants must be given the opportunity to provide direct but
anonymous feedback. A washup session at the end of the course,
preceded by a short questionnaire, will provide this opportunity. These
completed questionnaires should be fed directly to management to
provide useful information on how the crews feel aspects of the
operation could be improved. Crews should be encouraged to take this
opportunity to provide constructive criticism rather than 'bitching'. In
my experience, aircrew relish the opportunity to make an anonymous
input.

Stage 4

This stage sees the implementation of the course. It is most important that
the first few courses go smoothly with a minimum of administrative
blunders. Anything that may reflect badly on the course will travel around

381
Aviation Instruction and Training

the company's grapevine at 500 knots. Trivia such as bad food in a

restaurant can cause damage far in excess of their true value. Aircrew will
give you some allowances, but only up to a point. In general, they will
expect the course to run smoothly and efficiently. It's worth putting in the
effort to make sure it does - the rewards are tenfold.
Keep the information process going. An article or two in a crew magazine
describing initial reactions (favourable ones) will help maintain a positive
outlook and minimize the grumblings of the sceptics. Take photos at the first
few courses and put them on a crew noticeboard to foster interest.
Seek feedback from facilitators and make sure they realize their efforts
are appreciated. Parties or nights out at a restaurant on company expense
will help maintain their morale and may reduce the burnout problem
mentioned earlier. If they are having fun being a facilitator they will be
inclined to hang in longer.
A problem I have often noticed when CRM courses are introduced to a
company is an attitude amongst some aircrew that CRM training is an
optional extra and not something to be taken seriously. This can be
overcome by having a management captain introduce each course to give it
the stamp of authority it deserves.
Line crew will expect management aircrew to attend the course as
participants, not onlookers. This point cannot be too strongly emphasized.
Avoid at all costs the possibility of a 'them versus us' attitude. Most aircrew
feel that management aircrew need the course more than they do and will
resent it if management are seen to be avoiding attendance. Get
management support so line crews understand the course is for all aircrew
regardless of status. Some managerial aircrew feel their authority may be
threatened by courses of this nature. Do what you can to reassure them and
rely on their insecurity being assuaged by reports from their peers who have
done the course. These problems usually disappear once they have done the
course and realize their fears are groundless.
Avoid having too many check and training or management personnel on
any one course. Line crew participants may feel threatened and this will
temper their participation.
Work hard at maintaining a good standard amongst the facilitators.
Regular meetings and workshops where they are asked to present a section
of the course in front of the other facilitators will assist you in assessing their
strengths and weaknesses and provide remedial training if required. Give
them the opportunity to modify sections of the course if they all feel it is
warranted.
Last but not least, is the modus operandi of the organization consistent

382
 Design, Development and Implementation o f a CRM Program

with what you are espousing with CRM? Discuss this with other aircrew
and the facilitators. If it is thought that there are logical inconsistencies
within the Standard Operating Procedures or in other areas (e.g. the manner
in which check flights are conducted), seek to discuss this with management.
This needs to be done with caution and tact, but nevertheless great
determination. There is no point spending literally millions of dollars on a
course to promote correct attitudes on the flight deck if those attitudes
cannot be found enshrined in the SOP and elsewhere within the flight
operations hierarchy and management style:

... the concepts and behaviors covered by the training need to be

continuously reinforced to remain operative. This means that CRM must
become an integral part of normal flight operations and not just a one shot
event or a topic for Recurrent Training if it is to live up to its potential.
(Helmreich et. al., 1991).

Conclusion

This chapter is by no means a comprehensive guide to designing and

implementing a CRM program, but will provide those that are researching in
the field and those that are considering implementing a CRM program, an
insight to some of the theoretical and practical considerations and problems.
The evidence is overwhelming that CRM training saves lives and helps
provide a much more satisfying work environment for aircrew. They are
however, extremely expensive when viewed in isolation. The cost pales into
insignificance if CRM contributes to the saving of one of your hulls. At least
one major airline attributes their CRM program with the saving of six hulls
so far.
Keep in mind that CRM is an educational and training tool, a means to an
end, not and end in itself. Make it do what you want. Have clear criteria for
the success of the program and if these are not met, find the problem areas
and work with your colleagues to remedy the situation so your CRM
achieves its potential.

References

Foushee H.C. and Helmreich R.L. (1988), 'Group interaction and flight crew
performance', in Wiener E.L. and Nagel D.C. (eds.), Human Factors in

383
Aviation Instruction and Training

Aviation, Academic Press, San Diego, pp 189-227.

Goldstein I.I. (1974), Training: Program Development and Evaluation,

Wadsworth Publishing Co., Belmont.

Helmreich R.L., Chidester T.R., Foushee H.C., Gregorich S. and Wilhelm

JA (1989), How Effective is Cockpit Resource Management Training?
Issues in Evaluating the Impact o f Programs to Enhance Crew
Coordination, NASA/UT Technical Report 89-2.

Helmreich R.L., Wilhelm J.A., Kello J.E., Taggart W.R. and Butler R.E.
(1991). Reinforcing and Evaluating Crew Resource Management:
Evaluator/LOS Instructor Reference Manual, NASA/UT Technical
Manual 90-2, Revision 1.

Hersey P. and Blanchard K.H. (1988), Management o f Organizational

Behaviour, (5th ed.), Prentice-Hall, New Jersey.

Johnston N. (1992), 'RM - Cross-cultural pespectives', in Wiener E.L.,

Kanki B. and Helmreich R. (eds.), Cockpit Resource Management,
Academic Press, San Diego.

NASA/MAC Workshop (1986), Cockpit Resource Management Training,

NASA Conference Publication 2455.

Robbins S.P. (1988), Essentials o f Organizational Behavior, (2nd ed.),

Prentice-Hall, New Jersey.

Ruffell Smith H.P. (1979), A Simulator Study o f the Interaction o f Pilot

Workload with Errors, Vigilance, and Decisions, NASA Technical
Memorandum 78482, NASA-Ames Research Center, Moffett Field.

Sloan S.J. and Cooper C.L. (1986), Pilots Under Stress, Routledge and
Kegan Paul, London.

384
26 Crew management in
supplementary airlines
T. Wilson

The sixties and seventies were decades of considerable progress in aviation

safety. Most of the advances were brought about by technological
improvements such as the jet engine, the transistor, ground and airborne
radar, better approach and landing aids and flight simulators.
While the eighties saw further advances in the technological field, the
Ground Proximity Warning System probably being the most notable, it
became increasingly clear that further advances in safety would have to come
not from better hardware but from improving the efficiency of flightcrew.
Training aircrew in management has gradually been accepted around the
world's major airlines as an essential part of training. Many have embarked
on multi-million dollar programs. Regional airlines have been slower to
adopt the idea, even though their aircraft spend much more time in the area
where most accidents happen - namely, near the airport. Perhaps tight
budgets have been the constraint. Yet the training need not be expensive.
This chapter describes how one regional airline has gone about it.

Kendell crew management

Kendell Airlines had its beginnings as a small charter firm operating out of
Wagga, New South Wales, Australia, in 1967. Having acquired a reputation
for service and reliability, the company began Regular Public Transport
services in 1971. The intervening years have seen steady growth. A fleet of
eight Metroliners and six SAAB 340s now covers an extensive route network
from Melbourne and Adelaide hubs, last year (1991) having carried 327,000
passengers and flown 117 million revenue passenger-kilometers, to become
Aviation Instruction and Training

Australia's largest Supplementary Airline.

With the first twenty-five years free of accidents, founder Don Kendell
recognized that statistics were beginning to mount to the point where
renewed effort would be needed to maintain safety standards. A
well-maintained fleet of turbine-powered airliners and thorough training of
personnel had ensured high safety standards, but world-wide experience
showed that, in four out of five accidents, there was some failure on the part
of the crew. Further, the crew involved in these accidents were almost
invariably very experienced. Bureau of Air Safety Investigation figures
clearly show that, while accidents due to handling errors tend to decrease
markedly with experience, those due to errors in judgment and
decision-making decrease at a much lower rate (Figure 26.1). A Crew
Management program therefore, offered the best potential for ensuring that
Kendell's record should remain unblemished. This form of training seemed
particularly appropriate for those pilots and flight attendants who had joined
the company without experience in multi-crew operations, since it would give

Figure 26.1
Accidents versus experience

386
 Crew Management in Supplementary Airlines

them the opportunity to look at and practise the different skills required. The
fact that this could be done away from the immediate technical and
commercial concerns of the cockpit or cabin meant that full attention could be
devoted to the techniques presented, and therefore maximum benefit gained.
Experience reported from other operators suggested strongly that, apart
from the enhancement of safety, Crew Management training could result in
better relationships between crew members, and between crew and
management - in short, higher morale. Though morale was not considered a
problem at Kendell, reports from people who have attended the course so far
have been very positive.

Research came first

Meetings were originally held to ask crews whether they were interested in
the idea of such training, and these were followed up by contact with each
individual pilot, to identify possible problems in the company. These proved
very valuable. Not only were the pilots interested, but flight attendants also
contributed, even though they had not been included in the initial survey. As
a result of their keenness, and in the light of such incidents as the British
Midlands crash at Kegworth, it was recognised that Flight Attendants were an
integral part of the team, and should be included in any training program.
A steering committee was formed in each base, and after analysis of the
feedback from crews, a program was drawn up to cover as many issues as
possible. It was decided that, in order to make an effective impact, the
course would need to be at least two full days, and would be best conducted
at a venue remote from the airport and from home, so that participants could
devote their full attention to the subject. The venue chosen had excellent
accommodation and conference facilities, which helped ensure that this
happened.

Getting under way

A prototype course was held in October, 1991. Representatives of the

steering committees attended, and it was decided that, with a few minor
adjustments, the program should go ahead as it was.
Kendell has about 75 pilots and 25 flight attendants, and it is not possible
to take more than one crew at a time from each base, so that the course must
be run with an average of only six people - though this is sometimes

387
Aviation Instruction and Training

augmented by visitors from other organizations. Most airline courses have

bigger numbers, and there was some initial doubt about the value of such a
small group, but it has been found that in fact there are advantages, such as
the greater opportunity for individuals to give their views. Since crews in
Kendell Airlines consist of either just two pilots or two pilots and a flight
attendant, exercises with small numbers are quite appropriate.
It is now nearly fifteen years since the first serious training began in
Human Factors, and particularly that part of Human Factors which has come
to be known as CRM (Cockpit Resource Management, or more recently
Crew Resource Management). It is probably fair to say that there has been
some refinement of the subject matter in that time, as each new course has
been able to build on the experience of those that have gone before. While
KLM’s KHUFAC course was a trail-blazer, it was designed purely as an
awareness course. United Airlines, in their Command Leadership Resource
(later CRM) course, attempted to go much further, in that they actively
involved their crews in the program by getting them to take part in
problem-solving exercises which resembled real line situations. This gave the
course the stamp of practicality - a characteristic dear to the hearts of all
pilots - although it was recognized that no 'one-off course could be expected
to make lasting changes to attitudes without a regular program of follow-up
training. Most subsequent courses around the world have followed this
pattern, and so it was decided to adopt the workshop method.

The course content

The Kendell steering committees settled on the following subject areas as

essential to the course:

- teamwork and leadership;

- problem solving and decision making;
- communication skills;
- stress management;
- information processing;
- self awareness; and
- situational awareness.

The program is structured around the basic premises that:

1. The safety record shows the high incidence of crew-related accidents.

388
 Crew Management in Supplementary Airlines

2. Attention to management or team-building skills can improve crew

performance.
3. Team performance is clearly dependent on an awareness of the strengths
and limitations of the individual team member, as well as the combined
effort of the group.

Introductory written material is sent to each participant about two weeks

before the course, asking him or her to consider the world safety record and
the implications for each individual. This is reinforced during the opening
session of the course by a video and discussion period, ending by spotlighting
the essential characteristics of high-performing teams.
Problem solving and decision making in the team environment are dealt
with initially by looking at some past examples of breakdowns resulting in
accidents or incidents, and then by the introduction of a model which can be
used to guide the problem solving process (the PILOT model, Figure 26.2).

Figure 26.2
The PILOT problem solving model

389
Aviation Instruction and Training

The use of this model is then practised in some simulated flying scenarios.
Psychologists are still undecided about the exact process followed by
experienced crews when confronted with a problem, but there is no doubt
that many have come unstuck because they failed to give sufficient thought to
one or more of the steps in the model. Crews in this course have made it
clear that they find the model a practical aid when dealing with a problem.
The salient features of the model are that:

- ALL relevant information is acquired;

- a problem is properly identified before solutions are attempted; and that
- the action taken is reviewed, both after a particular event, and at the end
of each flight.

This is a key strategy for any crew which wishes to achieve a high level of
performance.
The importance of communication in flying safety is looked at first from
the point of view of PA and radio transmissions. This focus is then broadened
to that of cockpit interaction and crew relationships in general, as well as the
wider workplace. Some general principles are drawn, the application of
which can help the crew member in his or her daily duties. In particular,
emphasis is placed on the importance of listening and actively seeking
contributions from others in the crew, particularly for the captain. Because so
many accidents have occurred in which the supporting crew have failed to
make their concerns forcefully known to the captain, crew members are
encouraged to develop the skill of assertiveness where the situation demands
it.
About halfway through the course, emphasis changes from the crew as a
unit to the individual crew member, and how he or she can contribute to the
joint effort. Particular topics covered are stress, information processing and
self awareness.
A discussion period on stress, after looking briefly at the basic theory, gets
down to practical details such as noting the signs of stress and particularly
what to do to run one’s life in such a way that stress is not only kept to
manageable proportions, but is used to make work enjoyable and rewarding,
as well as more efficient.
A basic understanding of information processing is encouraged by a
discussion of perception, memory and computing ability. Some of their
characteristics and limitations are illustrated by simple demonstrations, and
then related to situations likely to occur on the job. Visual illusions in flight,

390
 Crew Management in Supplementary Airlines

the characteristics of short and long term memory, the dangers of mindset
and precautions to deal with the bottle-neck of the single channel processor
are typical of attributes discussed.
Since it is widely accepted that personality is a major influence in any
group situation, it was considered important that participants in the course be
given the opportunity to look at their personal styles of relating to others and
going about their work, and also to recognize that others in the crew might
have different preferences and styles. The instrument used as a tool here is
the Myers-Briggs Type Indicator, which is probably the most widely-used
psychological assessment system in the world. It has the advantage that it is
a self evaluation, rather than being imposed by an outside 'expert'. Because it
deals with just four parameters, it is readily understood by non-psychologists,
and is therefore well-accepted by crew members.
The three modules of stress management, information processing and self
awareness combine to encourage participants to consider their own positions
in the crew, and to optimize their talents to make the crew as efficient as
possible.
Perhaps the key element in the program is that of situational awareness,
because it ties together all the earlier modules. A high-performing crew will
be continuously aware of all the various factors relevant to the flight, and will
use the management skills discussed throughout the course to ensure that the
flight is successfully concluded. The concept of the error chain is used to
help crews recognize ways in which safety might be jeopardized. To enable
participants to identify links in an error chain they are given a summary of an
actual accident and asked to report where the operating crew could have
broken the chain.

The flight attendant and the course

The Kendell Crew Management course is unusual among CRM courses in

that flight attendants take part fully in the program - the first time this has
been done in Australia, and among very few in the world. There are several
reasons.
First, in contrast to the position in the major airlines, the flight attendant on
the SAAB 340 works alone with up to 34 passengers. He or she must
therefore be capable of showing leadership skills of a high order, particularly
if an emergency should arise. Many of these skills are identical to those
needed by crew members in the cockpit.
Secondly, the crew of two pilots and one flight attendant work closely

391
Aviation Instruction and Training

together in preparing for a flight and during turnarounds, and it is common

practice for tasks to be carried out by any one of the three, depending on the
circumstances. Good relationships are vital to efficient work at these times,
and an appreciation of the other person's responsibilities and problems helps
to meet this end.
The third, and perhaps the most compelling, reason for including flight
attendants on the course is related to situational awareness, which can
perhaps be defined as an alert appreciation of everything that is going on at
any time. For a flight attendant, this will include not only a continuous
assessment of the needs, comfort and demeanour of passengers, but also an
appreciation of what the aircraft is doing and where it is, as far as that is
possible from the passenger cabin. There have been many occasions when an
alert flight attendant has made a useful - and sometimes vital - contribution to
the safety of the flight through bringing some abnormality to the attention of
the operating crew. The most unfortunate experience of the Kegworth B737
is a classic illustration of an accident which might well have been avoided if
there had been more input from the cabin crew, or more searching
consultation from the captain. Even on a normal flight, an appreciation of the
aircraft’s position and its progress on the leg can be useful in answering
queries from passengers, but when something out of the ordinary occurs, the
flight attendant's contribution can be critical.

Kendell crew management: does it work?

At the time of writing, just half of Kendell's crews have attended the crew
management course, so it is too early to be able to make any objective
judgments about its value. There is, however, no doubt at all that it has been
received very enthusiastically by both cabin and cockpit crews. It is
therefore already apparent that some of the spin-offs expected from the
program, such as an improvement in the already high level of morale, and
better understanding between cockpit and cabin crew, are being achieved.
Judging by results from earlier programs, it can confidently be predicted that,
when the principles have been incorporated into the training program, it will
prove to be a powerful tool in the quest for enhanced safety levels. How to
ensure that the principles put forward during the initial course are applied in
further training and on the line are questions that are being actively pursued.

392
27 Establishing a cockpit
resource management
training program

R.L. Wilkinson

This chapter is intended as a broad practical guide to the introduction of

Cockpit Resource Management (CRM) training to a flying organization. I
believe that the principles are the same, regardless of whether it is a small
charter company or a large airline. Each organization has its own problems,
but finding the answers is a common problem.
Predominantly, the chapter is written from the experience gained in
establishing a CRM program in a large international airline with modem
training equipment. However, the same principles apply to organizations
without expensive training aids.
The chapter deals with establishing the need for CRM training and then
explores whether the training program should be bought or built. For those
who have made the latter decision, it looks at building the program, training
those who will conduct CRM training, the integration of the concepts into
the entire training environment, introducing LOFT and evaluating the
program. A table of suggested reading and references concludes the
chapter.
Probably the first questions asked when the introduction of CRM training
is contemplated will concern cost. The cost depends on many factors, as we
will explore. However, among the decisions necessary at the costing stage
is the length of the course. Industry standards in this area vary from one day
to one week, or more. For reasons that will be apparent later, even if the
Aviation Instruction and Training

budget will allow, a long course for practising aircrew is probably not a
good idea. It is better to deal with portions of the subject thoroughly than
attempt to cover the field in one course. 'Little and often' is preferable to 'a
lot once'. A two day course may be the ideal, allowing the fundamentals of
the subject to be well presented. If possible, expand on that base with a new
course every couple of years.
Line Oriented Flight Training, or LOFT (also known as Full Mission
Simulation), is the conduct of real time en route exercises in flight
simulators. It is an essential and integral part of CRM training and, to a
great extent, failure to include LOFT in a CRM program is akin to learning
to fly aeroplanes from a book. You know how - in theory.
The more realism that is achieved in LOFT, the more acceptance it
receives from crews and their perception is that the value is higher.
Certainly, modem flight simulators with 210 degree daylight visual systems,
good airport models, voice synthesization, automatic ATIS transmissions
and recorded continuous ATC background are superb training aids in the
teaching of CRM. However, reports suggest that a great deal of value can
be had on a much lower budget. One reasonably large military group in the
USA conducted LOFT, from which their crews perceived they had gained
great value, using paste - on instrument panels and broomsticks.
It is probable that LOFT is seen by different organizations in different
ways. Pilots working for a big airline look upon it as an opportunity to
practise possible situations that they may one day encounter in the air. The
fact that they are practising the art of cockpit resource management may not
be obvious to them. To their less affluent confreres, LOFT is the
opportunity to play out difficult scenarios in readiness for putting CRM
skills into practice in the air. To my knowledge, aviation psychologists have
yet to research whether dollars give proportionate value. I suspect that they
do not. It is more probable that you reach a certain level at little expense
and then progress from there, but at vast expense.

Establishing the need

You are probably establishing a CRM program for one of three reasons:
you have a CRM problem, you are being forced into it by regulatory
pressure, or the more likely and practical reason - what if you had a human
factors related incident and you have no CRM course?
In the long run, it is of no consequence which of these reasons motivates

394
 Establishing a Cockpit Resource Management Training Program

you; the approach is the same in each case. For whatever reason you need
the program, it is necessary to first discover the major deficiencies of your
operation. Until you have done this, you do not know what problems you
need to solve and you can never know, later on, whether you have solved
them.
The basis of any evaluation of the problems must, quite clearly, be
expertise and it may be that the task should be contracted out to persons or
companies with that expertise. That is not to say that an in-house evaluation
is without merit. In fact it forms a valuable part of the process. However,
the experience of the large airlines who were forced by circumstance to
embark upon CRM training, appears to have been that the necessity came
about due to their own lack of perception of the problems. How Effective Is
CRM Training? by Robert L. Helmreich et al. is essential reading at this
stage.
The advantage of an independent evaluation is that flight crews are far
more likely to discuss the problems they see with such a source rather than
the company's own management or Check and Training organization. This
discussion is a vital part of the evaluation.
The role of the management and Check and Training organization in the
evaluation process is in their intimate knowledge of cyclic and promotional
training problems within the airline and the deficiencies uncovered by their
licence renewal checks.
The evaluation must gain information in two crucial areas: the
deficiencies of the operation and the attitudes of the crews; and, of prime
importance, the Check and Training organization and its instructors. The
point will be made frequently in this chapter that the ultimate success of the
entire program hinges upon the commitment and expertise of the
instructional staff charged with teaching CRM.
Data must be gathered, therefore, but it must be objective, standardized
data. It is important at this stage to predict the performance indicators that
will later be used to gauge the success of the program and gather the
baseline data for comparison.

Buy or build?

Cockpit Resource Management has been taught for many years now and
essentially the concepts are well documented and easily researched. If all
aircrew could simply be given a book to read, then one of the more

395
Aviation Instruction and Training

significant costs of constructing a program could be eliminated.

Unfortunately, experience has shown that issuing books does not solve
CRM problems. The way the subject matter is presented is of prime
importance. The fact remains, though, that the subject matter of the course
is available in text books and is unlikely to be subject to copyright.
There is no reason, therefore, why any organization cannot construct its
own program, provided it has the expertise, or can import that expertise.
Buying a CRM program can be expensive. If you hire a team of experts to
design a custom built program, the cost will probably be proportional to
your ability to pay. A large airline could be looking at seven figures, a
smaller organization six. What you will get for your money is a course
tailored to your company, hopefully responding to its specific needs and put
together with the professional skill. However, in the long run you will get a
course that does not differ in essence from the other courses that your
consultant has put together.
The success of your program will depend not upon the CRM knowledge
of your consultant, but rather the consultant's understanding of the culture of
your organization and its staff. This culture not only depends upon the
nationality, but also the traditions of your organization. It is one thing to
build a CRM program for a newly formed airline and quite another to build a
program for a long-established one.
An alternative to handing the entire job over to a consultant is to purchase
a ready made course or contract out the training to an airline with an
established program. This may be a cheaper proposition than the previous
option. How much you pay for an established course is a matter of market
forces. There are plenty of companies willing to sell CRM courses.
Sending crews to attend established courses will probably be the more
expensive option of these two because of the profit expected from the
manpower, facilities and materials used.
The advantages of buying a course, either off-the-shelf or custom made,
are that it can be achieved quickly and does not require the commitment of
your company's resources to the task. The disadvantage is that there may be
few of your company personnel with a commitment to CRM in general and
the course in particular. An off-the-shelf course has the added disadvantage
of possibly being foreign to the culture of your organization. If you have the
time and the resources, the preferable option is to build your own course. By
so doing you get a CRM course which is designed to meet your needs and,
most importantly, a course with which you and your people can identify.

396
 Establishing a Cockpit Resource Management Training Program

Building the program

As part of a CRM survey, I observed a short one hour sector. On joining

the crew at pre-flight planning, the captain said that he had welcomed the
introduction of CRM training and expressed the opinion that for some of his
colleagues such training was long overdue. He wondered, however, that I
did not observe longer sectors for preference, 'where there's a chance for
some Cockpit Resource Management to take place'. The captain then
proceeded to deal with a complicated weather/fuel/curfew decision by
making calculations privately on a scrap of paper, announcing his decision to
the dispatcher, to the consternation of his first officer. Most pilots will tell
you that CRM training is a good thing - for other pilots. Mark Twain
summed it all up when he said that to be good is noble, but to teach others to
be good is nobler - and less trouble.
Aircrew rarely enter training programs by choice and generally only do so
in order to maintain the job they have or to gain promotion to another.
Faced, therefore, with the introduction of a training program that the
consumers do not really want, or believe they need, the potential for failure
is clearly great.
The starting point in building the program is the appointment of a project
manager. The person selected must be committed to CRM and be willing to
learn. Lack of knowledge of the subject, at this stage, is not important. The
person must have credibility with both management and the aircrew. The
first task of the Project Manager is to become familiar with the subject.
Aviation psychology journals and the proceedings of the various conferences
discussing CRM are fruitful starting places to acquire knowledge of the
subject. There is a wealth of reading available and attendance at one of the
better conferences will establish contacts which may be invaluable. The
very essence of CRM is synergy and now is the time to make contact with
those who can give advice, the people who have trodden the path and know
where the potholes are.
Attendance at CRM courses is useful. If possible, a course should be
selected from your organization's own country and another from a different
country. Thus, a comparison of cultural differences can be made of the
various approaches to the problem of teaching CRM. By comparing another
operator in his or her own country with one outside the country, the manager
will be able to more easily see the unique cultural influences which must be
dealt with in his or her own organization.
Having gained a feel for the subject, the Project Manager should appoint a

397
Aviation Instruction and Training

Steering Committee to oversee the construction of the course. It is essential

to involve the aircrew in the design and construction of the course from the
start. It is most important that they believe it to be their own course, not one
imposed upon them. The involvement of the aircrew in 'their' course is one
of the keys to success.
The other key to success is the involvement of the instructional staff who
will ultimately teach the subject. I am not referring here to the facilitators of
the course itself, but the simulator instructors and check pilots who will
eventually be responsible for the introduction of the CRM program into the
cyclic, promotional, line training and general day to day life of your
organization. If aircrew unions are involved, it is best that they are involved
from the very beginning by allowing them to appoint members of the
Steering Committee. Inevitably, they will be concerned with negotiating the
conditions of the course for their members and their early participation will
be of benefit.
Whilst restricting the Steering Committee to a workable level, it is
important that all interest groups are given the opportunity to participate,
including Line and Training management, along with Union and
Instructor/Check Pilot bodies. Once again I make the point that the training
personnel must be convinced of the merit of the program and it will be
useful if the Instructor representative can report back to his colleagues to
receive feedback and input. Remember that your instructor group may be
your more senior people and the group who will take the most convincing
that the program is necessary. It will later become their task to teach CRM
at the coal face and the sooner they can be indoctrinated, the better.
The content of the course will greatly depend upon the proposed length
and the results of your initial investigations into the problems of your
organization. However, it is more likely that this will affect the presentation
rather than the subject matter, peripheral issues rather than the core. There
are certain basics of CRM that qualify to gain a place in any program.
These are:

- Decision making
- Communication
- Attitudes
- Leadership
- Followership
- Teamwork

398
 Establishing a Cockpit Resource Management Training Program

There are, of course, other topics, such as stress management, arousal,

prioritization, and so on - an awareness of which is ultimately essential to
the safe operation of aircraft. However, the above subject matter will
inevitably form the framework upon which the course is built.
Having decided the content of the course, the form of presentation should
the next consideration. At this point, resist the temptation to teach too much
theory! A certain amount is necessary, but it should be minimal. The aim
should be to give your people tools that they can take to the workplace.
They are accustomed to being presented with tools. Standard Operating
Procedures are tools that aid their normal operation of the aircraft.
Checklists are the tools that support not only that normal operation, but the
non-normal as well; mnemonics and learning of recall emergency items by
rote are further tools supporting crew members in the technical operation of
their aircraft. They must now be given CRM tools to support their
management of the aircraft.
Modem flight training involves a progression from classroom theory (be it
face to face, carrels or CBT) to part task trainer (if available), to simulator,
to aircraft base training (if required), to aircraft en route training. CRM
training must follow the same progression. If you start on the basis that
nothing will be taught in the classroom that cannot be demonstrably
practised in the simulator and in the air, this will make the first step a step in
the right direction. A few illustrations of the handling of the staple topics
may be of assistance.
Starting at the decision making process, you should invent a mnemonic
that is easily remembered and either has some special significance to your
organization or is capable of being employed as a double entendre in the
promotion of your course. For instance, the decision making process may
be defined as the 'GRADE' process:

- Gather information
- Review priorities
- Analyse alternatives
- Decide
- Evaluate the outcome

- Do your decisions make the grade?

Whatever the process and the supporting mnemonic, it must be practical and
stand the test of practical use. Do not wait until you are committed to the

399
A v i a t i o n I n s t r u c t i o n a n d T r a in in g

process before trying it out on a complicated decision!

Communication covers a lot of ground and is not always easy to teach in
the practical sense, i.e. the classroom/simulator aircraft progression.
However, there are aspects, especially those that directly inter-relate to other
subjects, such as leadership and followership, that do lend themselves to that
progression. An example may be the working tool of the DARTS board for
practical debriefing skills.

Figure 27.1
DARTS diagram

The approach to teaching 'Attitude' is reasonably diverse amongst those

who teach CRM. Frequently the subject is taught by introducing each
course member to some form of psychological appraisal aimed at providing
them with an insight into which preset attitude pattern they fit. Fascinating
though this approach may be, once again we need to give crews the tools
with which to determine the correct attitudinal responses, to recognize poor
attitudinal response in others and deal with attitudinal problems.

400
 E s ta b lis h in g a C o c k p it R e s o u r c e M a n a g e m e n t T r a in in g P r o g r a m

Figure 27.2
Attitude diagram

A typical attitude diagram may be similar to the one illustrated above. It

is deliberately styled in a familiar aviation format and is clearly a
representative rather than a quantitative presentation. Its purpose is simply
as a tool to aid practical discussion by crews (and also supervisors and
instructors) of perceived attitude problems.
'Leadership' is a subject that has been exhaustively researched and is the
subject of many publications. There are a number of leadership models of
which the Blake/Mouton managerial grid is perhaps the most widely used in
CRM circles. Its use is not as a theoretical model, but rather a practical tool
allowing an impersonal descriptive language of leadership style, for
example, 'a bit 9:1', or 'too 1:9 to be effective'. Although there is a point of
view that the basis of the managerial grid is too narrow, it remains a quite
workable tool.
Another (generally more preferable), approach is that of Adair, who
differentiates between the team and the individual.
Teaching followership suffers the initial and obvious difficulty that the
word does not appear in the average dictionary. However, we are

401
A v i a t i o n I n s t r u c t i o n a n d T r a in in g

Figure 27.3
Leadership grid diagram

Figure 27.4
Leadership circles diagram

402
 E s t a b l i s h i n g a C o c k p i t R e s o u r c e M a n a g e m e n t T r a in in g P r o g r a m

essentially discussing 'support' or, as one airline has termed it, 'managing
upwards'.
The attitude graph shows the optimum level of support and gives crews a
visual presentation of the problems of support. However, the major problem
that has to be addressed in most airlines, is lack of assertiveness on the part
of crew members. It is in this area that your course can provide invaluable
tools by provision of an official language of assertiveness.

Figure 27.5
Assertiveness diagram

'Teamwork' is another subject that is supported by a plethora of literature.

Once again, you should look for tools that crews can use.
Having decided the content of your course, teaching methodology is the
next concern. If you are to avoid the theoretical approach to the subject,
you must ensure that there are practical exercises supporting each concept.
There is a danger that the trainers, possibly unwittingly, will alter the
philosophy of your course by injecting their own interpretation. This can be

403
Aviation Instruction and Training

avoided to a great extent by ensuring that the theory is taught in pre­

prepared video or audio/visual presentations.
A suggested format for the course is:

1. Short oral introduction by the facilitator.

2. Video presentation of the theory.
3. Group discussion of the theory.
4. Video presentation of the practical application of the theory.
5. Group discussion of that practical application.
6. Practical exercises to consolidate.

Once the Steering Committee has decided upon the subject matter and
methodology, the actual construction of the various segments may well be
handed over to a team of course developers, people with expertise and
aptitude for this work, preferably, for the 'ownership' that it will produce,
nominated by the unions and the instructor body.

Training the trainers

There are two levels of trainers that we are referring to here, the CRM
course facilitators and the instructors who will conduct LOFT. Note the use
of facilitators to guide courses through the process of discovering CRM
concepts, rather than instructors to teach those concepts. The CRM course
facilitators should be selected from volunteers with a special interest in
CRM and the presentation of training courses in general. The greater the
enthusiasm displayed by these people, the greater the impact of the course.
As we have already mentioned, it will not be easy to convince your
people that the course is necessary and worthwhile and it is therefore
essential that your course is presented by people who are not seen to be part
of the Establishment and yet who have credibility as competent operators.
The starting point in the training of the facilitators is a thorough grounding in
the CRM principles used in the course. They must understand the concepts
to be taught and also the reasoning behind the chosen presentation of those
concepts. Training your facilitators to actually conduct the course is best
done by experts. It is as well to remember that the first impressions of those
exposed to the course will be entrusted entirely to your facilitators and
economies in this phase of the operation may cost dearly later. A good
presentation by well trained facilitators is essential to the program.

404
 Establishing a Cockpit Resource Management Training Program

The training of the training captains and check airmen who will teach and
evaluate CRM en route, and the instructors who will conduct LOFT in the
simulator, has to be approached with a great deal of preparation and thought.
Failure to properly train these people will have a seriously detrimental effect
upon your program.
The training should cover the following as a minimum:

1. CRM concepts.
2. Recognition of CRM concepts.
3. Running LOFT.
4. LOFT construction.

It is most important that your instructor corps have more than a passing
knowledge of the CRM concepts taught in your course. Although they will
have completed the course themselves, they must be given further insight
into the rationale behind the concepts and be able to engage in informed
discussion on the concepts. Knowing the theory is one thing; recognizing it
in action is another. The instructor course should include plenty of
opportunity to observe good and bad application of the principles. This can
be achieved cheaply and easily, especially if your simulators are fitted with
video recorders, by making video vignettes for group discussion.
Running LOFT may be very different to anything your instructors have
been called upon to do before and requires instruction. It is necessary to lay
down some basic rules of LOFT and ensure that instructors are familiar with
these rules and the practicalities of achieving the levels of realism that you
require. A practical method of ensuring that your instructors understand
what LOFT is all about is to give them instruction and practice in the
construction of LOFT. When doing so we can get across the key to LOFT
construction; the scenario writer must be able to identify the principle(s) of
CRM intended to be demonstrated in each major event of the LOFT. Far
too frequently a decision-making exercise, for instance, is structured in such
a way that there is no decision to be made!

Integrating

Once the CRM course has been introduced, it is time to integrate CRM
training into the entire training environment. CRM can be effectively
introduced into new hire training, promotional training, endorsement

405
Aviation Instruction and Training

training, and cyclic training and checking.

New hire training will clearly require the inclusion of your CRM course
or even an expanded version, depending upon the experience levels of your
intake and their previous exposure to Human Factors training. It will be
necessary to reinforce this training with simulator exercises, including
LOFT. It is inevitable that some type of sector training will be included in a
new hire program and CRM training can therefore be introduced without
greatly expanding the existing program.
Promotional training is a fruitful and essential area into which CRM
training must be introduced. If you influence nobody but your future
captains and first officers, then your program will have succeeded. An
expanded ground training program should be considered which focuses more
intently on the new role of the pilot concerned, be it the command/leadership
role or the first officer/support role. The ground training must be extended
to the simulator, where carefully constructed training sessions should be
devised to reinforce the ground training. It cannot be emphasized too
strongly that you are building the future of your organization here. Build
carefully and build strongly!
Endorsement training should be used to emphasize the CRM differences
of the new aircraft. You may be converting a pilot from a three person flight
deck to a two person, or vice versa. You may be converting a pilot from an
analog to a 'glass' cockpit. Each of these changes involves some cockpit
resource management facet that can be emphasized in the classroom and in
the simulator.

Establishing LOFT

Integrating practical CRM training into promotional and endorsement

training presents few logistical problems and relatively little expense.
However, providing LOFT for line crews can be expensive. The ideal is to
run a dedicated LOFT program, once or twice a year, and roster all crew
members to attend these sessions. Even with a small organization this would
involve considerable expense. With a large organization with several bases,
the cost would clearly be high. A practical solution is to incorporate LOFT
into the annual cyclic training and checking schedule.
LOFT is time consuming and if it is to be introduced into an already full
schedule of training and checking, something has to be removed. There are
two approaches: include a LOFT content in all or most sessions, or run one

406
 Establishing a Cockpit Resource Management Training Program

dedicated LOFT session, say, annually. As was mentioned earlier, LOFT

should be designed to give training in specific CRM principles, and it would
be possible to arrange small LOFT segments in any training session.
However, research and experience have shown that crews find LOFT
particularly worthwhile if they can take complicated situations to their
logical conclusions. So much of their training time in simulators involves
dealing immediately with specific situations and then passing on to other
sequences, the crews rarely having the opportunity to deal with a situation in
its entirety. It is likely therefore that your crews will gain the most benefit
from a single dedicated LOFT session annually, rather than several sessions
with a small LOFT content. Another disadvantage of including a LOFT
content in all sessions is the decrease in handling practice that this entails for
your crews.
If you are running a program in which your crews attend cyclic training
and checking at six monthly intervals, you face the problem that your crews
regard the cyclic training prior to the licence renewal as a practice and
would not welcome substituting LOFT for that practice.
A possible answer to the problem may be to change the timing of the
sessions from two at six monthly intervals, to single sessions at three
monthly intervals. This has the advantage of keeping your crews more
familiar with the simulator and over a period of time they will feel less need
for a practice prior to licence renewal. A better solution, however, is to
entirely change the licence renewal program so that instead of the familiar
biannual cyclic training/licence renewal dichotomy, you operate a cyclic
training program which guarantees to expose your crews to an acceptable
level of training that ensures competence in all required sequences over a
period of, say, three years. To achieve such a program obviously requires
the co-operation of the regulatory authority. However this has been
achieved, certainly in Australia, and has been successful.
The program involves agreeing with the regulatory authority the number
of times a particular sequence will be practised over a three year period and
the minimum time between completion of sequences. Thus you may agree
that a Cat ΙΠ II.S will be included in eight of the twelve sessions and that it
will be included at least every six months. This enables you to ran one
session every year that does not include a Cat ΙΠ ILS. You may agree that
an NDB approach will be included in six of the twelve sessions with a
minimum frequency of six months. You may agree that a hydraulic failure
will be included twice in the package with no minimum frequency. Thus a
matrix is built up showing training/checking items, their number and their

407
Aviation Instruction and Training

frequency. It is now possible to build a three year program in which you

have the flexibility to include LOFT, in which you do not necessarily have to
force crews into completing a particular sequence. For instance, if you write
a scenario which envisages an arrival at Los Angeles, it may detract from the
realism if you force an NDB approach to 24R when all four runways in fact
have an ILS at each end. Better to allow the crews the choice of approach
and write your NDB approach into another session.
The airline that persuaded the regulatory authority to drop formalized
licence renewals and substitute an ongoing training package, achieved this
on the basis that crews were not allowed to walk away from cyclic training
without having achieved a satisfactory standard in all sequences and
therefore ALL sessions were de facto licence renewals. Licence renewals,
as such, are stressful and, if the training department of your organization is
functioning properly, unnecessary.
If it is not possible to integrate your CRM training into your cyclic
training and checking package, then you may still be able to gain some
financial relief by persuading the regulatory authority to reduce the route
checking requirements in exchange for the extra training.

Evaluating the program

Evaluation of the program may start with a survey of the attitudes of

personnel to CRM for comparison with data gathered at the commencement
of the program. This comparison will provide pointers to the probable
success of your program as it relates to day to day line operations. If you
have convinced them that the concepts are useful and that they work, then
you have probably modified the behaviour of your crews.
Further evaluation can be achieved by surveying both crews and
instructors during LOFT. The NASA/UT LINE/LOFT evaluation form is
ideal for this purpose and can be obtained from the former organization.
Apart from its value in the independent assessment of your own results, the
facility exists to compare your results with data acquired from other
operators who have submitted their results to NASA. Additional useful data
can be provided by surveying the instructors who conducted the session and
comparing their perceptions with those of their students.
The evaluation will hopefully tell you which areas need further
concentration in the future and assist you with ongoing CRM curriculum
development.

408
 Establishing a Cockpit Resource Management Training Program

Conclusion

What can you expect to achieve if you implement CRM training? You
certainly cannot expect overnight miracles. Although there is plenty of
research that suggests that aircrew can be enthused by a superb program,
there are very few researchers who will state categorically that such
enthusiasm will, in all cases, translate into better teamwork on the flight
deck.
There are certainly intractables in every organization and it is excessively
ambitious to expect to influence them all, although you can, at the very least,
expect to train their unfortunate crews to deal with them. What you can
hope and expect to achieve, however, is an investment in the future. You
have the opportunity to introduce a CRM ethos into the organization which
will ensure that the future of aviation is safer than the past.

References and further reading

Foushee, H.C. (1990), The Current State Of CRM Training, Proceedings of

the 7th General Flight Crew Training Meeting, International Air Transport
Association.

Hawkins, F.H. (1987), Human Factors In Flight, Gower Technical,

Aldershot.

Helmreich, R.L., Chidester, T.R., Foushee, H.C., Gregorich, S. and

Wilhelm, J. (1990), 'How effective is CRM training?', Flight Safety
Digest, vol. 9, no. 5, May. Flight Safety International, Arlington.

Huntington, H. (1990), CRM Training For Small Airlines, Proceedings of

the 7th General Flight Crew Training Meeting, International Air Transport
Association.

International Civil Aviation Organisation (1989), 'Flight crew training:

Cockpit Resource Management and Line Oriented Flight Training',
Human Factors Digest, no. 2, Montreal.

Jensen, R.E. and Biegalski, C.S. (1989), 'Aviation psychology' in R.

Jenson (ed.), Cockpit Resource Management, Gower Technical,

409
Aviation Instruction and Training

Aldershot, Hants, pp. 176-209.

Maher, J.W. (1989), Beyond CRM to Decisional Heuristics, Proceedings

of the Fifth International Symposium on Aviation Psychology, Ohio State
University, Columbus, 439-444.

NASA (1986), Cockpit Resource Management Training, Proceedings of

the NASA/MAC Workship, San Francisco, May, NASA Conference
Publication 2455, Ames Research Center, Moffat Field.

Pendarvis, C.W. (1989), CRM Is Not a New Word, Proceedings of the Fifth
International Symposium on Aviation Psychology, Ohio State University,
Columbus, pp. 451-455.

Schwartz, D. (1987), Training For Situational Awareness', Paper presented

to the Fortieth Air Safety Seminar, Flight Safety International, Houston.

Telfer, R. (1989), 'Pilot decision making and judgement', Aviation

Psychology, Jensen, R.S. (ed.), Gower Technical, Aldershot, pp. 154-
175.

410
28 Using a generic non­
normal checklist to
train crew
co-ordination skills
A.N. Johnston

This chapter discusses the rationale underlying the creation and use of a
generic non-normal checklist by Aer Lingus (the national airline of Ireland).
The checklist is a key element in the simulator training component of a
generic Multi-crew Cooperation Training Course (MCC: Joint Aviation
Authorities, 1992) undertaken by all pilots entering the airline. Those pilots
are normally ab initio trainees graduating from a fourteen-month full-time ab
initio training course with an augmented CPL/IR ground-school and 200
flying hours (Harms, 1988; Johnston, 1989).
On first entering the airline a six week generic training course is used to
provide a 'bridge' between ab initio training and airline operations (Johnston,
1991; LATA, in press). During this training there are approximately three
weeks of ground-school, and twelve four-hour simulator sessions using a
simulator operating in 'generic twin-jet' mode. On completing their training,
trainees progress directly to type transition training on Aer Lingus/Aer
Lingus Commuter aircraft types, ranging from the Saab 340 through to all
variants of the Boeing 737. Substantial reductions in type-specific training
costs and time were achieved after the introduction of this generic training.
For instance, the entire six weeks training period is recovered through
reductions of up to 40 per cent in the training times previously required.
The generic airline entry training conducted by Aer Lingus has a number
of similarities (and contrasts) with that proposed for introductory AQP

This chapter is adapted from a paper first published in the International

Journal o f Aviation Psychology, vol. 2, no. 4. With permission from
Lawrence Erlbaum Associates, Inc.
Aviation Instruction and Training

(Advanced Qualification Program; FAA, 1991). Aspects of the checklist

discussed here will be of interest to those developing generic AQP
checklists, and to those already providing airline transition ’bridge’ training in
ab initio pilot training courses (LATA, in press).

Training and checklist design philosophy

In preparing the overall training program, it was necessary to develop a

generic checklist which captured the essentials of normal and non-normal
airline checklists. A basic design objective for the full training course was
to maximize training transfer to all aircraft types operated by Aer Lingus,
while avoiding negative training transfer. As Aer Lingus operates both
turbo-prop and turbo-jet aircraft, this presented the occasional challenge. In
practice, however, developing and refining a generic 'normal' checklist was
relatively straightforward and this will not be discussed further. On the
other hand, while the design of a generic non-normal checklist presented a
singular challenge, it also offered unique training opportunities.
During the early design phase it was determined that the simulator
training of preference would expose trainees to a representative series of
normal and non-normal operational situations. These would be flown in real
time as LOS (Line Operational Simulations) with periodic use of LOFT
(Line Oriented Flight Training) for quality control and confidence building
purposes - since LOS permits instructional intervention, while LOFT does
not (Lauber and Foushee, 1981; FAA, 1990). It was further decided that
during this training phase aircraft handling skills - traditionally the
predominant training concern with new pilots - would take second place to
multi-crew cockpit management skills. Cognitive task analysis (Reading,
1989; 1990) was used to isolate and abstract common elements in the non­
normal operational events covered in existing checklists and type
transition/LOFT training. By combining the common operational elements
of normal and non-normal events across various aircraft types, a
comprehensive and representative series of generic training scenarios were
developed (Schoemaker, 1989). It was eventually determined that just
under half of the simulator training sessions would be devoted to the
operational management of non-normal events. This, in turn, dictated the
development of a suitable non-normal checklist to support the non-normal
simulator training exercises.
The following general design criteria for the non-normal checklist were
first established:

412
 Using a Generic Non-Normal Checklist

1. it should facilitate successful operational management of the relevant

non-normal event by the crew;
2. it should not generate any negative training transfer and should avoid
extraneous technical detail;
3. it should mirror the general style of existing type-specific non-normal
checklists and contain both memory and 'read and do' items;
4. all memory items should have continuing relevance in normal airline
operations after the generic airline entry training phase concluded;
5. human factors principles and crew co-ordination issues should
explicitly figure in the checklist design and content;
6. all communications generated by non-normal events should be
addressed, including real time liaison with cabin crew and passengers;
and
7. there should be positive reinforcement of good operating practices
throughout i.e. training requirements should not lead to a temporary
suspension of operational 'reality'.

These basic design principles were applied when evaluating all suggested
checklist items.

Airline non-normal checklists

Airline non-normal checklists normally contain three distinct types of

information:

1. memory items - intended for completion by pilots from memory, and

subsequently re-checked by reference to the checklist itself;
2. second are 'read and do' items; these either augment memory items for
critical non-normal events, or provide a checklist of items to be
completed in the event of minor problems; and,
3. operational guidance, review or reference information.

Figure 28.1 illustrates a characteristic non-normal checklist for a critical

event (an engine fire). The boxed items must first be carried out from
memory and subsequently re-checked using the checklist, before moving to
the secondary or 'tidy up' items.
Figure 28.2 illustrates a typical non-normal checklist for a small
operational problem (an overheating air conditioning system). There are no
memory items and the relevant actions are accomplished only after reference

413
Aviation Instruction and Training

to the actual checklist. The checklists in Figures 28.1 and 28.2 are

Figure 28.1
Sample engine Are drill (items in the box must be memorised)

illustrative variations of actual non-normal checklists, though they not

actually applicable to any particular aircraft type.

Non-normal checklists: the operational and training context

Checklists are basically intended to support pilots in accomplishing various

technical tasks. In this regard the fire drill detailed in Figure 28.1 is a
typical non-normal checklist item. At first sight it meets all essential

414
 U s in g a G e n e r ic N o n -N o r m a l C h e c k lis t

Figure 28.2
Sample air conditioning overheat drills (no memory items)

requirements. In other words, it provides a list of actions necessary to

protect the aircraft (the memory items) and ensure that it is appropriately re­
configured for safe operation (the 'tidy up' items). However there are certain
basic assumptions which surround and precede the use of this checklist.
While these are in one sense obvious, the accident record suggests that they
are not always understood and actioned effectively by pilots. It is, for
instance, assumed that:

- the aircraft is being flown properly and situational awareness

maintained throughout;
- the drill is being executed on the correct engine;
- the captain has ensured an appropriate division of tasks between crew­
members;
- the crew will promptly and effectively address all operational
consequences of the failure; and that
- appropriate communications and coordination with the passenger cabin
will occur.

Clearly these contextual issues are not immediately relevant to the fire

415
Aviation Instruction and Training

drill itself. On the other hand, the fire drill only forms part - essential though
it may be - of the overall management of the operational dilemma faced by
the crew. It is, however, a reality that traditional simulator training for initial
type transition emphasizes technical drills in preference to seeking fidelity
with operational reality (Vicente, 1990). The application of various drills in
operationally realistic circumstances comes later - either in LOFT, or in 'real
world' experience. The cost of simulator time invariably ensures that type
transition training is focused on technical issues and aircraft handling.
Technical failures are deemed to finish when the pilot handles each
operational/training problem as a technical/manipulative issue. Hence the
training context for initial type training is often that of the re-positioned
simulator, boxes to be 'ticked', circuit breakers which never successfully 're­
set', technical problems which never reach the stage of actually using the
relevant operational documentation, and so forth.
The implicit assumption underlying these training emphases is that pilots
will have little problem in applying their technical training to problems in the
real operational world. It is also implied that pilots will readily link, in real
time, the discrete sub-tasks required to effect a successful operational
outcome to any technical problem. There is also the implication that co­
pilots will 'pick-up' the required insights and skills as they gain 'experience'.
Errors and deficiencies dating from early training exposure are often
assumed to have been 'over-written' by such means.
All of these assumptions are challenged in the training discussed here.
The training principles underlying the Aer Lingus generic training course
derive from a belief that the basic building blocks of optimal operational
behaviour need to be explicitly taught and practised before the first type
transition. It is further believed that the accumulation of aeronautical
'experience' can be accelerated through carefully structured training (Ortega,
1989). Additionally, if we want pilots to perform as a crew - as team
members - we should train them as a crew (Hackman, 1988), while
providing an appropriate emphasis on crew decision making behaviour
(Orasanu and Salas, in press). In the light of these training imperatives, all
the simulator training scenarios discussed here have been scripted in
considerable detail (Schoemaker, 1989; Thordsen and Klein, 1991).

Developing the generic non-normal checklist

The criteria above suggest that an effective generic non-normal checklist

should seek to assure certain aspects of crew performance, namely that:

416
 Using a Generic Non-Normal Checklist

1. one pilot is explicitly charged with flying the aircraft at any given
moment;
2. there is explicit agreement as to which tasks are to be accomplished by
each pilot;
3. there is explicit agreement as to the nature of the technical problem;
4. critical actions are actively monitored by the non-actioning pilot;
5. there is an explicit review of the operational consequences of any
technical failure; and that
6. appropriate communications with cabin crew and passengers are
initiated in an appropriate and timely fashion.

In this context the word 'explicit' simply means that the relevant activity
should be verbalized and actively co-ordinated between crew members. A
further ramification is that successful accomplishment of the checklist should
itself invoke crew communication and discussion. This should assure crew
coordination (Helmreich, 1979) and encourage joint situational assessment,
shared mental models of operational circumstances (Cannon-Bowers, Salas
and Converse, 1990; Orasanu, 1991), and focused planning aimed at the
successful resolution of any ensuing operational problems (Pepitone, King
and Murphy, 1988). Successful practical application of a suitable non­
normal checklist during training should also help build a general operational
framework or schemata (Brewer and Dupree, 1983) to help with future
problem management - mainly by providing generic mental 'triggers' to
facilitate anticipation of essential operational planning and management
tasks (Geiselman and Samet, 1980). It is believed that the relevant generic
insights and skills will transfer to normal airline operations and facilitate
recognition-primed decision-making (Klein, Calderwood and MacGregor,
1986). While this has yet to be definitively verified for the checklist
described below, there is no reason to anticipate that the level of training
transfer here will differ from the current simulator training norm.

Content and application of the non-normal checklist

The normal and non-normal checklists used during simulator training are of
different colours. They are designed to fit back-to-back in a transparent
page sized plastic cover. The current version of the full non-normal
checklist is shown in Figure 28.3. Memory drills are boxed and captioned
as illustrated in Figure 28.1. 'Read and do' drills are captioned as illustrated

417
418

Figure 28.3
Aer Lingus generic non-normal checklist
 Using a Generic Non-Normal Checklist

Figure 28.4
Generic Engine Failure and Shutdown check

in Figure 28.2. The first drill in the actual generic checklist, for an Engine
Failure and Shutdown, is shown in Figure 28.4. Those familiar with
Boeing checklists will recognize the general structure of this particular
check.
The three items in the box are the memory items for an engine failure,
while the subsequent items are ’read and do' elements of the check, designed
to 'tidy up’ after an engine failure or shutdown. The memory items are first
articulated by the flying pilot, who initiates the relevant actions and then
calls for the checklist to be read. The thrust of the check is to assure
appropriate crew communications and co-ordination. It will be noted that
the engines of the generic jet are defined to have an 'auto shut-down’ feature
and that the 'tidy up' drill is itself brief. This avoids extraneous technical
detail and ensures that only those checklist items common to all Aer Lingus
aircraft types are physically accomplished. By this means the checklist aims
to fulfil essential generic training and operational objectives, while avoiding
any emphasis upon technical requirements. The One Engine Inoperative
Landing checklist is a 'read only' checklist, and this is used during the
subsequent approach and landing. It has certain flight management

419
Aviation Instruction and Training

implications, which are considered below following the discussion of the

hydraulic system failure drill.
Figure 28.5 shows the generic Systems Malfunction checklist. This is
used in the event of a failure or malfunction of any aircraft system or flight
instrument.
It can be seen that the first three memory items mirror those used in the
Engine Failure and Shutdown checklist. The additional two memory
items ensure that circuit breakers are checked and that any operational
consequences of the failure arise promptly for crew discussion and
resolution. In the early stages of the Aer Lingus generic training course the
relevant circuit breakers always reset, thus providing positive feedback on
the successful completion of this particular checklist item. It will be noted
that there is only one 'tidy up' item required in this check - namely to switch
off the relevant system - something normally encountered in aircraft type-
specific checklists. It is felt that these memory items provide an effective
generic response to systems malfunctions, facilitating both simulator training
and the longer term operational needs of trainees.

Figure 28.5
Generic Systems Malfunction check

420
 U s in g a G e n e r ic N o n -N o r m a l C h e c k lis t

The third of four memory checks in the non-normal checklist is the Ground
Proximity W arning check. It reads: 'Simultaneously execute a positive
pull up and apply maximum power. Climb at best angle until warning
ceases'. This check simply enshrines Aer Lingus company policy in respect
of ground proximity warnings. It is introduced as a memory drill at an early
training stage in order to emphasize the critical importance of an immediate
response to ground proximity warnings. However, trainees do not receive a
ground proximity warning in the simulator training unless their conduct of a
LOS or LOFT merits such a warning. Contrived situations and warnings are
explicitly avoided during LOS/LOFT simulator training.
Figure 28.6 illustrates the Hydraulic System Failure checklist. This is a
'read and do' check. As can be seen it is very short, though some
background technical information is provided to trainees elsewhere. The
intent of this check is to introduce changes to the appropriate actions and
procedures for approach and landing; this then generates a need to plan for
the new operational circumstances. As the standby hydraulic pump in the
generic jet is defined to have limited capacity, the landing gear must be
selected down first, and a two minute delay is required before flaps can
subsequently be selected. The combination of the two minute delay and
reversal of the normal order for gear and flap selection impacts upon
planning for the approach and landing. As the landing must be made with

Figure 28.6
Generic Hydraulic System Failure check

421
Aviation Instruction and Training

less than normal flap, it is also necessary to check the landing distance
available - hence the guidance to check the Operational Procedures
Manual 1-4-4; this is a generic aircraft performance manual, itself based
upon the manual format used by Aer Lingus. Page 1-4-4 of the manual
contains information on the required landing distance in various weather
conditions when using 'Flap 2’. The manual also contains information on the
landing performance implications of using 'standby brakes'. An additional
consideration to be taken into account by trainees is the absence of nose
wheel steering after landing. This means that the aircraft cannot clear the
runway unaided. Appropriate forward planning will take this into account
and timely communications will be initiated with Air Traffic Control.
While this check contains only three lines, their operational impact is
considerable. There is a need for trainees to manage a wide number of
operational issues and anticipate the requirements for a successful
conclusion to the approach and landing. While it is short on technical detail,
the check introduces all issues relevant to the successful operational
management of a similar problem in the real operational world.
Figure 28.7 illustrates the final memory drill in the checklist, the
Emergency Briefing: Cabin check. This checklist format is used as a
template for all communications between the cockpit and the 'senior cabin
attendant', in the event of any non-normal or emergency event. The format
mirrors that already in use by Aer Lingus and is fully compatible with the
joint training given to cabin and cockpit crewmembers. In the event of an
emergency Aer Lingus cabin crew are taught to use these four checklist
items to ask questions, should the cockpit crew fail to action them or
communicate effectively. It can be seen that this established practice
enables the simulator instructor to continually interrogate the trainees during
LOS/LOFT, should their execution of the check be deficient. The 'tidy up'
items in this checklist speak for themselves. The PA (Public Address)
announcements at two minutes and at thirty seconds prior to landing are part
of the cockpitxabin co-ordination procedures used in Aer Lingus. This
check thus elevates the existing Aer Lingus procedure to the status of a
'memory drill' for the purposes of the generic training course. The use of this
check by e ach trainee is required on a number of occasions, including
circumstances in which there are continuing reports of steadily increasing
smoke in the passenger cabin and changes in the time available before
landing.

422
 Using a Generic Non-Normal Checklist

Figure 28.7
Generic Em ergency Briefing: Cabin check

Using the generic non-normal checklist

Each trainee flys one each of the representative series of non-normal

LOS/LOFT scenarios, acting in the role of 'co-pilot flying'. When a
colleague flies seats and the trainee plays the role of 'captain, acting as 'co­
pilot'. Their preparatory training includes role definition and execution as
both captain and co-pilot. Contrary to some expectations, this has been
achieved with minimal difficulty and it presents a number of training
advantages.
The non-normal checklist provides the new airline pilot with a concise
introduction to the use of non-normal checklists within a multi-crew
operational environment. The checklist facilitates the completion of a
comprehensive spectrum of operational problems during actual simulator
training. It should also be mentioned that several other non-normal events
are addressed during training, for instance the operational and administrative

423
Aviation Instruction and Training

actions which follow a high speed rejected take-off. A number of other

practical issues arise because training exercises are full mission simulations
conducted in real time - an example is the need for trainees to make
appropriate public address announcements to passengers.

The impact of the checklist

The checklist meets the original design objective of avoiding technical detail
while facilitating joint crew member problem management in an authentic
real time operational environment. It also generates crew member
communication, problem resolution and forward planning. It thus enables
the achievement of Aer Lingus' overall training objectives for airline entry
training. It is difficult to isolate the training impact of the checklist from the
various other training elements contributing to the overall training course.
For instance, successful management and completion of a non-normal event
requires the application of traditional manipulative, procedural and
operational skills. Indeed, it is arguable whether the impact of any one
element of the training can be isolated from the other training elements
(Lave and Wenger, 1991). An important consideration to be kept in mind is
that the changes in training practice discussed here took place within an
applied industrial setting where experimental control was not considered
relevant. However, notwithstanding such limitations, it is hoped to clarify
some issues of potential interest to students of training transfer in a
forthcoming paper.
What can be said is that subsequent observations by type-specific
instructors indicate that these trainees have markedly improved cockpit
management skills in comparison to their predecessors. In addition, their
joint 'error trapping' and problem resolution skills are notably better than
those of many 'experienced' pilots (personal communication from a senior
flight instructor). In many ways the potential merits of the non-normal
checklist speak for themselves. After all, it simply formalizes and structures
pilot training in essential multi-crew pilot competencies. It will be noted that
these competencies are essentially those involved in CRM (Cockpit
Resource Management; FAA, 1989). However, in this Aer Lingus training
course these competencies are not addressed within a conventional CRM
training format. This raises a number of issues which hopefully will be
discussed in a future paper.

424
 Using a Generic Non-Normal Checklist

Conclusion

Two significant advantages follow from the use of the checklist and the
allied training scenarios.

Building multi-crew cockpit management skills

First is the rapidity with which trainees grasp the realities and implications
of non-normal multi-crew aircraft operations. The training format rapidly
breaks down the individualistic ethos which inevitably characterizes most ab
initio training (Boeing, 1990). Co-operative cockpit management and joint
error identification rapidly becomes the norm. The success of this change in
operating ethos only becomes clear to the trainees following the
'incapacitation' of their 'captain' during the latter simulator sessions. Because
the training involves the structured management of a wide spectrum of non­
normal operational events, many of the underlying problem areas identified
in past accidents and incidents are addressed in a highly effective manner.
Joint situational awareness and focused inter-crew communications are
married to practices which structure actual problem management. This
engenders early recognition of the operational and management demands of
various technical problems. In turn, this further generates forward planning
and anticipation. These problem management issues all arise in a full
mission, real time, multi-crew operational environment, with 'passengers'
and 'cabin crew' fully integrated into the training situation. This rich learning
environment for new pilots is experientially based and has a high fidelity
with the real operational world. The training ends with a challenging LOFT
session, which includes a single-engined diversion to an alternate and
various flight planning problems. That this can be successfully completed
by trainees who have yet to undertake their first transition to a multi-crew
airliner suggests that they are now able to produce generic responses to a
wide range of operational problems. The working assumption underlying
this training phase is that these newly acquired skills will subsequently
underpin their operational behaviour. Certainly, it seems reasonable to
expect that these skills will more readily emerge in those who have had such
training exposure at a formative stage in their early training.

Innovations to initial type transition training

A second significant advantage to this training is now manifesting itself.

Trainees complete their initial training with a wide repertoire of basic

425
Aviation Instruction and Training

cockpit management skills. They also know, and can operationally apply,
all the relevant Aer Lingus standard operating procedures and callouts. The
Aer Lingus Boeing 737-200 fleet sought to build upon these training
advantages in preference to seeking further reductions in simulator training
time. It was decided to completely rewrite the initial transition syllabus to
this aircraft, with a view to determining if die advantages of the initial entry
training could be further exploited. This resulted in a highly innovative
initial transition syllabus, in which traditional stimulator training practice has
been largely ignored. The new syllabus calls for eight simulator sessions.
Only three of these are similar to traditional Manoeuvre Oriented Flying
Training (MOFT). The remaining five sessions involve Line Operational
Simulations (LOS). This means that during initial type transition, normal
flight planning and operational documentation become part of the actual
simulator training situation. It further means that many of the individual
training topics - for example, electrics and hydraulic failures - are first
encountered in a realistic line context, where both the technical problem and
the real time operational management of that problem are treated as an
integrated unit for training purposes. This training approach was successfully
tested on one cohort of trainees and has now been extended to other Aer
Lingus fleets. That 200 hours ab initio pilots can transition to their first jet
aircraft using such an eight session type-transition makes its own statement
as to the efficiency of the non-normal checklist and the overall training
course described above. As a result of these successes the training course
was subsequently tested on two pilots approaching command. This also was
very successful and formal adoption of such training for all future command
upgrade candidates is now pending.

References and further reading

Boeing Commercial Aircraft Company (1990), 'Cockpit management

techniques', Boeing Flight Operations Safety Seminar, Paris, Seattle,
Author.

Brewer, W.F. and Dupree D.A. (1983), 'Use of plan schemata in the recall
and recognition of goal-directed actions', Journal o f Experimental
Psychology: Learning, Memory, and Cognition, vol. 9.

Cannon-Bowers, J.A., Salas, E. and Converse, S.A. (1990), 'Cognitive

psychology and team training: Training shared mental models of complex
systems', Human Factors Society Bulletin, vol. 33, no. 12.

426
 Using a Generic Non-Normal Checklist

Colley, A.M. and Beech, J.R. (1989), Acquisition and Performance o f

Cognitive Skills, Wiley, London.

Degani, A. and Wiener, E.L. (1991), 'Philosophy, policies and procedures:

The three P's of flight-deck operations', in Proceedings o f the Sixth
International Symposium on Aviation Psychology, Ohio State University,
Columbus.

Federal Aviation Administration (1989), Cockpit Resource Management

Training, Advisory Circular 120-51, Washington.

Federal Aviation Administration (1990), Line Operational Simulations,

Advisory Circular 120-35B, Washington.

Federal Aviation Administration (1991), Advanced Qualification Program,

Advisory Circular 120-54, Washington.

Fisk, A.D. and Rogers, W.A. (1988), 'The role of situational context in the
development of high-performance skills’, Human Factors, vol. 30.

Foushee, H.C. (1982), 'The role of communications, socio-psychological,

and personality factors in the maintenance of crew co-ordination',
Aviation, Space and Environmental Medicine, vol. 53, pp.

Geiselman, R.E. and Samet, M.G. (1980), 'Summarizing military

information: an application of schema theory', Human Factors, vol. 22.

Glickman, A.S., Zimmer, S., Montero, R.C., Guerette, P.J., Campbell, W.J.,
Morgan, B J . and Salas, E. (1987), The Evolution o f Teamwork Skills: An
Empirical Assessment with Implications fo r Training, Technical Report
87-016, Naval Training Systems Center, Human Factors Division,
Orlando.

Hackman, J.R. (1988), 'Crew coordination issues’, Address to the Seventh

General Flight Crew Meeting, International Air Transport Association,
New Orleans, Montreal.

Harms, D. (1988), 'Ab initio training at Bremen', in Report o f the Seventh

General Flight Training Meeting, International Air Transport
Association, New Orleans, Montreal.

427
Aviation Instruction and Training

Helmreich, R.L. (1979), 'Theory underlying CRM training: Psychological

issues in flight crew performance and crew coordination', in Cooper,
G.E., White, M.D. and Lauber, J.K. (eds.), Resource Management on the
Flight Deck, NASA Conference Publication 2120, NASA-Ames Research
Center, Moffett Field.

Hoc, J.M. (1988), Cognitive Psychology o f Planning, Academic Press,

London.

International Air Transport Association, (in press), IATA ab initio Pilot

Training Course, International Air Transport Association, Montreal.

Johannsen, G. and Rouse, W.B. (1983), 'Studies of planning behavior of

aircraft pilots in normal, abnormal and emergency situations', IEEE
Transactions on Systems, Man and Cybernetics, SMC-13.

Johnston, A.N. (1989), Ά review of airline sponsored ab initio pilot training

in Europe', in Proceedings o f the Fifth International Symposium on
Aviation Psychology, Ohio State University, Columbus.

Johnston, A.N. (1991), An Introduction to APT, Aer Lingus Training

Document, Dublin, Ireland.

Johnston, A.N. and Mauriflo, D. (1990), 'Human Factors training for aviation
personnel', ICAO Journal, May.

Joint Aviation Authorities (1992), Flight Crew Licencing Proposals (fourth

draft), JAA, Hoofddorf, Netherlands.

Klein, G.A., Calderwood, R. and MacGregor, D. (1986), 'Rapid decision

making on the fire ground', Proceedings o f the 30th Annual Meeting o f
the Human Factors Society, Human Factors Society, Dayton, Ohio.

Lauber, J.K. and Foushee, H.C. (1981). 'Guidelines for line-oriented flight
training’, NASA Conference Publication 2184, NASA-Ames Research
Center, Moffett Field.

Lave, J. and Wenger, E. (1991), Situated Learning, Cambridge University

Press.

428
 Using a Generic Non-Normal Checklist

Orasanu, J. (1991), 'Information transfer and shared mental models of

decision making', in Proceedings o f the Sixth International Symposium on
Aviation Psychology, Ohio State University, Columbus.

Orasanu, J. and Salas, E. (in press), 'Team decision making in complex

environments', in Klein, G, Orasanu, J. and Calderwood, R. (eds.).
Decision Making in Action: Models and Methods, ABLEX, Norwood.

Ortega, K.A. (1989), 'Problem-solving: Expert/novice differences', Human

Factors Society Bulletin, vol. 32, no. 3.

Pepitone D., King, T. and Murphy M. (1988), 'The role of flight planning in
aircrew decision performance', SAE Technical Paper 881517, Society of
Automotive Engineers, Warrendale.

Prince, C., Chidester, T.R., Cannon-Bowers, J. and Bowers, C. (in press).

'Aircrew coordination: Achieving teamwork in the cockpit', in Swezey, R.
and Salas, E. (eds.), Teams: Their Training and Performance, ABLEX,
Norwood.

Reading, R.E. (1989), 'Trainers teaching thinking skills: Applications of

recent research in metacognition to training', Proceedings o f the 33rd
Annual Meeting o f the Human Factors Society, Human Factors Society,
Santa Monica.

Reading, R.E. (1990), 'Taking cognitive task analysis into the field:
Bridging the gap from research to application', Proceedings o f the 34th
Annual Meeting o f the Human Factors Society, Human Factors Society,
Santa Monica.

Schoemaker, P.J.H. (1989), Scenario Thinking, Center for Decision

Research, Graduate School of Business, University of Chicago.

Thordsen, M.L. and Klein, G. (1991), Training implications of a team

decision model', in Proceedings o f the Sixth International Symposium on
Aviation Psychology, Ohio State University, Columbus.

Vicente, K.J. (1990), Ά few implications of an ecological approach to

human factors', Human Factors Society Bulletin, vol. 33, no. 11.

429
29 A skills-based
approach to training
aeronautical decision­
making
M. Wiggins andD. O'Hare

Over the previous two decades, errors in decision-making have been cited
consistently as the predominant cause in over 50 per cent of fatal general
aviation accidents in Australia, New Zealand and the United States (Copas,
1984; O'Hare, Wiggins, Batt and Morrison, in Press; Jensen and Benel,
1977). Despite this longstanding association however, aeronautical
decision-making remains a subject which is poorly understood and lacks the
theoretical framework upon which sound, prescriptive models of decision­
making can be built and tested. This paper presents a critical review of
previous research into aeronautical decision-making in general, and weather-
related decision-making in particular. An alternative approach to the study
of weather-related decision-making is proposed, based on the ACT* model
of skill acquisition. This tripartite model outlines the cognitive processes
involved in the transition from novice to expert, and the implications for
weather-related decision training are discussed in detail. It is hoped that this
paper will stimulate discussion in this important area, and lead to the
development of a structured training program for weather-related decision­
making, based on a well-founded theory of skill acquisition.

Visual flight rules and decision-making

According to the United States National Transport and Safety Board (1987),
weather conditions were a significant factor in approximately 24 per cent of
 A Skills-Based Approach to Aeronautical Decision-Making
general aviation accidents during 1983. Moreover, weather-related crashes
accounted for 40 per cent of general aviation fatalities and typically involved
non-instrument rated pilots who flew into meteorological conditions for
which they were not trained, and were legally obliged to avoid. The
difficulties associated with flying in the absence of terrain reference have
been emphasized clearly by Bryan, Stonecipher and Aron (1955) in a study
involving 20 pilots, none of whom held an instrument rating. Although a
number of subjects had extensive aeronautical experience flying in visual
conditions, 95 per cent entered a 'graveyard' spiral within eight minutes of
entering conditions which required sole reference to the aircraft's
instruments. The results of this study indicate that the majority of pilots
authorized to fly exclusively in visual conditions are markedly ill-prepared to
cope with the demands associated with operations in instrument or near
instrument meteorological conditions (IMC).
In a report issued by the British Civil Aviation Authority (1988), the
predominant causes of weather-related crashes in the United Kingdom were
attributed to errors of judgement due to the inadequate use of meteorological
information. Similarly, Rockwell and McCoy (1988) noted that a significant
proportion of pilots experienced difficulty in predicting accurately the
weather conditions during computer-simulated flights under visual weather
conditions. Typically, these pilots had accumulated relatively fewer hours of
cross-country flying experience and made a disproportionate number of
weather-related, in-flight errors including flight above a broken cloud base,
flight at improper altitudes with respect to the weather conditions, and flight
at altitudes with insufficient ground clearance. Collectively, these results
suggest that the majority of weather-related crashes are not the product of
unforeseen encounters with extreme weather conditions, but rather, are due
to systematic errors in the decision-making process through which pilots
recognize, diagnose and formulate strategies to cope with various weather
phenomena.
In the context of the present research, effective weather-related decision­
making is defined as occurring during operations under Visual Flight Rules
(VFR) in which a pilot recognizes and avoids conditions which are below
that necessary to maintain flight within visual meteorological conditions,
ensuring both the safety of the aircraft and its occupants, and achieving as
closely as possible, the stated objectives of the flight. VFR are those
regulations which make it incumbent upon pilots to maintain flight in Visual
Meteorological Conditions (VMC). The dimensions which define VMC
vary from country to country and may differ according to the altitude of the
aircraft. In Australia for example, VMC at or above 5000 ft, is said to exist
when the forward visibility is at least 8 kilometres, and vertical and

431
Aviation Instruction and Training
horizontal separation from cloud can be maintained at or beyond 1000 ft and
2000m respectively. Below 5000 ft, these dimensions reduce to 5000 m
foward visibility with 500 ft vertical separation, and 600 m horizontal
separation from cloud (Civil Aviation Authority, 1991)

The difficulty of learning from experience

It is tempting to view the problems of aeronautical decision-making as

arising from lack of experience, and to assume that as pilots gain greater
experience they will automatically learn to overcome the difficulties that
beset the neophyte. There are two major difficulties with this view. Firstly,
there is no evidence to suggest that older pilots or those with many hours
logged are under-represented in accident statistics. Pilots aged 40 to 55, for
example, made up approximately 32 per cent of the active US pilot
population between 1975 and 1986, but were involved in approximately 43
per cent of the VFR into IMC crashes during this period. It is the younger,
and on the whole less experienced groups, who are actually under­
represented in these accident statistics! Some epidemiological studies show
increased accident rates with age and cumulative flight time (Booze, 1977).
The second difficulty is that tasks which consist primarily of decision­
making and judgement show little evidence of learning from experience,
despite the high levels of confidence which people exhibit in their
judgemental ability. This has been labelled the 'illusion of validity'
(Kahneman and Tversky, 1973). For example, despite considerable
evidence showing that there is no relationship between clinical experience
and the validity of clinical judgement (Garb, 1989), clinician's faith in their
abilities remains undiminished. Einhom and Hogarth (1978) have suggested
that the difficulty of learning from experience arises because of the way
most judgemental tasks are structured. This can be seen most clearly in the
following hypothetical example involving aircrew selection, but the same
points apply to any other aeronautical judgement or decision-making task.
Suppose a prestigious airline recruits only pilots with a military training
background. A very high proportion, say 90 per cent, subsequently graduate
through training to line flying. The company can only observe two of the
relevant 'action-outcome' combinations: those who were selected and
succeeded, and those who were selected and failed. There is considerable
evidence that people are strongly biased towards the information concerning
positive liits' (i.e those selected who were successful), and as long as this
seems satisfactory, they will conclude that their judgement concerning the
value of a military training background is sound. The company is not in a

432
 A Skills-Based Approach to Aeronautical Decision-Making
position to observe the other two 'action-outcome' combinations which are
necessary to establish the validity of the judgement. These are the possible
successes and failures of non-military trained, and therefore non-selected
pilots. Whilst the 'positive hit' rate may seem to provide good evidence of
sound judgement, if 95 per cent of the non-military trained pilots would have
been successful, then the judgement is entirely worthless. Since it is very
difficult for anyone to be in a position to observe all four 'action-outcome'
combinations, it is easy to see why worthless judgements can so easily lead
to high confidence - the 'illusion of validity'.
Another excellent example of this illusion can be drawn from experiences
during flight training (O'Hare and Roscoe, 1990). Extreme responses, such
as an outstandingly good landing or a particularly poor attempt at a steep
turn, are likely to be followed by relatively less extreme responses - a poorer
landing or better steep turn - at the next attempt. If the good attempts were
praised by the instructor and the poor ones punished, then the instructor will
learn from experience that punishment is effective in improving performance
whilst reward is harmful. This is exactly the reverse of what the scientific
literature shows is actually the case.
In addition to the basic difficulty of being unable to observe the outcomes
of actions which were not selected, various additional factors such as
'treatment effects' (being selected and trained by a top company might
improve one's performance) and the chance element (poor decisions can
'luck out'), make it still more difficult to learn anything from experience at
all. The situation which typically prevails in flight training is one in which
pilots are expected to learn the subtleties of probabilistic judgement (e.g.
Will the en-route weather deteriorate?') by being given dichotomous
feedback (Yes/No; Good/Poor) in a situation naturally structured to prevent
learning at all
According to Rockwell and McCoy (1988), effective weather-related
decision-making is dependent not on the pilot’s total flying experience but on
the extent to which the pilot has been exposed previously to weather
phenomena during flight. Thus, irrespective of their qualifications, there
exists a preponderance of weather-related decision errors amongst pilots
with limited experience as the pilot-in-command of cross-country flights.
This suggests that existing flight-training perspectives may not prepare pilots
adequately to recognize and avoid situations in which the meteorological
conditions deteriorate below that necessary to maintain flight within VMC.
Indeed, incorrect and/or insufficient flight-training is at least partially
responsible for the majority of factors identified by Perrett (1991) as
contributing to weather-related accidents in Australia (See Table 29.1).
Most significant of these in terms of flight-training were the failure of the

433
Aviation Instruction and Training
pilot to conform to Visual Flight Rules; the lack of exposure to operations
at low level; the lack of experience in conducting precautionary landings;
and the failure to choose a flight-plan which minimized the effect of terrain
en-route.

Table 29.1
Factors identified as contributing to weather-related
crashes in Australia

- Pilots involved were generally low time, low experience.

- They felt an overwhelming desire to 'press on'.
- They felt subject to peer pressure, particularly from
passengers.
- They flew high-performance aircraft with which they were not
completely familiar.
- They did not know or wish to conduct a precautionary search
configuration.
- They did not plan to minimize the effect of terrain on their
route.
- They did not know, or conform to VFR.
- They had little or no exposure to low-level operations under
low cloud bases.
- They had little or no supervision from flying organizations.
- They had little or no experience in precautionary search
and landing procedures.

(Adapted from Perrett, 1991, p. 11.)

Whereas most aeronautical skills are taught explicitly during pilot

training, proficiency in weather recognition and avoidance is a skill,
generally assumed to develop implicitly as a pilot gradually accumulates
aeronautical experience (Moll, 1982; O'Hare and Roscoe, 1990).
Furthermore, the nature of weather-related decision-making is such that
creating situations which provide the cues necessary for instruction is
considerably more difficult than creating learning environments for other
aeronautical skills such as forced landing procedures. Consequently,
students are usually presented with a plethora of information relating to the
specifications of VMC, but rarely during their flying training is this

434
 A Skills-Based Approach to Aeronautical Decision-Making
information associated directly with the practical knowledge necessary for
effective weather-related decision-making.
The results described by both Rockwell and McCoy (1988) and Perrett
(1991) suggest that poor weather-related decision-making is not the product
of any single error, but is the outcome of a complex interaction between a
variety of factors, both intrinsic and extrinsic to the pilot. Motivation to
reach a destination, or 'get-home-itis' as it is known colloquially, is one such
factor which has received a great deal of emphasis in the accident prevention
literature. Although this represents a rather convenient explanation for
crashes involving VFR flight into IMC, it should not be regarded as a cause
since it simply replaces one unknown ('pressing on’) with another unknown
('get-home-itis'). Moreover, this line of reasoning does little to explain the
underlying cognitive processes which cause a pilot to risk flight through
meteorological conditions for which he/she has had little or no training.
Knowledge of these cognitive processes is vital in devising effective training
strategies to reduce the occurrence of crashes involving continued VFR
flight into IMC.

Theoretical accounts of aeronautical decision-making

According to Jensen and Benel (1977), effective aeronautical decision­

making is dependent upon the ability to derive and evaluate information
relevant to a critical situation and, on the basis of this knowledge, select and
implement actions which are both timely and appropriate. Consistent with
this definition, Wickens and Flach (1988) have developed a model of
aeronautical decision-making which comprises three distinct stages of
information processing: a situation assessment stage, decision formulation
stage and a response execution stage. During the initial, situation
assessment stage of decision-making, environmental cues are sampled and
the pattern of information matched against knowledge recalled from long­
term memory. Having diagnosed the situation, possible courses of action
are generated during the decision formulation stage. The selection of a
particular strategy is determined by its expected utility, a value defined as
the product of the extent to which the strategy satisfies a desired goal and
the probability that the outcome will occur under the circumstances
(Wickens and Flach, 1988). According to 'Subjective Expected Utility
Theory', the strategy with the highest expected utility is considered the most
favourable and is thus, the most likely to be executed during the response
stage.

435
Aviation Instruction and Training
Underlying Wickens and Flach's (1988) conceptualization of decision­
making is the assumption that effective decisions result from the careful,
deliberate analysis of both the environmental cues and the possible courses
of action which relate to a particular situation. This approach is
characteristic of an analytical class of decision strategies which includes
normative models such as Bayesian Theory and Decision Analysis (Klein,
1989; Kleinmuntz, 1990). Briefly, Bayesian Theory incorporates a
mathematical equation which enables the calculation of the probability that a
given hypothesis is true on the basis of prior evidence (Massaro, 1989). It is
particularly useful in conditions of uncertainty and has been applied
successfully to the prediction of suicides amongst psychiatric patients
(Meehl and Rosen, 1955; Gustafson, Tianen and Greist, 1986). Decision
Analysis is a well-established method derived from Subjective Expected
Utility Theory, in which complex problems are decomposed and the options
available to the decision-maker represented graphically. Each option is
assigned a utility value in conjunction with its probability of occurrence, and
their cumulative values are used to determine the most desirable option
under the circumstances.
Hammond, Hamm, Grassia and Pearson (1987) suggest that analytical
strategies occupy one extreme of a cognitive continuum, the other being
occupied by intuitive strategies. Whereas analytical strategies are relatively
time-consuming and require an evaluation of options concurrently, intuitive
decision strategies are generally less time-consuming and are characterized
by a serial evaluation of options (Klein, 1989). Recognition-Primed
Decision-Making (RPD) for example, is an intuitive strategy for which the
evaluation of options is conducted in series, through a process of mental
simulation (Klein, 1989; Klein and Klinger, 1991). More specifically, RPD
involves the location of previous experiences which embody features similar
to the present state, and the evaluation and application of those procedures
which were successful previously.
For several years, there has been considerable disagreement evident
within the decision-making literature regarding the relative efficacy of
analytical strategies over those which incorporate intuitive principles (Beech
and Mitchell, 1978; Waller and Mitchell, 1984; Wickens and Flach, 1988).
Waller and Mitchell (1984) for example, contend that intuitive strategies are
incapable of eliciting skilled performance since they are subject to biases
and thus, lack internal reliability. Under conditions of uncertainty, there is
clearly some support for this assertion, since individuals have been shown to
apply a limited number of heuristic principles which, in certain situations,
result in extreme and systematic errors (Tversky and Kahneman, 1986).
Moreover, Kahneman and Tversky (1973) note that the accuracy of intuitive

436
 A Skills-Based Approach to Aeronautical Decision-Making
judgements is influenced markedly by variations in fatigue, recent
experience, and the sequence with which information is presented to
subjects.
In 1954, Paul Meehl conducted a review of 20 clinical judgement studies
in which he concluded that analytical procedures for combining information,
consistently out-performed intuitive strategies in diagnosis and prediction.
Furthermore, Meehl (1954) established a methodological framework for
conducting evaluations between the two classes of strategies, and thereby
created the impetus for a vast array of subsequent comparative research.
One of the earliest, and most important studies in this regard compared the
accuracy of clinician's diagnoses of psychosis with those made by statistical
models from scores on the Minnesota Multiphasic Personality Inventory
(MMPI) (Goldberg, 1970). Surprisingly, a simple linear model,
incorporating the five MMPI scales, out-performed all 13 clinicians despite
their having access to identical information. Similar results are apparent in
variety of domains lending support to the argument favouring the efficacy of
those analytical strategies of decision-making for combining information
(see Dawes, 1971; Einhom, 1972).
According to Klein (1989) however, the efficacy of one class of strategies
over another is dependent upon both the characteristics of the decision-task
and the experience of the individual involved. Hammond et al. (1987) for
example, have demonstrated the efficacy of intuitive decision strategies
amongst expert highway engineers' predictions of highway aesthetics, safety
and capacity. Specifically, a within-subjects comparison was undertaken
between analytical, quasi-analytical and intuitive strategies, the results of
which provided empirical support for the relationship between the properties
of the task, the type of cognitive strategy employed by the decision-maker,
and the subsequent accuracy of judgements. Moreover, the results also
indicated that extreme errors in judgement were more likely to occur during
analytical rather than intuitive cognition.
A concern perhaps more fundamental than those discussed previously is
highlighted by O'Hare (in press) in suggesting that analytical models, such as
that proposed by Wickens and Flach (1988), are simply too cumbersome
and time-consuming to be effective in naturalistic situations requiring prompt
responses. This has led directly to O'Hare's (in press) development of an
alternative model of aeronautical decision-making based, in part, on the
intuitive principles of recognition-primed decision-making, espoused by
Klein and Klinger (1991). Unique in its approach, the ARTFUL model
brings together the notions of pattern recognition and planning in
accounting for decision-making behaviour during familiar and unfamiliar
situations respectively.

437
Aviation Instruction and Training
According to O'Hare (in press), decision-making is inherently, a goal-
directed process, the intention of which is to select responses likely to attain
the goals intended. During familiar situations where an individual has a
stored set of procedures or rules, the appropriate responses are activated
when certain preconditions have been met as in IF (condition) - THEN
(response)' statements. This is entirely consistent with the principles of RPD
in which it is proposed that proficient decision-makers generate options on
the basis of typicality or what is normally the response under particular
circumstances (Klein, 1989).
In situations where there exists some threat to the achievement of a
particular goal, O'Hare (in press) postulates the intervention of a planning
mechanism which regulates a search for alternative responses. Mediating
between pattern recognition and planning is a risk assessment process
whereby cues threatening the attainment of a desired goal are examined in
conjunction with the time available to conduct the search for alternatives.
Each alternative generated during planning, is evaluated in terms of its risk,
and those for which it is deemed negligible are accepted with a concomitant
change in the desired goal. Although the ARTFUL model of decision­
making has yet to be examined empirically, it nevertheless reflects a
growing awareness amongst researchers as to the significance of intuitively-
based decision strategies in naturalistic environments. The dearth of such
aeronautically-oriented theories however, has resulted in empirical research
concerned more with the development of an appropriate methodology rather
than the validation of decision-making theory (O'Hare, in press).

Weather-related decision research

Despite its significance as a causal factor in fatal aircraft accidents, the

characteristics of weather-related decision-making and in particular, it’s
association with certain in-flight errors, have been subjects largely under­
researched. Some progress however has been made towards an
understanding of weather-related decisions, due principally to a
comprehensive series of studies undertaken at the Ohio State University
(Giffin and Rockwell, 1987; Layton and McCoy, 1989; O'Hare, 1987;
Potter, Rockwell and McCoy, 1989; Rockwell and McCoy, 1988). Using
various editions of a computer-simulated flight task, researchers examined
pilot's acquisition of aeronautical information during a simulated flight
involving progressively deteriorating weather conditions. In each case, pilots
were able to seek weather information, make altitude selections and conduct
diversions at selected decision-points en-route.

438
 A Skills-Based Approach to Aeronautical Decision-Making
Initial versions of the computer-simulation were such that a substantial
proportion of pilots chose at the outset not to fly, since the weather
conditions at the departure point violated their own personal weather
minimums (Rockwell and McCoy, 1988). Not surprisingly, those pilots who
did decide to take-off were described generally as older pilots with
comparatively greater experience in flying under marginal weather
conditions. O'Hare (1987) found that those pilots who elected to take off
had higher scores on a measure of personal invulnerability from typical
accident causes and poorer knowledge of the phases of flight in which
accidents are most likely. There was some evidence to suggest that, of those
subjects who undertook the flight, pilots with instrument ratings sought more
weather information, committed fewer weather-related errors and were more
knowledgeable about weather than their VFR rated counterparts.
To ensure that subjects would undertake the flight, the revised computer-
simulation involved pilots flying from solid VFR conditions into
deteriorating weather conditions not forecast on departure (Rockwell and
McCoy, 1988). Thirty six VFR rated pilots with a low number of flying
hours were used in the study, the aim of which was to characterise the in­
flight errors associated with weather-related decision-making. A number of
distinct errors were noted including the failure to obtain trend information;
the failure to obtain information relating to alternates; and the failure to
develop a global depiction of the weather conditions. No difference was
evident between the demographic characteristics of those subjects who
made 'good decisions' and those who made decisions which were considered
inappropriate.
In a related study, Layton and McCoy (1989) used the computer-
simulated flight to model low time, VFR rated pilot's perceptions of
deteriorating weather conditions. Meteorological information was presented
in the form of menus from which a pilot could select reports at any one of 12
stations in the vicinity of the flight. Six of the 34 subjects were asked to
'think aloud' as they were performing the tasks whilst all pilots were asked a
series of questions following each decision point to determine that
information which was most influential in their decisions. According to
Layton and McCoy (1989), the information of most concern to pilots was
the current ceiling and visibility, which, in combination with subject’s
decisions, suggested that avoiding bad weather rather than seeking good
weather was the consideration uppermost in subject's responses.
On the basis of these results, Layton and McCoy (1989) proposed a
model of decision-making in which positive and negative values are assigned
to alternative destinations depending upon the weather conditions, distance
and azimuth. Using the author's terminology, pilots are repelled from the

439
Aviation Instruction and Training
'negative' zones and attracted towards the 'positive' zones, the repulsion
being stronger than the attraction for pilots with limited flying experience.
Unfortunately, the verbal protocols extracted during the study, failed to
elucidate any further, the cognitive processes which might underlie such a
theory.
Although these studies provide useful characterisations of weather-related
decision-making particularly amongst low-time pilots, the majority
emphasize symptomatic strategies such as the inappropriate acquisition of
information, rather than the fundamental cognitive processes from which
these strategies arise. Without knowledge of the root cause/s, it is difficult
to develop effective training procedures to combat inadequate weather-
related decision-making.

Aeronautical decision training

Due to the lack of well-founded, theoretical models of aeronautical decision­

making, the development of decision-related training programs has been
extremely limited. Arguably, the most extensive training program to date
has been devised by Berlin, Gruber, Holmes, Jensen, Lau, Mills, and O’Kane
(1982) under contract to the United States Federal Aviation Administration.
This research group was tasked with developing and optimizing training
materials which would augment the quality of pilot judgement in a variety of
aeronautical situations. Lester, Diehl, and Buch (1985) define pilot
judgement as:

... the mental process by which pilots recognise, analyse, and evaluate
information about themselves, the aircraft, and the operational
environment. The final step in the process is to make a decision pertaining
to the safe operation of the aircraft and to implement the decision in a
timely manner (Lester ei al., 1985, p.354).

The distinction between the terms judgement and decision-making is

unclear, although Telfer and Ashman (1986) suggest that judgement is a
broader term which encompasses both complex decision tasks and relatively
less complex perceptual-motor activities. Consistent with this assertion,
Berlin et al. (1982) target their training program towards a broad spectrum
of aeronautical activities ranging from decision skills during flight planning
to the location of anomalies during pre-flight inspections of aircraft. In this
regard, the program emphasizes a series of general strategies which are
presumed conducive to skilled judgement in a variety of circumstances.

440
 A Skills-Based Approach to Aeronautical Decision-Making
The pilot judgement training program comprises two sections, the first of
which defines the relevant terms and introduces concepts such as 'Action
Ways' and ’Judgement Chains'. According to Berlin et al. (1982), the six
Action Ways (Do, No Do, Early Do, Late Do, Under Do, Over Do) are
behavioural outcomes which follow errors of judgement. On this basis they
conclude that good judgement involves the performance of the action
required, combined with the performance of the action to the degree
necessary, and its execution at the appropriate time. The characteristics of
poor judgement chains are also stressed since previous research has
suggested that most accidents result from a sequence of judgement errors
rather than a single cause. In addition, the training program outlines a
general strategy through which chains of poor judgement can be recognized
and eventually broken.
Berlin et al. (1982) cited an extensive analysis of aircraft accidents
(Jensen and Benel, 1977) as a basis for their training initiatives, and
concluded that the most prevalent causal factors could be embodied within
the notion of 'Hazardous Thought Patterns'. The five types of thought
patterns and their respective antidotes are listed in Table 29.2. A major
proportion of the training program is devoted to the recognition of these
patterns, their significance in precipitating errors of judgement, and the
application of appropriate antidotes. In particular, the training manual
contains a number of scenarios which, in combination with in-flight
instruction, are designed to highlight the factors contributing to poor
judgement and provide the skills necessary for improvement.
Following a series of evaluations undertaken in the United States, Canada
and Australia (Berlin et al., 1982; Lester et al., 1985; Telfer and Ashman,
1986), some doubt has emerged concerning the efficacy of the training
program in augmenting pilot judgement within the operational environment
(O’Hare and Roscoe, 1990). In each of these experiments, groups of
control-subjects received normal flight training whilst an experimental group
received pilot judgement training in conjunction with the normal flying
curriculum. Comparisons between groups were made on the basis of their
scores on an academic test of pilot judgement and their performance during
an observation flight.
In the United States study, Berlin et al. (1982) reported a statistically
significant difference between the experimental and control group scores on
both the academic test and the observation flight. The observation flight in
this case comprised 20 situations during which the observer rated the
subject's judgement as either 'poor' or 'good'. Whereas Berlin et al. (1982)
utilized full-time university students, Lester et al. (1985) evaluated the
training program amongst Canadian students from conventional flying

441
Aviation Instruction and Training

Table 29.2
Hazardous thought patterns and their respective antidotes

Hazardous Example Antidote

Thought

Anti-authority 'Don't tell m e ...’ 'Use the rules: They are

usually right.'

External control 'What's the use?' Ί can make a difference.'

Impulsivity 'Do something.. quickly.' 'Not so fast. Think first.'

Invulnerability 'It won't happen to me.' 'Why not me? I'm human
too.'

Macho Ί can do it.’ 'Risks don't make me fly

better'.
(Adapted from Berlin et al., 1982, pp. 10-2)

schools. Using an abridged, 12-situation version of the original observation

flight protocol, they reported a greater percentage of correct responses in
subjects who had undergone judgement training, although the difference
between the experimental and control group was not statistically significant.
Telfer and Ashman (1986) noted a similar difference between the responses
of experimental and control subjects in Australia, however the limited
sample size in this study precluded the application of statistical analyses.
Undoubtedly, the variability between these evaluations reflects at least
partly, the difficulty in evaluating pilot judgement performance objectively
within the operational environment. In addition however, the pilot
judgement training program is based intrinsically on the assumption that the
skills taught will be generalized and applied immediately to novel judgement
tasks. This view of generalization contradicts a large body of literature
which suggests that skills taught in one domain are highly specific and only
become generalized following extensive experience with the specific
knowledge and the process through which that knowledge is applied
(Anderson, 1987; Gray and Orasanu, 1987; Simon and Hayes, 1976).
Perfetto, Bransford, and Franks (1983) for example, found that relatively

442
 A Skills-Based Approach to Aeronautical Decision-Making
inexperienced subjects having acquired knowledge in one situation, failed to
transfer the information to a novel situation even though the two possessed
similar characteristics. Consequently, the lack of success of the pilot
judgement training programs reviewed above, may reflect an inherent lack of
skill-generalization to novel, operational situations. Alternatively, the results
may also be indicative of the fact that the skills taught are not relevant to the
decision-making process.
The immediate transfer of general skills is an assumption also held by
Jensen (1988) in his preliminary evaluation of the DECIDE model of
effective pilot judgement. Adapted from research involving fire fighting
personnel, the DECIDE model consists of a sequence of six steps as
depicted in Table 29.3. Adhering to these steps ensures that information is
not overlooked and facilitates the decision-making process, particularly
during situations of uncertainty. Using a procedure similar to that employed
by Berlin et al. (1982), Jensen (1988) compared the judgements made by
control and experimental groups of private pilots, the latter of whom had
received five hours of instruction using accident scenarios to illustrate the
application of the DECIDE model.

Table 29.3
The DECIDE model of aeronautical decision-making

D - Detect Detects the fact that a change in the system state has
occurred.
E - Estimate Estimates the significance of the change to the flight.
C - Choose Chooses a safe outcome for the flight.
I - Identify Identifies plausible actions to control change in the
system state.
D-Do Acts on the option considered most suitable.
E - Evaluate Evaluates the outcome on the change in the system state.

(Adapted from Jensen, 1988, p. 3)

The evaluation was conducted during a simulated cross-country flight in

which subjects were presented with changes in the operational environment
designed to evoke the decision-making process. Of the five subjects in the
experimental group one chose not to fly, another turned back following a

443
Aviation Instruction and Training
failure of the attitude indicator, whilst the remaining subjects turned back at
the onset of deteriorating weather conditions. In contrast, three of the five
control subjects chose to fly, of which one lost control following the failure
of the attitude indicator, and the remainder 'crashed' having failed to
recognize the onset of carburettor icing. Due to the sample size and extreme
variability between groups however, statistical analyses could not be
performed.
The nature of Jensen's (1988) evaluation procedure is such that it remains
unclear as to whether the results were the product of a more effective
decision-making process, or whether judgement training induced a
heightened situation-awareness which encouraged subjects in the
experimental group to be more vigilant than those in the control group. For
example, two of the three subjects in the control group 'crashed' due to a
failure to recognize the onset of carburettor icing. Whilst it must be
acknowledged that recognition is an intrinsic aspect of decision-making, the
success or failure to recognize the onset of a change in the system state may
be attributable equally to the efficacy of the DECIDE model or to
differences in the sensitivity to changes in the operational environment
(Massaro, 1989). Thus, rather than improving their decision-making ability,
the training may simply have 'primed' the experimental group to be more
vigilant than their control group counterparts in recognizing and responding
to system failures. This suggests that the judgement-tasks fail to
discriminate effectively between the two theoretical perspectives and
consequently, leave the validity of Jensen's (1988) results in some doubt.
The inconclusive results arising from these evaluations lends support to
the assertion that both the pilot judgement training program and the
DECIDE model are conceptualizations which fail to capture effectively the
nature of pilot judgement and aeronautical decision-making. Indeed, O'Hare
and Roscoe (1990) suggest that the development of these programs has been
somewhat premature given the dearth of sound, theoretical research
concerning decision-making in the aeronautical environment. Moreover,
they note the scarcity of information pertaining to the characteristics of
expert pilots, and in particular, the process through which they formulate
decisions during flight.

Experts versus novices in the aeronautical domain

Dreyfus and Dreyfus (1986) have characterized expertise in a particular

domain as the ability to recognize spontaneously, similar classes of
problems, and apply solutions as appropriate. Novice performance in

444
 A Skills-Based Approach to Aeronautical Decision-Making
contrast, is characterized by the application of previously acquired, general
problem-solving strategies, irrespective of context. Consequently, the
performance of novices has usually been regarded as markedly inferior to
that of experts, although empirical research suggests that this is not always
the case. Giffen and Rockwell (1984) for example, noted that during paper
and pencil aeronautical simulations, neither the number of flight hours,
training, nor type of flying experience was, predictive of performance in
diagnosing critical in-flight events such as a vacuum pump failure or a frozen
static port. Despite such results however, Lintem (1986) has argued that the
identification of differences between expert and novice pilots should hold a
priority amongst researchers, since it provides a useful approximation of that
knowledge which must be learnt in order to achieve expertise.
Although researchers have yet to examine explicitly the characteristics
which differentiate between expert and novice weather-related decision­
making, a number of distinguishing features have emerged during the course
of related studies. Beck (1987) for example, compared the weather
forecasting procedures of high-time and low-time pilots, and noted
differences in both the accuracy of forecasts and the acquisition of
meteorological information. High-time pilots were defined as pilots whose
total cross-country flying experience exceeded 400 hours, whereas those
with fewer than 400 hours were designated as low-time pilots. In addition to
a superiority in forecasting performance, Beck (1987) noted that pilots with
relatively greater cross-country experience also utilized more effective
information search strategies than their low-time counterparts. This, the
author attributed to differences between the pilots' internal representation of
the principles which underlie the behaviour of weather systems.
Beck's (1987) findings are consistent with differences apparent between
experts and novices in other domains. Hershey, Walsh, Read and Chulef
(1990) for example, observed that in comparison with their expert
counterparts, novice financial planners accessed more information with
greater frequency, and were thus, less efficient in formulating solutions to
financial investment problems. Similarly, Chi, Glaser and Rees (1982)
reported differences between the information search strategies of expert and
novice physicists during their analyses of physics problems. Collectively,
these results suggest that in comparison with novices, experts possess a
more comprehensive understanding of the particular problem environment
which enables problem solutions to be formulated with greater efficiency.
Empirical support for this assertion can be derived from research in which
experienced and inexperienced subjects are asked to sort domain-related
concepts into as many discrete categories as they consider appropriate

445
Aviation Instruction and Training
(Glaser and Chi, 1988; Terranova, Snyder, Seamster and Treitler, 1989).
According to Glaser and Chi (1988), experts organize fewer groups, based
on the relationship between underlying principles, whereas novices construct
a larger number of categories, the basis of which are superficial features
embodied within the original problem-statements (Chi, Feltovich and Glaser,
1981; Larkin, McDermott, Simon and Simon, 1980).
Using a more naturalistic methodology than that employed by Beck
(1987), Barnett (1989) examined the performance of experienced and
inexperienced pilots during a simulated cross-country flight. The study was
designed to examine the validity of the Wickens and Flach (1988) model of
decision-making in differentiating between the characteristics of pilots with
greater than 1000 hours and those with fewer than 500 hours of total flying
experience. As mentioned earlier, the model is divided into three
information-processing components based on the assumption that decisions
comprise an assessment and diagnostic phase, a decision formulation phase
and a response execution phase.
According to Wickens and Flach (1988), situations with which a pilot is
familiar invoke the application of a previously established set of procedures
resident in long-term memory. Unfamiliar situations however, require a
more deliberate process in which alternative strategies are generated and
examined on the basis of probabilities and risk. Consistent with this
theoretical perspective, Barnett (1989) suggests that the performance of
experienced pilots is dependent upon their ability to represent the situation
accurately, since a correct diagnosis will lead to the activation of appropriate
procedures. In contrast, the performance of inexperienced pilots is
dependent upon information processing ability, since these skills are
necessary for the generation and evaluation of alternative strategies.
Barnett (1989) examined the knowledge representation and information
processing ability of thirty subjects using a battery of cognitive measures
including a test of mathematical reasoning and a memory-recall task
involving air traffic control messages. The values derived from these tests
were assumed predictive of performance on a series of aeronautical decision
scenarios which engaged diagnostic and information processing skills.
Scenarios comprised both routine decisions and in-flight emergencies for
which the performance was rated on the basis of optimality, confidence and
decision latency. Between-group comparisons however, failed to reveal any
differences between experienced and inexperienced pilots in either their
cognitive abilities or decision-making performance.
Nevertheless, some support for a relationship was evident through
subsequent within-group analyses which identified those cognitive variables
that best predicted performance during the decision scenarios. According to

446
 A Skills-Based Approach to Aeronautical Decision-Making
Barnett (1989), multiple regression analysis indicated that whereas measures
of information processing were amongst the most prevalent predictors of
inexperienced pilot performance, measures of knowledge representation
were predominant amongst the predictors of experienced pilot performance.
This qualitative dissimilarity in decision performance is consistent with that
proposed by Wickens and Flach's (1988) model of decision-making and
suggests that there exists a difference between the strategies of decision­
making employed by experienced and inexperienced pilots.
Support for the distinction between expert and novice decision strategies
can be derived from empirical research in domains such as trouble-shooting
and problem-solving, which demand information processing skills similar to
those required during decision-making tasks. Differences between expert
and novice problem-solving have been established both in the frequency of
hypothesis generation and the level of qualitative analysis undertaken prior
to the development of solutions (Glaser and Chi, 1988; Voss, Tyler and
Yengo, 1983). Voss et al. (1983) for example, noted that relative to novice
political scientists, a greater proportion of experts' protocols involved
elaborations on the original problem state rather than the formulation of
solutions. Similarly, a comparison between Fireground Commanders
indicated that experts deliberated on the nature of the problem more often
than did their novice counterparts during retrospective recall of real-life
critical incidents (Klein, 1989). These results suggest that whereas expert
problem-solving strategies are dependent upon the accurate diagnosis and
appropriate representation of a problem, novice strategies are dependent
upon the successful generation of numerous alternatives which can be tested
subsequently against a predetermined goal.
Using a model of aeronautical decision-making similar to that adopted by
Barnett (1989), Flathers, Giffen and Rockwell (1982) examined in greater
detail, the process of decision formulation amongst experienced and
inexperienced pilots during simulated deviations from planned flight. The
authors conceptualize the decision formulation process as a compromise
between the information derived during diagnosis, psychological factors
such as motivation and social influence, and the subjective 'worth' attributed
to various alternatives. The latter value is thought to be obtained through the
integration cognitively of relevant physical and descriptive information for
each alternative strategy. It is this process in particular, which Flathers et al.
(1982) suggest differs between experienced and inexperienced pilots and
accounts for differences in performance.
Thirty instrument-rated pilots participated in the paper and pencil
simulation involving a planned flight from Bangor, Maine to Glen Falls,
New York. Part-way through the simulation, the pilots were advised that

447
Aviation Instruction and Training
their alternator had failed and that the battery would last a maximum of 50
minutes before all navigation and communication equipment would become
inoperative. Pilots were thus made aware that a diversion was necessary
and that an alternative destination must be reached within a specified time­
frame. In addition, the nature of the introductory statement was such that
psychological factors and diagnostic ability were removed as confounding
variables from the decision formulation process. Consequently, individual
differences in the outcome of this process were considered solely the
product of individual differences in the value of the worth function.
Following the presentation of an introductory statement, pilots were
tasked with ranking a series of 16 alternative airports in order from the 'most
preferable' to the 'least preferable' on the basis of the attributes presented.
These varied across four orthogonal dimensions, each comprising two levels
with high or low relative values (See Table 29.4). According to Flathers et
al. (1982), the worth function associated with a particular airport is an
additive function which can be derived from the summed products of the
values for the four dimensions and their respective beta coefficients. These
coefficients represented standardized values of the relative importance of a
particular variable in accounting for the total sample variance. Regression
analysis was used to determine the estimates of the four beta coefficients for
each airport, and these values were combined subsequently to yield an
overall worth function for each subject.
For the purpose of statistical analysis, Flathers et al. (1982) divided the
sample on the basis of total flying experience, instrument flying experience
and single-engined aircraft experience using the 50th percentile as the point
of division in each case. Statistical analysis however, failed to reveal any
significant differences in the mean values for the beta coefficients.
Nevertheless, statistically significant differences were apparent in the type of
training and the type of flying in which pilots were most commonly engaged.
Civilian trained pilots for example, scored higher beta coefficients for the
presence of air traffic control facilities than their military trained
counterparts. Similarly, pilots for whom flying was related to business or
pleasure scored higher coefficient values for air traffic control facilities than
those pilots involved primarily in airline or military flying.
Significant differences were also evident between levels of pilot
certification with private or commercial pilots scoring higher beta values
than those with air transport licences for both weather information and the
time to reach the airport. On the basis of these results, Flathers et al. (1982)
conclude that during deviations from planned flight, pilots with relatively
greater levels of skill place less emphasis on factors such as air traffic
control facilities and weather information than pilots with lower levels of

448
 A Skills-Based Approach to Aeronautical Decision-Making

Table 29.4
The four dimensions and their respective levels which pilots used
to rate 16 airports from the most preferable to the least preferable
during a deviation from planned flight

1. Air Traffic Control a. Tower with Radar (High Value)

services b. Unicom (Low Value)

2. Airport weather a. 1000ft ceiling/ 3 miles visibility (High Value)

b. 500ft ceiling/ 1 miles visibility (Low Value)

3. Time to reach airport a. 15 minutes (High Value)

b. 30 minutes (Low Value)

4. Instrument approach a. Instrument Landing System (High Value)

facilities b. Non-directional Beacon (Low Value)

(Adapted from Flathers et al., 1982, p. 961)

skill. Whilst the external validity of this type of interpretation is arguable,

the results suggest that rather than the quantity of experience as measured by
total flying hours, it is the quality of that experience as indicated by the type
of training, which predicts the value of worth functions and thus, differences
in decision-making performance.
Giffen, Rockwell and Smith (1985) provide additional support for this
assertion through an evaluation of pilots' acquisition of meteorological
information during a computer-based simulated flight. Descriptions of the
weather conditions visible from the cockpit were provided in addition to
forecasts, advisories of significant meteorological activity, and the reported
winds aloft. Of the 19 subjects who elected to conduct the flight, the
characteristics of eight subjects whose weather-seeking strategies were
considered poor were contrasted with eight subjects whose strategies were
rated as good. Performance was graded on the basis of the frequency of
each of five weather seeking errors made prior to the take-off decision (See
Table 29.5).
According to Giffen et al. (1985), those subjects in the poor group had
accumulated fewer cross-country hours, achieved lower scores on a test of
aeronautical knowledge, and were less likely to have an instrument rating

449
Aviation Instruction and Training

Table 29.5
Weather seeking strategy errors observed during simulated
VFR cross-country flights

1. Failed to check destination weather.

2. Failed to have an escape route in case of deteriorating weather
en-route.
3. Incorrect altitude selection.
4. Failed to make use of all the information available.
5. Failed to obtain forecast weather en-route.

(Adapted from Giffen et al., 1985, p. 284)

than their counterparts in the good group. The results therefore suggest a
relationship between the efficacy of weather-seeking strategies and
aeronautical experience in situations which require the interpretation of
meteorological information. In generalizing these results however, it should
be noted that differences in the frequency of weather-seeking errors may not
necessarily reflect differences in the efficacy of weather-related decision­
making, since the validity of this relationship has yet to be established
empirically.
Compared with research in areas such as medicine or psychology, there
have been relatively few studies in the aeronautical domain designed to
differentiate between the performance of experts and novices. Nevertheless,
a number of useful differences have become evident through those
investigations which have been conducted within the aeronautical domain,
and these are summarized in Table 29.6. Consistent with research in other
areas, these results suggest that through experience, pilots develop a detailed
understanding of the problem domain and on this basis, devise problem­
solving strategies which maximize both accuracy and efficiency.
Whilst comparisons between experts and novices provide an indication of
the knowledge which must be gained to become an expert, they provide little
insight however, into the processes through which the transition from novice
to expert occurs. The development of instructional programs requires not
only information concerning the nature of expertise, but also knowledge of
the cognitive processes involved in the progression towards expertise. In
particular, a theoretical model of skill acquisition is required, which

450
 A Skills-Based Approach to Aeronautical Decision-Making
describes accurately the behaviour of decision-makers at various stages
during the transition from novice to expert. Establishing the validity of such
a model will provide the foundation necessary to facilitate the development
of expertise in weather-related decision-making.

Table 29.6
A summary of the differences which have been established
between experienced and inexperienced pilot performance

Experienced Pilots Inexperienced Pilots

More accurate meteorological Less accurate meteorological

forecasting ability forecasting ability

More efficient acquisition of Less efficient acquisition of

meteorological information meteorological information

Less frequent weather-seeking More frequent weather-seeking

strategy errors strategy errors

Decision performance predicted by Decision performance predicted

knowledge representation measures by information processing
measures

Detailed understanding of Limited understanding of

meteorological principles meteorological principles

Production systems approaches to skill acquisition

Production systems are distinguished from other models of skill acquisition

by the fundamental assumption that the cognitive processes which underlie
human behaviour operate as a sequence of condition-action rules. Each rule
comprises both a conditional statement, preceded typically by the
conjunction ’IF, and an action statement denoted by the adverb 'THEN'. In
some cases, more than one qualifying statement may be included either to

451
Aviation Instruction and Training
increase ('OR') or decrease ('AND') the range of application. PI and P2
below are examples of general production rules which might be used in
response to the failure of an aircraft's engine during a take-off roll:

PI
IF the goal is to respond to an engine failure during a take-off roll in
Aircraft X
and Action (IX) is the first response to an engine failure in Aircraft X
THEN execute Action (IX)

P2
IF the goal is to respond to an engine failure during a take-off roll in
Aircraft X
and Action (IX) has just been executed
and Action (2X) follows Action (IX) in response to an engine failure
in Aircraft X
THEN execute Action (2X)

Anderson's (1983) ACT* theory is a production-based model of skill

acquisition which includes a detailed representation of the learning
mechanisms involved in the transition towards expertise. Derived from
earlier research by Fitts and Posner (1967), it comprises three stages
arranged hierarchically; an interpretive stage, a knowledge compilation
stage, and a tuning stage. The main feature distinguishing ACT* from
other production systems is the delineation between declarative and
procedural forms of knowledge. Declarative knowledge comprises
statements of fact or static knowledge such as that derived from textbooks.
This type of information is represented in declarative memory as a series of
cognitive units containing elements with particular relationships (Anderson,
1983). Procedural knowledge in contrast, encompasses strategies which
specify the process through which declarative knowledge is applied to a
given task (Lesgold, 1988). According to ACT*, the relationship between
declarative and procedural knowledge is mediated by the information
present in working memory as illustrated in Figure 29.1.

Interpretive procedures

Anderson argues that during the initial stage of skill acquisition, incoming
information is encoded declaratively as a sequence of facts which, in
combination with existing productions, control goal-directed behaviour

452
 A Skills-Based Approach to Aeronautical Decision-Making
(Anderson, 1983). More specifically, existing, task-independent productions
such as general problem solving skills and analogy are used to interpret
the declarative knowledge resident in working memory and produce
behaviour which approximates that required for problem resolution. Such
strategies are referred to commonly as weak methods since they fail to take
into account the characteristics of a particular domain. They are thus,
neither as efficient nor as effective as the domain-specific productions which
emerge during later stages of skill acquisition.

Figure 29.1
The inter-relationship between the structural components
which comprise the ACT* model of skill acquisition

(from Anderson, 1983, p. 19)

During the interpretive stage, the particular strategy adopted during

problem resolution is determined by the form in which the declarative
knowledge was encoded into long term memory (Anderson, 1987).
Situations in which the declarative knowledge comprises templates and
worked examples are likely to invoke strategies such as analogy, which
transforms the solution for one problem into the solution for another. In
contrast, declarative knowledge in the form of rules is likely to induce

453
Aviation Instruction and Training
general problem solving strategies such as means-ends analysis, working
backwards or hill climbing (Anderson, 1983; 1987). Since some strategies
are more applicable than others in particular problem domains, the
application of an appropriate strategy is related intrinsically to the accuracy
with which the declarative knowledge is encoded into memory.
In comparison with problem solving during the later stages of skill
acquisition, performance at the interpretive stage is relatively protracted, due
to the time-consuming retrieval of declarative information from long-term
memory. Furthermore, the interpretive stage is characterized by a relatively
large number of goals which are expressed with greater frequency, reflecting
the small size of the steps which comprise interpretive productions
(Anderson, Greeno, Kline and Neves, 1981). Finally, problem solving
during the initial stage of skill acquisition requires that declarative
information be present in working-memory, a mental structure which is both
capacity-limited and fails to retain information over extended periods of time
(Anderson, 1983; Miller, 1956). Thus, the interpretive stage is
characterized further by the repetition of declarative information verbally to
ensure its availability for interpretive processing.

Knowledge compilation

Central to the ACT* theory is the presupposition that the development of

productions forms the foundation of expertise in a particular domain
(Anderson, 1982). Productions are condition-action pairs for which the
declarative and procedural knowledge relevant to a task has been integrated
into compact production rules which embody specific problem solutions
(Gray and Orasanu, 1987; Lesgold, 1988). These mental structures provide
the facility for experts to recognize, categorize and solve problems with
maximum efficiency and their emergence constitutes the knowledge
compilation stage of skill acquisition (Anderson, 1983).
The transition from the interpretive to the knowledge compilation stage is
a process which occurs gradually through the mechanisms of
proceduralization and composition. Proceduralization involves the
recognition of repeated associations between specific mental actions and
declarative memory patterns, and the creation of productions which
incorporate this relationship (Anderson, 1983; Lesgold, 1988). For
example, if the production PI, encountered previously, was applied
repeatedly following engine failures during take-off rolls in a Cessna 152A
(Cessna, 1981), the process of proceduralization would result in the
following transformation:

454
 A Skills-Based Approach to Aeronautical Decision-Making

P3 (Proceduralized)
IF the goal is to respond to an engine failure during a take-off roll in a
Cessna 152A
THEN Close throttle
given that as shown above:
Action (IX) = Closing the throttle

Since the transformed production incorporates discrete values rather than

variables, there is no requirement for the time-consuming retrieval of
declarative information from long-term memory. Similarly, proceduralization
eliminates the verbal rehearsal necessary to maintain declarative knowledge
in working memory.
In addition to proceduralization, knowledge compilation also comprises
the mechanism of composition whereby unnecessary steps are eliminated
from a sequence of productions (Anderson, 1983). More specifically,
composition involves the creation of a single production which embodies the
actions and objectives of a sequence of productions oriented towards a
particular problem domain. Thus, the initial productions PI and P2 can be
composed into the following production for which two condition statements
have been eliminated:

P4 (Composed)
IF the goal is to respond to an engine failure during a take-off roll in a
Aircraft X
and Action (IX) is the first response to an engine failure in
Aircraft X
and Action (2X) follows Action (IX)
THEN execute Action (IX) and then Action (2X).

Although the mechanism of composition reduces processing time

considerably, composed productions remain dependent upon the retrieval of
declarative information from long-term memory. Therefore, both
composition and proceduralization are necessary to transform productions
effectively from the general, interpretive stage of skill acquisition to the
domain-specific productions which characterize knowledge compilation.
Compilation eliminates the need to consult declarative memory, and results
in far fewer steps being required to perform an action. Practice, therefore,
speeds up the performance of a skill and its superiority over formal
instruction is an assertion fundamental to the ACT* model.

455
Aviation Instruction and Training
Tuning procedures

The final stage in the ACT* model of skill acquisition involves optimizing
productions through the tuning stage sub-processes of generalization,
discrimination and strengthening (Anderson, Kline and Beasley, 1980).
Generalization is a mechanism whereby the common elements of domain-
specific production rules are used to create domain-general productions
which can be applied successfully to novel situations (Anderson, 1983;
Colley and Beech, 1989). This is accomphshed through either replacing
constants with variables or deleting a condition clause as shown in the
example below:

P3a (Proceduralized)
IF the goal is to respond to an engine failure during a take-off roll in a
Cessna 152A
THEN Close throttle

P3b (Proceduralized)
IF the goal is to respond to an engine failure during a take-off roll in a
Cessna 172N
THEN Close throttle

Through generalization, ACT* will delete the condition clauses 'Cessna

152A' and 'Cessna 172N' and create the production:

P3a and P3b (Generalized)

IF the goal is to respond to an engine failure during a take-off roll
THEN Close throttle

To ensure that the structure of the original productions is not lost, the
generalization process must also determine whether constraints are
necessary and incorporate these into subsequent procedures. In the example
outlined above, the constraint '... during take-off roll in a single engined
piston aircraft' may be added to limit the possibility of over-generalization.
Contrasting generalization is the mechanism of discrimination in which
the application of productions is restricted to those situations in which they
have been successful previously (Colley and Beech, 1989). Discrimination
is activated when production rules are confronted with an impasse and
cannot be executed (Lesgold, 1988). If feedback concerning the correct
action is available, action discrimination will be initiated and a specific
production formulated which recommends the appropriate action. If

456
 A Skills-Based Approach to Aeronautical Decision-Making
however, information concerning the correct action is unavailable, condition
discrimination will restrict the production to the last situation in which it
was applied successfully (Anderson, 1983). A schematic representation of
the progression towards expertise is shown in Figure 29.2.

Figure 29.2
The processes involved in the transition from the interpretive
stage to the tuning stage of skill acquisition

Strengthening is the final mechanism which comprises the tuning stage

and is based upon the assumption that productions possess different levels of
activation depending on the degree to which the condition clauses match
(Anderson, 1983). The role of the strengthening mechanism is to eliminate
those production rules which are either too general, too restrictive or simply
incorrect. Productions with relatively lower levels of activation will be
overshadowed during the matching process by productions with higher
strengths (Anderson, 1987). Prolonged disuse of productions leads to a
gradual decay in production strength and these are eventually discarded from
procedural memory. Due to the specificity and extensive range of
productions during the tuning stage, problem solving behaviour is
characterized by a relatively low decision-time coupled with a limited
information search.
The notion that expertise is characterized by an array of domain-specific
productions is consistent with the theory of decision-making under
uncertainty espoused by Klein and Klinger (1991) and O'Hare (In press).
Specifically, they assert that effective decision-making involves the location
of previous experiences which exhibit characteristics similar to the present
state and the application of strategies which have been successful

457
Aviation Instruction and Training
previously. Thus, it involves the recognition and classification of situations
as either familiar or prototypical. According to Klein (1989), the recognition
of a situation as familiar leads spontaneously to the recognition of plausible
goals, salient cues and typical reactions.
Since the recognition of familiar situations immediately evokes the most
likely course of action, decision-makers need only acquire information
sufficient to recognize it as comparable with previous encounters.
Consequently, expert decision-makers with a vast repertoire of experiences
need scan relatively fewer sources of information than those decision­
makers with limited experience.

The s-r-k model and the mental representations of experts and novices

Anderson's (1983) conceptualization of skill acquisition corresponds closely

to a tripartite model of human behaviour proposed by Rasmussen (1983;
1988) in which knowledge-based, rule-based and skill-based performance
relate to distinct levels of cognitive control. Knowledge-based performance
is similar to Anderson's declarative stage of skill acquisition, whereas rule-
based and skill-based performance equate with the knowledge compilation
and tuning stages respectively. The two models can be considered
complimentary, and whilst Anderson provides the most extensive account of
the cognitive processes which underlie skill acquisition, Rasmussen (1983)
outlines in greater detail, the changes in the mental representation of
information which occurs during the transition towards expertise. Thus, the
relationship between mental representations and performance can also be
considered within the theoretical framework of skill-acquisition.
According to Rasmussen (1983), knowledge-based performance
represents the highest conceptual level of cognitive control and arises during
unfamiliar situations for which an individual has no concomitant rules or
procedures. Behaviour at this level is oriented primarily towards the
satisfaction of goals which are formulated explicitly on the basis of both an
analysis of the particular situation and a consideration of the aims of the
individual. Since appropriate solutions to problems are not immediately
apparent during knowledge-based performance, a series of goal-oriented
plans must be developed and tested sequentially either through their physical
application and subsequent evaluation, or by predicting the outcomes
mentally (Rasmussen, 1983). The principles upon which these plans are
based, derive from the individual's conceptualization or mental model of the
inter-relationship between variables which comprise a particular problem
domain.

458
 A Skills-Based Approach to Aeronautical Decision-Making
Due to its computational and deductive nature, Rasmussen (1983)
suggests that the classes of mental models most effective during knowledge-
based reasoning are those which incorporate a functional representation of
the relationship between variables. It has been noted previously however,
that in comparison with their expert counterparts, novice representations of
problem domains are frequently composed of features which are relatively
superficial (See deGroot, 1966; Larkin et al., 1980). This, combined with
the dependence upon mental models to guide knowledge-based reasoning,
leads directly to the inference that novice problem-solving behaviour is
directed by features which are superficial to the problem, and is thus
influenced by the form in which information is presented. Such an assertion
is wholly consistent with the characteristics associated with Anderson’s
(1983) interpretive stage in which problem-solving strategies are determined
supposedly by the form in which declarative information is encoded into
long-term memory.
Like Anderson (1982), Rasmussen (1983) suggests that gradual increases
in experience give rise to the development of a repertoire of stored rules or
procedures which direct subsequent behaviour. This conscious ability to
both recognize familiar situations and implement appropriate problem­
solving strategies, constitutes the second, rule-based level of cognitive
control. In contrast to knowledge-based reasoning, goals are not formulated
explicitly at this level, but are determined implicitly through the execution of
procedures. Rule-based behaviour is thus, controlled by the perception of
salient cues, rather than an understanding of the functional principles which
underlie a particular situation. Nevertheless, it should be noted that the
formulation of effective rules demands that the functional inter­
relationships between variables be taken into account (Rasmussen, 1983).
Skill-based performance represents the final level in Rasmussen's (1983)
hierarchy of cognitive control and is characterized by the automatic selection
and application of appropriate strategies from an extensive repertoire of
procedures. Behaviour at this level is controlled continuously by the
perceptual-motor system which updates internal representations of the
environment, identifies changes in the system state and executes appropriate
responses accordingly. According to Rasmussen (1983), the flexibility of
skilled behaviour derives from the ability to select from a large repertoire,
the sequence of procedures best suited to satisfy the requirements of the
particular situation. However, the choice between procedures and thus, the
efficacy of skill-based performance, is instrinsically dependent upon the
flexibility and efficiency of a dynamic mental representation of the
functional relationship between variables.

459
Aviation Instruction and Training
Although Rasmussen (1988) makes few predictions concerning the
specific properties of knowledge-based performance, empirical research
provides some support for the relationship between mental models and the
choice of strategies for problem resolution. Hayes and Simon (1974) for
example, observed that despite a similar underlying structure, slight
modifications in the presentation of problems induced different mental
representations in inexperienced subjects. These differences were reflected
in the degree of difficulty perceived, and in subsequent problem-solving
performance. Table 29.7 provides a summary of the characteristics which
comprise the three stages of skill acquisition as detailed by Anderson (1983;
1987) and Rasmussen (1983).

Application of skill acquisition theory to weather-related decision­

making

Given the proposed division of skill acquisition into three distinct categories,
each mediated by well-defined cognitive processes, it is possible to deduce
certain behavioural characteristics likely to be exhibited by pilots of varying
levels of experience during weather-related decision-making. Pilots at the
interpretive stage for example, would initially be expected to recall from
long-term memory, any information considered relevant to the decision­
making process. In particular, an authoritative statement such as the
requirement for VFR rated pilots to maintain VMC may be recalled and
provide the basis upon which an initial sub-goal would be formulated.
Although the flexibility of the interpretive stage makes definitive
predictions difficult, it is likely that this sub-goal will be oriented towards
the location and avoidance of areas in which poor meteorological conditions
are forecast (cf. Layton and McCoy, 1989). Subsequent considerations may
include avoiding areas of high terrain or choosing between alternative
destinations on the basis of runway dimensions and fuel requirements.
Irrespective of their particular intention however, all sub-goals require a
comprehensive search for relevant information to determine those
alternatives which satisfy their preconditions. Furthermore, since this
information is encoded declaratively, it must be rehearsed to ensure that it
remains in working memory and is available for subsequent processing.
Weather-related decision-making at the interpretive stage can thus, be
characterized as a relatively time-consuming and ineffective process during a
situation in which the time available is limited and the possible
consequences disastrous.

460
 A Skills-Based Approach to Aeronautical Decision-Making
Table 29.7
Summary of the behavioural characteristics compiled from the
three stages which comprise Anderson's (1983) ACT* model of skill
acquisition and Rasmussen's (1983) S-R-K model of human behaviour

Stage 1 Stage 2 Stage 3

(Novice) (Intermediate) (Expert)

Extensive verbal No verbal rehearsal of No verbal rehearsal of

rehearsal of declarative knowledge declarative knowledge
declarative knowledge

Extensive information Moderate information Limited information

search search search

Rules expressed Rules expressed Rules not expressed

Rules associated with Rules associated with Rules not expressed

general cues specific cues

High decision time Moderate decision Low decision time

time
General problem General problem
solving reflects - solving does not
performance reflect performance

Mental representation - Mental representation

affects task has no effect on task
performance performance

Following repeated exposure to situations involving weather-related

decision-making, pilots would be expected to progress to the knowledge
compilation stage of skill acquisition and develop productions which
incorporate both the conditions signifying the requirement for a weather-
related decision, and the actions which are most effective in dealing with the
particular situation. These procedures eliminate the time-consuming
information retrieval process and the verbal rehearsal necessary during the

461
Aviation Instruction and Training
interpretive stage. Thus, knowledge compilation should elicit gains in both
the efficiency and the efficacy with which pilots conduct the weather-related
decision-making process. It should be noted however, that although there are
significant gains over the interpretive stage in terms of efficiency, knowledge
compilation does not guarantee error-free performance.
Whereas errors during the interpretive stage can be described as the
product of ineffective problem-solving strategies, errors at the knowledge
compilation phase can be characterized as the result of inappropriate
problem-solving strategies. Rasmussen (1988) for example, lists a number
of errors associated with the knowledge compilation/rule-based stage, of
which the most serious is the failure to recognize differences between
previous encounters and the features which characterize the present state.
Therefore, a pilot may apply a well-established and previously successful
rule to a weather-related decision only to find that slight differences between
previous situations and the present circumstances render the rule
inappropriate. In the aeronautical domain, inappropriate decision strategies
will usually lead to errors more serious than those resulting from ineffective
strategies.
Empirical support for this assertion can be derived from an examination of
aircraft crashes conducted by O'Hare, Wiggins, Batt and Morrison (In press)
in which accidents were classified according to a six stage taxonomy of
errors. These included the failure to recognize or diagnose the situation
correctly, the failure to select a procedure, strategy or goal appropriate for
the circumstances, or the failure to implement a required action. Analysis of
261 crashes showed that whilst procedure-selection and action or
implementation errors become relatively less frequent with increasing
cumulative flight time, the relative importance of goal and strategy-selection
errors was greatest at intermediate levels of between 100 and 1000 hours of
total flying experience. Moreover, 24 per cent of crashes involving goal
errors were fatal, compared to only 7 per cent for other kinds of errors.
O'Hare et al. (In press) suggest that the prevalence of goal and strategy-
related crashes amongst 100-1000 hour pilots reflects a stage of
development in which pilots are no longer being guided by their instructors,
but have yet to develop the skills necessary to select appropriate goals and
strategies independently. This is consistent with the notion that errors are
most likely to occur during the intermediate stage of skill acquisition in
which procedural knowledge is being developed (Anderson, 1983;
Rasmussen, 1988).
According to ACT*, the progression towards skilled weather-related
decision-making will be dependent upon the development of a series of
productions which can be applied successfully to a wide variety of

462
 A Skills-Based Approach to Aeronautical Decision-Making
situations. This constitutes the tuning stage of skill acquisition and requires
a considerable level of experience with situations in which the particular
skills are applied, and feedback is provided concerning the outcome. In
contrast to the interpretive and knowledge compilation stages, productions at
the tuning stage represent a compromise between general and specific
productions, and ensure an optimum level of operating performance.
Weather-related decision-strategies should thus, be both effective and
appropriate, irrespective of the conditions which constitute the particular
decision environment. More specifically, the expert weather-related
decision-maker should be able to recognize accurately, those meteorological
conditions which are below that necessary to maintain flight within VMC,
ensuring both the safety of the aircraft and passengers, and achieving as
closely as possible, the stated objective of the flight.

Implications for flight-training

Due to the dynamic and time-constrained nature of aeronautical operations,

flight training for the most part, has been designed to provide pilots with
both the declarative and procedural knowledge necessary to recognize and
respond to situations efficiently and effectively. In this regard, the student is
provided with ground instruction initially, in which the declarative and
procedural knowledge relevant to a task is introduced and explained in
detail. Subsequent training is carried out in the operational environment
where the student is required to execute the procedures repeatedly until
some performance criteria is achieved (Telfer and Biggs, 1988). According
to ACT* theory, this approach to instruction is particularly effective in
facilitating the process of composition, since practice ensures that sequential
productions are present in the working memory simultaneously, and can thus
be combined into a single production.
As mentioned earlier however, flight training often provides few
opportunities for pilots to develop and practice their weather-related
decision skills. This is due both to the widespread assumption that such
skills are unable to be taught explicitly, combined with the difficulties
associated with the construction of appropriate training scenarios. One of
the most significant consequences of this deficiency is the requirement for
relatively inexperienced pilots to develop their procedural knowledge
unassisted, through a process of trial and error. As noted previously, it is
extremely difficult to learn from experience in this way. Moreover,
Rockwell and McCoy (1988) have noted that some pilots choose not to
extend their experience in weather-related decision-making and conduct

463
Aviation Instruction and Training
flights only when the meteorological conditions exceed certain self-imposed
minimas. Considering their procedural inexperience, pilots in either
situation, when confronted unexpectedly with a requirement for effective
weather-related decision-making, are likely to exhibit characteristics of
problem-solving consistent with ACT*’s interpretive stage of knowledge
acquisition.
One of the most effective techniques with which to develop and fine-tune
skills is through exposure to relatively unfamiliar situations and experiences.
It is only through this type of experience that the pilot can acquire a broad
set of rule-based strategies based upon a sound understanding of the
particular domain. Since relatively few cross-country pilots place
themselves deliberately in situations with which they are unfamiliar, they
tend to avoid developing problem-solving strategies outside their normal
sphere of operations. However, it is precisely the rare and unfamiliar
occurrence, such as when a VFR pilot is caught in deteriorating
meteorological conditions with limited daylight and fuel, that fatal accidents
occur.
In providing weather-related decision-making experience for their
students, it is necessary for training organizations to develop programs
which are both safe, and encourage the development of the appropriate
skills. These programs may involve either dual cross-country flights or
simulated flights for which the objectives are to recognize weather
conditions which are likely to deteriorate below VMC; ensure the safety of
the aircraft and its passengers; and achieve as closely as possible the stated
objectives of the flight. Pilots should be encouraged to note the sequence of
these objectives, and should receive feedback immediately concerning their
level of performance.
The provision of delayed or immediate feedback is an important issue to
be considered from the ACT* perspective since both forms impact on the
efficacy with which productions are compiled. It will be recalled that
proceduralization is dependent upon the appropriate condition and action
segments being resident in working memory simultaneously (Anderson,
1987). Since working memory is prone to rapid loss of information,
delaying feedback may result in the loss or substitution of information salient
to the task, and thereby create associations which are either ineffective or
inappropriate. Immediate feedback in contrast, should facilitate the
development of productions, since the pilot is made aware of the association
between condition and action segments already contained within working
memory.
The role of the instructor in providing feedback is to observe the process
through which weather-related decisions are made, and indicate immediately

464
 A Skills-Based Approach to Aeronautical Decision-Making
and clearly to the student when an error has or has not been made. Errors in
weather-related decision-making may include miscalculating distances or
endurance, or misinterpreting certain meteorological information relevant to
the decision. Irrespective of the type of error however, the instructor should
not provide the correct solution, but rather should note that an error has been
made, and encourage the pilot to re-examine the problem. According to
Anderson (1987), simply copying the correct answer from an external source
will not elicit the procedures necessary to formulate solutions correctly in
the future. Moreover, it is unlikely that the skills will be transferred and
applied successfully to novel situations.
The ability to transfer or generalize existing knowledge to relatively
unfamiliar tasks is the ultimate goal of skill acquisition, and reflects the
achievement of expertise (Woods, 1988). ACT* predicts that a positive
transfer of skill will occur if the productions used in two or more skills are
identical, and can be perceived as identical by the operator (Singley and
Anderson, 1989). Pirolli and Anderson (1985) examined the validity of this
prediction using recursive programming functions in the artificial intelligence
language LISP. One group of subjects was provided with instruction which
emphasized the functional aspects of recursion, whilst the other group
received training in the structure of recursion problems. Following their
training, both groups were required to devise four recursion problems, and it
was noted that those subjects who had received the functional perspective
were 33 per cent slower than those who had received the structural training.
Pirolli and Anderson (1985) attribute the relative success of the structural
group to the development of procedural knowledge at an earlier stage of
their training. Thus, rather than a similarity between the superficial features
of problems, it is the similarity between the procedures used that facilitates
the transfer of skills.
During the early stages of flight training, it is necessary to provide
students with examples of situations which highlight the procedures common
to weather-related decision-making tasks. In other words, knowledge of
what to do is more important in the early stages of skill acquisition than
why do you do it. Possibly the first step in weather-related decision-making
is to recognize that a decision is required either to continue the flight or
divert to an alternate. Subsequent considerations may include determining
the lowest safe altitude, the endurance, or the distance to alternate airports.
This sequence represents the foundation of a procedure which would be
common to all weather-related decision-making tasks despite differences in
the particular circumstances. It should, therefore, facilitate the successful

465
Aviation Instruction and Training
transfer of weather-related decision-making skills to situations which would
otherwise be solved using relatively ineffective interpretive procedures.
Since weather-related decision-making is a relatively complex skill
comprising numerous sub-tasks, it may be necessary to part-task the
training, particularly during the initial stages of skill acquisition. The
process of part-tasking involves dividing the skill into separate components
and teaching these in isolation. Once a specified level of performance has
been achieved, the components are then re-integrated, and the entire task is
executed (Lintem, 1986). In the aeronautical domain, part-task training has
been demonstrated as superior to whole-task training in the acquisition of
skills such as air-to-ground combat manoeuvres and simulated landings
aboard aircraft carriers (Wightman and Sistrunk, 1987). Wightman and
Sistrunk (1987) suggest that the relative superiority of part-task training in
teaching complex skills is attributable to the opportunity to practice more
intensely those segments of the task which present the greatest difficulty.
Although the success of part-tasking is consistent with that predicted by
the ACT* model of skill acquisition, an alternative explanation is proposed
to that provided by Wightman and Sistrunk (1987). According to ACT*, the
development of productions is dependent upon the components being
resident simultaneously within the capacity-limited working memory. Since
part-tasking reduces the number of components to be compiled at any one
time, it should also reduce the likelihood that important items will be
replaced before knowledge compilation and proceduralization can occur.
Consequently, the development of productions, and thus the development of
skilled performance, is accomplished with greater efficiency than would
otherwise be the case.
Weather-related decision-making can be partitioned into a number of sub­
tasks from the recognition of deteriorating weather conditions, to the
calculation of range and endurance whilst maintaining control of the aircraft.
These are skills which can be learnt in isolation initially, and then integrated
to provide the basis for the development of effective weather-related
decision skills.
Fundamental to this approach to aeronautical decision-making training is
the assumption that pilots will develop the skills necessary to make effective
and appropriate weather-related decisions only through practical experience.
Unfortunately, all the existing approaches to training aeronautical decision­
making have taken the opposite course and have concentrated on enhancing
the pilot's declarative knowledge of decision-making styles. From the
perspective outlined here, it is clear that there is unlikely to be much transfer
from these training programs to decision-making in flight situations. The

466
 A Skills-Based Approach to Aeronautical Decision-Making
empirical evidence reviewed earlier is indeed consistent with such a
conclusion.

Conclusion

Crashes involving VFR flight into IMC account for a significant proportion
of accidents involving general aviation aircraft. Nevertheless, researchers
have yet to establish the causes of such crashes, and propose strategies
which will improve the quality of weather-related decision-making amongst
pilots. This paper takes the view that such crashes are due partly to
limitations in the pilot's level of experience in recognizing and avoiding
situations in which the conditions are below that required for flight under
VFR. Consistent with this notion, weather-related decision-making is
examined from the perspective of the ACT* model of skill acquisition. This
model comprises three distinct behavioural stages through which an
individual must progress in the transition from novice to expert in a
particular domain.
Pilots whose weather-related decision-making experience is relatively
limited would be expected to operate at the interpretive stage of skill
acquisition in which performance is governed by general problem solving
strategies. As pilots gradually accumulate experience in making weather-
related decisions, they will begin to devise productions during the
knowledge compilation stage, and increase the efficiency with which
decisions are made. With further experience, pilots will progress to the
tuning stage of skill acquisition and would be expected to make weather-
related decisions which are both efficient and appropriate in the particular
circumstances.
Central to the development of skilled performance is the association
between declarative and procedural forms of knowledge and the
development of productions which embody this relationship. According to
Anderson (1983), this can occur only through practical experience in a
variety of situations relevant to the task. During their initial flight training,
pilots should thus, receive less emphasis on formal instruction, and more
emphasis should be directed towards the information necessary to perform
weather-related decision-making successfully. This can be achieved through
dual cross-country flights or simulated exercises in which the trainee is
taught to recognize situations which will preclude flight under VFR,
formulate strategies which will ensure the safety of the aircraft and its
passengers, and achieve as closely as possible, the objectives of the flight.
As with other aeronautical skills such as engine failure or fofced landing

467
Aviation Instruction and Training
procedures, it is only through continuous practice that pilots will develop
and maintain the skills necessary for effective weather-related decision­
making.
Whilst it would be extremely optimistic to suggest that this approach will
eliminate weather-related crashes completely, it nevertheless, may go some
way to reducing that proportion of general aviation accidents for which the
probable cause is continued VFR into IMC.

References

Anderson, J.R. (1982), 'Acquisition of cognitive skill', Psychological

Review, vol. 89, pp. 369-406.

Anderson, J.R. (1983), The Architecture o f Cognition, Harvard University

Press, Cambridge.

Anderson, J.R. (1987), 'Skill acquisition: compilation of weak-method

problem solutions’, Psychological Review, vol. 94, pp. 192-210.

Anderson, J.R., Greeno, J.G., Kline, P.J. and Neves, D.M. (1981),
'Acquisition of problem solving skill', in Anderson, J.R. (ed.), Cognitive
Skills and their Acquisition, Erlbaum, Hillsdale.

Anderson, J.R., Kline, P.J. and Beasley, C.M. (1980), 'Complex learning
processes', in Snow, R.E., Frederico, P.A. and Montague, W.E. (eds.),
Aptitude, Learning and Instruction, Vol. 2, Erlbaum, Hillsdale.

Barnett, B.J. (1989), 'Information processing components and knowledge

representations: An individual differences approach to modelling pilot
judgement', in Proceedings o f the Human Factors Society 33rd Annual
Meeting, pp. 878-882.

Beck, G.A. (1987), The Problem o f 'Visual Flight Rules’ Flight into
'Instrument Meteorological Conditions': A Protocol Analysis o f General
Aviation Pilots' Subjective Weather Forecasting Procedure,
Unpublished MSc Thesis, The Ohio State University.

Beech, L.R. and Mitchell, T.R. (1978), Ά contingency model for the
selection of decision strategies', Academy of Management Review, vol. 3,
pp. 439-449.

468
 A Skills-Based Approach to Aeronautical Decision-Making

Berlin, J., Gruber, E.V., Holmes, C.W., Jensen, P.K., Lau, J.R., Mills, J.W.
and O'Kane, J.M. (1982), Pilot Judgement Training and Evaluation -
(Volumes 1, 2, 3), DOT/FA-79NA-6040, Federal Aviation Administration,
Washington.

Booze, C.F. (1977), 'Epidemiological investigation of occupation, age, and

exposure in general aviation accidents’, Aviation, Space, and
Environmental Medicine, vol. 48, pp. 1081-1091.

Bryan, L.A., Stonecipher, J.W. and Aron, K. (1955), AOPA's 180o Rating,
AOPA Foundation and Aircraft Owners and Pilots Association,
Washington.

Cessna Aircraft Company (1981), 152 Aerobat: Information Manual,

Wichita.

Chi, M.T.H., Feltovich, P.J. and Glaser, R. (1981), 'Categorisation and

representation of physics problems by experts and novices', Cognitive
Science, 5, 121-152.

Chi, M.T.H., Glaser, R. and Rees, E. (1982), 'Expertise in problem solving’,

in Sternberg, R. (ed.), Advances in the Psychology o f Human Problem-
Solving, Lawrence Erlbaum, Hillsdale.

Civil Aviation Authority (1988), CAP 452: General Aviation Accident

Review, 1987, Cheltenham, England.

Civil Aviation Authority (1991), Aeronautical Information Publication,

Canberra, Australia.

Colley, A.M. and Beech, J.R. (1989), 'Acquiring and performing cognitive
skills', in Colley, A.N. and Beech, J.R. (eds.), Acquisition and
Performance of Cognitive Skills, Wiley, Chichester.

Copas, C. (1984), 'Accident myths', Bureau o f Air Safety Investigation

Journal, vol. 4, no. 6.

Dawes, R.M. (1971), Ά case study of graduate admissions: Application of

three principles of human decision-making', American Psychologist, vol.
26, pp. 180-188.

469
Aviation Instruction and Training

deGroot, A.D. (1966), 'Perception and memory versus thought', in

Kleinmuntz, B. (ed.), Problem Solving Research, Methods and Theory,
Wiley, New York.

Dreyfus, H.L. and Dreyfus, S.E. (1986), 'Why skills cannot be represented
by rules’, in Sharkeye, N.E. (ed.), Advances in Cognitive Science, I, Ellis
Harwood, Chichester.

Einhom, H.J. (1972), 'Expert measurement and mechanical combination',

Organizational Behaviour and Human Performance, vol. 7, pp. 86-106.

Einhom, H.J. and Hogarth, R.M. (1978), 'Confidence in judgement:

Persistence in the illusion of validity', Psychological Review, vol. 85, pp.
395-416.

Fitts, P.M. and Posner, M.I. (1967), Human Performance, Brooks Cole,
Belmont, CA.

Flathers, G.W., Giffen, W.C. and Rockwell, T.H. (1982), Ά study of

decision-making behaviour of aircraft pilots deviating from a planned
flight', Aviation, Space, and Environmental Medicine, vol. 53, pp. 958-
963.

Garb, H.N. (1989), 'Clinical judgement, clinical training, and professional

experience', Psychological Bulletin, vol. 105, pp. 387-396.

Giffen, W.C. and Rockwell, T.H. (1984), 'Computer-aided testing of pilot

response to critical in-flight events', Human Factors, vol. 26, pp. 573-
581.

Giffin, W.C. and Rockwell, T.H. (1987), Ά methodology for research on

VFR flight into IMC', in Jensen, R.S. (ed.), Proceedings o f the Fourth
International Symposium on Aviation Psychology, The Ohio State
University, Columbus.

Giffin, W.C., Rockwell, T.H. and Smith, P.E. (1985), Ά review of critical
in-flight events research methodology', in Jensen, R.S. (ed.), Proceedings
o f the Third International Symposium on Aviation Psychology, The Ohio
State University, Columbus.

470
 A Skills-Based Approach to Aeronautical Decision-Making
Glaser, R. and Chi, M.T.H. (1988), Overview', in Glaser, R., Chi, M.T.H.
and Farr, M.J. (eds.), The Nature o f Expertise, Lawrence Erlbaum,
New Jersey.

Goldberg, L.R. (1970), 'Man versus model of man: A rationale plus some
evidence for a method of improving clinical inference’, Psychological
Bulletin, vol. 73, pp. 422-432.

Gray, W.D. and Orasanu, J.M. (1987), Transfer of cognitive skills', in

Cormier, S.M. and Hagman, J.D. (eds.), Transfer of Learning:
Contemporary Research and Applications, Academic Press, San Diego.

Gustafson, D.H., Tianen, B. and Greist, J.H. (1986), Ά computer-based

system for identifying suicide attemptors', in Arkes, H.R. and Hammond,
K.R. (eds.), Judgement and Decision-Making, Cambridge University
Press, New York.

Hammond, K.R., Hamm, R.M., Grassia, J. and Pearson, T. (1987), 'Direct

comparison of the efficacy of intuitive and analytical cognition in expert
judgement', IEEE Transactions on Systems, Man, and Cybernetics,
SMC-17, pp. 753-770.

Hayes, J.R. and Simon, H.A. (1974), 'Understanding written instructions', in

Gregg, L.W. (ed.), Knowledge and Cognition, Erlbaum, Hillsdale.

Hershey, D.A., Walsh, D.A., Read, S.J. and Chulef, A.S. (1990), 'The
effects of expertise on financial problem solving: Evidence for goal
directed, problem solving scripts', Organizational Behavior and Human
Decision Processes, vol. 46, pp. 77-101.

Jensen, R.S. (1988), 'Creating a '1000 hr' pilot in 300 hours through
judgement training', in Proceedings of the Workshop on Aviation
Psychology, University of Newcastle, Newcastle, Australia.

Jensen, R.S. and Benel, R.A. (1977), Judgement Evaluation and Instruction
in Civil Pilot Training. Final Report FAA-RD-78-24, National Technical
Information Service, Springfield.

Kahneman, D. and Tversky, A. (1973), 'On the psychology of prediction',

Psychological Review, vol. 80, pp. 237-251.

471
Aviation Instruction and Training
Klein, G.A. (1989), 'Recognition-primed decisions (RPD)', in Rouse, W.B.
(ed.), Advances in Man-Machine Systems, vol. 5, pp. 47-92, JAI Press,
Greenwich.

Klein, G. and Klinger, D. (1991), 'Naturalistic decision-making', CSERIAC

Gateway, vol. 2, pp. 1-4.

Kleinmuntz, B. (1990), 'Why we still use our heads instead of formulas:

Toward an integrative approach', Psychological Bulletin, vol. 107, pp.
296-310.

Larkin, J.H., McDermott, J., Simon, D.P. and Simon, H.A. (1980), 'Expert
and novice performance in solving physics problems', Science, vol. 208,
pp. 1335-1342.

Layton, C.F. and McCoy, E. (1989), 'General aviation pilot perceptions of

deteriorating weather conditions', in Jensen, R.S. (ed.), Proceedings o f
the 5th Symposium on Aviation Psychology, The Ohio State University.
Columbus.

Lesgold, A. (1988), 'Problem solving', in Sternberg, R.S. and Smith, E.E.

(eds.), The Psychology o f Human Thought, Cambridge University Press,
New York.

Lester, L.F., Diehl, A. and Buch, G. (1985), 'Private pilot judgement training
in flight school settings: A demonstration project', in Jensen, R.S. (ed.),
Proceedings o f the Third Symposium on Aviation Psychology, The Ohio
State University, Columbus.

Lintem, G. (1986), Behavioural Principles fo r Flight Instruction, Essex

Corporation, Orlando.

Massaro, D.W. (1989), Experimental Psychology: An Information

Processing Approach, Harcourt Brace Jovanovich, New York.

Meehl, P.E. (1954), Clinical versus Statistical Prediction: A Theoretical

Analysis and a Review o f Evidence, University of Minnesota Press,
Minneapolis.

472
 A Skills-Based Approach to Aeronautical Decision-Making
Meehl, P.E. and Rosen, A. (1955), 'Antecedent probability and the
efficiency of psychometric signs, patterns, or cutting scores',
Psychological Bulletin, vol. 52, pp. 194-216.

Miller, G.A. (1956), 'The magical number seven, plus or minus two: Some
limits on our capacity for information processing', Psychological Review,
vol. 63, pp. 81-97.

Moll, N. (1982), More I Learned about Flying from That, Ziff-Davis,

London.

National Transport Safety Board (1987), Annual Review o f Accident Data:

U.S. General Aviation Operations, Calender Year 1983, Report
NTSB/ARC-87-01, National Technical Information Service, Springfield.

O'Hare, D. (1987), 'Pilots’ perception of risks and hazards in general

aviation', Aviation, Space, and Environmental Medicine, vol. 61, pp.
599-603.

O'Hare (in press), The "ARTFUL" decision maker: A framework model for
aeronautical decision making', The International Journal o f Aviation
Psychology.

O'Hare, D. and Roscoe, S.N. (1990), Flightdeck Performance: The Human

Factor, Iowa State University Press, Ames.

O'Hare, D., Wiggins, M., Batt, R. and Morrison, D. (in press), 'Cognitive
failure analysis for aircraft accident investigation', Ergonomics.

Perfetto, G.A., Bransford, J.D. and Franks, J.J. (1983), 'Constraints on

access in a problem-solving context', Memory and Cognition, vol. 11, pp.
24-31.

Perrett, M. (1991), 'Scudrunniri or plane-rain-terrain-pain', Aviation Safety

Digest, vol. 148, p. 11.

Pirolli, P.L. and Anderson, J.R. (1985), 'The role of learning from examples
in the acquisition of recursive programming skill', Canadian Journal of
Psychology, vol. 39, pp. 240-272.

473
Aviation Instruction and Training
Potter, S.S., Rockwell, T.H. and McCoy, C.E. (1989), ’General aviation
pilot error in computer simulated adverse weather scenarios', in Jensen,
R.S. (ed.), Proceedings o f the Fifth Symposium on Aviation Psychology,
Ohio State University, Columbus.

Rasmussen, J. (1983), Skills, rules, and knowledge; Signals, signs and

symbols, and other distinctions in human performance models', IEEE
Transactions o f Systems, Man and Cybernetics, SMC-13, pp. 257-266.

Rasmussen, J. (1988), 'Human error mechanisms in complex work

environments', Reliability Engineering and Systems Safety, vol. 22, pp.
155-167.

Rockwell, T.H. and McCoy, C.E. (1988), General Aviation Error: A Study
o f Pilot Strategies in Computer Simulated Adverse Weather Scenarios,
Contract Report DTRS-57-85-C-00101, United States Department of
Transportation, Cambridge.

Simon, H. and Hayes, J.R. (1976), 'The understanding process: Problem

isomorphs', Cognitive Psychology, vol. 1, pp. 19-40.

Singley, M.K. and Anderson, J.R. (1989), The Transfer o f Cognitive Skill,
Harvard University Press, Cambridge.

Telfer, R. and Ashman, A. (1986), Pilot Judgement Training - An

Australian Validation Study, University of Newcastle, Newcastle,
Australia.

Telfer, R. and Biggs, J. (1988), The Psychology o f Flight Training, Iowa

State University Press, Ames.

Terranova, M., Snyder, C.E., Seamster, T.L. and Treitler, I.E. (1989),
'Cognitive task analysis: Techniques applied to airborne weapons
training', Proceedings o f the Human Factors Society 33rd Annual
Meeting, pp. 1358-1362.

Tversky, A. and Kahneman, D. (1986), 'Judgement under uncertainty:

Heuristics and biases', in Arkes, H.R. and Hammond, K.R. (eds.),
Judgement and Decision Making: An Interdisciplinary Reader,
Cambridge University Press, New York.

474
 A Skills-Based Approach to Aeronautical Decision-Making
Voss, J.F., Tyler, S,W. and Yengo, L.A. (1983), 'Individual differences in
the solving of social science problems', in Dillon, R.F. and Schmeck, R.R.
(eds.), Individual Differences in Cognition (Volume 1), Academic Press,
New York.

Waller, W.S. and Mitchell, T.R. (1984), 'The effects of context on the
selection of decision-strategies for the cost variance investigation
problem', Organizational Behavior and Human Performance, vol. 33, pp.
397-413.

Wickens, C.D. and Flach, J.M. (1988), 'Information processing’, in Wiener,

E.L. and Nagel, D.L. (eds.), Human Factors in Aviation, Academic Press,
San Diego.

Wightman, D.C. and Sistrunk, F. (1987), 'Part-task training strategies in

simulated carrier landing final-approach training', Human Factors, vol.
29, pp. 245-254.

Woods, D.D. (1988), 'Coping with complexity: The psychology of human

behaviour in complex systems', in Goodstein, L.P. Andersen, H.B. and
Olsen, S.E. (eds.), Tasks, Errors and Mental Models, Taylor and Francis,
London.

475
Author Index

Adams, J.A., 11, 14, 226, 233 Bowden, J.A., 121, 128
Alessi, S.M., 81, 86 Bowers, C., 429
Alexander, K. 2, 13 Bowsher, J., 288
Alexander, W., 15, 276, 289 Brandt, R., 215, 217, 233, 289
Allerton, D.J., 3, 93, 163 Branyan-Broadbent, B., 233
Anastasi, A., 32 Brewer, W.F., 417, 426
Anderson, J.R., 72, 86, 227, 233, Brookfield, S.D., 288
452-458, 467, 468 Brown, S.W., 236
Arends, R.I., 216, 217, 233 Browning, R.F., 157
Argyris, C., 333 Bruggink, G.M., 111
Ashman, A., 12, 14, 440, 474 Bryan, G.L., 33, 431, 469
Buch, G., 440, 472
Bangert-Drowns, R.L., 233 Buchmann, M., 216, 217, 233
Bardre, A.N., 5, 6 Burkman, E., 212, 217
Bartram, D., 2, 13, 34, 40, 44,
48-50 Calderwood, R., 417, 426
Beech, J.R., 436, 468, 469 Campbell, W.J., 427
Benel, R.A., 430, 435, 441 Cannon-Bowers, J.A., 417, 427,
Bent, J., 3, 92, 114, 215,218 429
Berlin, J„ 440-443, 469 Cantwell, R.H., 125, 128
Besco, R.O.,341 Caretta, T.R.,40, 50
Biddle, B J., 220, 233 Camine, D., 234, 340, 341
Biegalski, C.S., 409 Caro, P., 92, 94
Biggs, J.B., 5, 7, 76, 86, 121-128, Charlwood, D., 53, 54, 56, 66
218, 221, 233, 236, 282, 289, Chidester, T.R., 384, 409, 429
342, 474 Clark, N., 59, 67
Blinkhom, S., 44, 50 Colley, A.M., 456, 469
Bloom, B.S., 278, 288 Converse, S.A., 417, 428
Blyth, B., 6, 206, 237 Cooper, C.L., 368, 384
Boekarts, M., 14 Copas, C., 430, 469
Bond, N.A., 33 Costello, M., 260, 266, 269
Borko,H ., 5, 6 Crook, C., 70, 86
Aviation Instruction and Training

Cruickshank, D.R., 220, 233 Glickman, A.S., 427

Curtis, R.V., 235 Globerson, T., 93, 94
Goldstein, D.A., 157, 368, 370,
Dale, H.C.A., 40, 48-50 384
Davis, P., 268, 269 Gradwell, D., 208
Degani, A., 427 Grandjean, A., 14
De Hart, R.L., 307 Green, R.C., 207
Diehl, A., 440, 472 Green, R.L., 208
Dolgin, D.L., 39, 50 Gregorich, S., 384, 409
Dooley, R.P., 151, 157 Gronlund, N., 278, 288
Droog, A., 6, 207, 323, 333 Guerette, P.J., 427
Dunkin, M.J., 220, 233 Gum, D.R., 141, 158
Dupree, D.A., 417, 426
Hackman, J.R.,416, 427
Eissfeldt, H., 43, 50 Haertal, G.D., 220, 236
Elshaw, C., 6, 206, 253, 256, 270 Hamilton, R.M., 67
Engelmann, S., 234, 340, 341 Hanes, L.F., 132-134, 161
Entwistle, N.J., 121, 128, 129 Hannan, G.J., 3, 13, 63, 67
Erensting, J., 307 Harding, R.M., 307
Eysenck, M.W., 267, 269 Harms, D., 411, 427
Harsveld, M., 50
Farr, M.J., 231, 232, 234 Hass, G., 289
Fester, B.C., 266, 269 Hattie, J., 121-123, 125, 128, 129
Fisk, A.D., 427 Hawkins, F.H., 97, 328, 409
Fitts, P., 151, 227, 234, 257, 270, Haynes, A.C., 111
470 Helmreich, R.L., 99, 111, 368,
Flach, J., 65, 67, 475 369, 383, 395, 409
Flathers, G.W., 448, 449, 470 Hilton, T.F., 39, 50
Flexman, R.E., 135, 157 Hobbs, A., 58, 63, 67
Foushee, H.C., 99, 112, 368, 383, Hoc, J.M., 428
409 Holding, D.H., 234
Hughes, C., 289
Gagne, R.M., 72, 86, 217, 234 Hunt, G.J.F., 3, 13, 70-72, 75, 86,
Garrison, P., 307 Hunt, L.M., 3, 13
Geiselman, R.E., 417, 426 Hunter, D., 39, 50
Gibb, G„ 64, 67 Huntington, H., 409
Giffen, W.C., 445, 449, 450, 470
Given, K., 4, 93, 188 Imhoff, D.L., 40, 50
Glaister, D.H., 299, 307
Glaser, R„ 72, 86, 446, 447, 469, Jacobs, R.S., 135, 158
471 James, M., 208

478
 Index

Jarvis, P., 92, 94 Levine, J.M., 40, 50, 148, 150,

Jensen, R.S., 93, 94, 111, 409, 158
430, 435, 443, 444, 471 Levison, W.H., 141, 158
Jenson, W.R., 234 Lewis, A., 276, 289
Johannsen, G., 428 Linn, R., 278, 288
Johnson, C., 44, 50 Lintem, G., 3, 5, 7, 92-94, 130,
Johnston, A.N., 5, 108, 110-112, 136, 137, 151, 158, 159, 161,
338, 371, 384, 411, 428 220, 229, 234, 236, 466, 472
Jones, B., 57, 289 Livingston, C., 5, 6
Jordan, J.L., 307
MacGregor, D., 417, 426
Kember, D., 234 Maher, J.W., 410
Kemp, J., 274, 289 Martin, E.L., 135, 159
Kent Wood, R., 233 Marzano, R., 289
King, A., 92, 94 Maurino, D.E., 3, 92, 95, 112
King, P„ 307 McCoy, C.E., 433, 435, 438, 439,
King, T., 417, 429 474
Kirby, J.R., 121, 128 McDonald, F.J., 4, 7
Klein, G.A., 416, 428, 429, 437, McMillan, G„ 3, 92, 130
438, 447, 457, 472 McNeil, J., 289
Knowles, M., 280, 289 Mellor, A., 4, 338, 368
Koonce, J.M., 137, 151, 158 Melton, A.E., 235
Krahenbuhl, G.S., 268, 270 Merrill, M.D., 72, 87
Krathwohl, D.R., 278, 289 Meyer, R.C., 52, 53, 67
Kravetz, D.J., 91, 94 Michael, A.L., 132, 133, 134, 161
Kulik, C.C., 233 Miller, G.A., 257, 270, 473
Kurstler, J.H., 4, 6 Miller, W.R., 235
Millis, B.J., 235
Lake, R., 58, 63, 67 Mills, F.J., 307
Lambeth, L., 6, 206, 291 Mintzberg, H., 333
Lane, E.L., 227, 230, 231, 234 Mold, E., 254, 270
Lauber, J.K., 99, 428 Moll, N., 473
Laurillard, D., 234 Montero, R.C., 427
Lave, J., 424, 428 Moore, P.J., 3, 92, 121-123, 125,
Leary, K., 58, 67 126, 128
Lee, P., 58, 67 Morgan, M., 233
Lehrer, H.H., 6, 206, 271 Muckier, F.A., 142-147, 151, 153,
Leinhardt, G., 216, 217 154, 159, 160
Lesgold, A., 234, 341, 454, 456, Muir, H., 208
472 Murphy, M., 417, 429
Lester, L.F., 440, 441, 472

479
Aviation Instruction and Training

Nagel, D.C., 383 Ronden, J.M., 254, 255, 270

Newell, A.A., 72, 87 Rogers, W.A., 427
Norwood, G., 307 Roscoe, S.N., 5, 7, 92, 94, 135,
136, 151, 159, 160, 207, 253,
O’Hara, R., 3, 13, 67 270, 433, 434, 444, 473
O’Hare, D., 5, 207, 430, 433-435, Rose, H.C., 235
437, 438, 444, 457, 462, 473 Rouse, W.B., 428
O’Reilly, S.J.A., 256, 270 Rudnitsky, A., 289
Orasanu, J., 417, 428, 429, 442, Ruffel-Smith, H.P., 112, 368, 384
454, 471 Ryan, R., 235
Orlady, H., 112
Ortega, K.A., 416, 429 Saettler, P„ 213, 217, 235
Salas, E., 417, 428, 429
Paechter, M., 43, 50 Samet, M.G., 417, 426
Pendarvis, C.W., 410 Saylor, J., 276, 289
Pepitone, D., 417, 429 Schmidt, R.A., 11, 12, 14, 226-
Perkins, D.N., 93, 94 228, 235
Perrett, M., 433, 434, 473 Schneider, W., 259, 270
Phillips, D.C., 209, 217, 235, Schoemaker, P.J.H., 412, 429
340, 341 Schofield, N.J., 121, 129
Pippin, W.E., 6, 207, 309 Schon, D.A., 211, 217, 235
Posner, G., 289 Schwartz, D., 410
Posner, M.I., 234, 470 Simon, H.A., 72, 87, 442, 446,
Povenmire, H.K., 135, 160 460, 472, 474
Presseisen, B., 289 Simpson, E.J., 278, 289
Prince, C., 429 Sinclair, B., 5, 337, 356
Sind,P.M., 207
Ramsden, P., 121, 128, 129 Skinner, B.F., 266, 269
Rankin, S., 289 Sloane, H.N., 234, 368
Rasmussen, J., 227, 235, 459-462, Smith, P., 40, 49, 470
474 Solomon, G., 93, 94
Ratner, R., 59, 67 Somerville, R., 5, 337, 343
Reading, R.E., 412, 429 Stead, G., 2, 13, 15, 33, 43, 50
Reason, J., 112 Suhor, C., 289
Reiguluth, C.M., 235
Reynolds, A., 211, 217, 220, 235 Taba, H., 273, 289
Rigney, J.W., 33 Tallman, K., 60, 65, 67
Ritchie, M.L., 132, 134, 161 Taylor, H.L., 161
Robbins, S.P., 370, 384 Telfer, R.A., 3, 5-7, 12, 14, 91,
Rockwell, T.H., 433, 435, 437, 92, 121-123, 126, 129, 205,
445, 463, 470, 474 209, 215, 218, 219, 221, 236,

480
 Index
254, 270, 282, 289, 337, 340-
342, 410, 440, 474
Thordsen, M.L., 416, 428, 429
Travis, G.W., 289
Trollip, S.R., 81, 86, 93, 94
Tyler, R., 272, 290

Underwood Ground, K.E., 308

Vandermark, M.J., 112

Vicente, K.J., 416, 429
Voss, J.F., 447, 475

Waag, W.L., 135, 138, 141, 161

Walberg, H.J., 5, 7, 220, 236
Walker, J.Y., 93, 94
Wang, M.C., 220, 236
Warren, N.D., 33
Watkins, D., 121-123, 129
Waxman, H.C., 5, 7, 220, 236
Wenger, E., 424, 428
Westra, D.P., 140, 161
Wiener, E.L., 113, 262, 270, 383,
427
Wiggins, M., 5, 430, 462, 473
Wightman, D.C., 229, 236, 466,
475
Wilhelm, J., 383, 409
Wilkinson, R.L., 4, 338, 393
Williams, C.J., 236
Wilson, T., 4, 338, 385
Wittrock, M.C., 236
Woods, D.D., 465, 475

Yonge, G.D., 92, 94

Young, K.R., 234

Zimmer, S., 427

481
Subject Index

ab initio, 3, 19, 41, 136, 188, airline, 2, 13, 18, 115, 126
203, 207, 238, 411 supplementary, 385-386
ability, 15-51, 71 airmanship, 95, 109, 207
general ability, 39 Airways Systems Data Officers,
academic, 5, 237, 252 64
acceleration, 101, 199, 299 alcohol, 102, 302, 303
accident, 115 American Educational Research
achieving approach, 121-123 Association, 220
ACT* model, 430-467 androgogy, 92, 122, 216
ADF, 185 anxiety, 261, 262, 267-269
adult learning approaches to learning, 92-127
see androgogy aptitude, 11, 35, 327
advance organizers, 222 testing, 2, 11-13, 15-33, 34-51,
Aer Lingus, 5, 110, 411-429 57, 58, 63
Aeradio, 52-53 Army Air Corps, 40, 44, 45
Aerodrome Controller, 53-54 arousal, 267-269
aeronautics, 311 assessment, 15-51 , 124, 188-202,
Aerosoft-200 flight trainer, 179 251, 252
affective, 277-278 see evaluation, testing
aims, 182-184 group, 29
aim point, 182 multiple, 29, 30
Air New Zealand, 5, 92 simulator, 11, 177, 178
Air Traffic Control (ATC), 43, associative, 258
52-68, 118, 275, 276 ATPL, 99
Air Traffic Controller, 13, 52-68, attitude, 117, 369, 400, 401
102 attribution, 263-265
aircrew, 126 attrition rate, 15
Aircrew Team Management, audio-visual aids, 359-361
116 Australian Airlines, 338
autonomy, 124, 227, 228, 258,
 Index
259 check list, 338, 411-429
Aviation Curriculum Design non-normal, 338, 413-416
Matrix (ACDM), 206, 274-288 non-normal generic, 338, 416-
aviation instruction, 5, 92, 93, 95- 429
113 chunking, 231, 257, 258
traditional, 311-314 see coding
aviation degree, 310, 311 Civil Aviation, 52, 237, 293, 294
aviation medicine, 123, 206, 291- Authority, 13, 58, 64
307 clarity, 224
Aviation Technical Education closed-loop, 190, 226
Council, 275 Cockpit (or Crew) Resource
avionics, 286, 287 Management (CRM), 102-104,
AWA, 52 116,119,127,332,351, 368-
awareness, 225,372 410
coding, 257, 258
B747-400, 325, 349 see chunking
Bachelor of Science (Aviation), cognitive, 74, 219, 227, 228, 257,
206 277
backward chaining, 229 command, 124
base rate, 37 Commercial Pilot Licence, 100,
basic flying training, 166-187, 238, 245
256, 257 communication, 52, 53, 91, 102,
behavioral, 97, 219, 222 266, 267, 330
profile, 27 Communications Section, 52, 54
Bernoulli, 345 Communications Officer, 54, 55
block, 200, 251 computer-based testing, 69-86
briefing, 91 co-ordination
broad-based course, 66 hand-eye, 39-42
B ureau of A ir Safety hand-foot-eye, 39-42
Investigation, 300, 302, 307, 386 see aptitude, selection
com petency based training
caffeine, 305, 306 (CBT), 14, 93, 167, 187, 195,
carrel, 360 213
case study, 279, 280 competency, 14, 69-87, 93
Cathay Pacific Airways, 3, 5, 40, objective, 317
91, 92 subjective, 317, 318
Central Flying School, 254 specification, 70, 77
Centre for Air Traffic Services, Comprehensive Occupational
59
certification, 99, 287 Data Analysis Programs
change, 114, 238, 252, 254, 337 (CODAP), 194, 201

483
Aviation Instruction and Training

computer, 280, 366 disorientation, 101

based aptitude tests, 13, 26, drag, 346-354
34-51 drill, 424
based assessment, 31, 32, 34, aircon unit overheat, 415
366 emergency briefing: cabin,
Aviation Licensing and 418, 423
Examination System engine failure and shutdown,
(CALES), 13, 14, 69-87 418-419
technology, 14, 26, 79, 164, engine fire, 414
213 ground proximity warning,
based training (CBT), 164, 418
249, 250 hydraulic system failure, 421
concepts, 74, 225 systems malfunction, 418, 420
conditioning, 265, 266 drugs, 102
confidence, 329, 330
context, 77, 107, 219, 248 education, 1, 97, 98, 122, 209,
social, 314 250, 318, 368
political, 314 basic, 320
conversion course, 356-367 discipline-related, 320
correlation coefficients, 11-13, future, 314, 315
36, 37 general, 207, 311
cost benefit, 17, 46, 118, 166, professional, 97, 309-322
211 educational experience, 250, 271
Cranfield Institute of Technology, effectiveness, 122, 221, 281, 282,
3 352
Crew Dynamics, 4, 338, 385 efficiency, 122, 201
Crew Management Training, 100, Embry Riddle Aeronautical
329, 330, 385-392 University, 206
curriculum, 100, 240, 318-322 encoding, 257-258
design, 271-288 see chunking
see syllabus energy manager, 337, 346-355
engineer, flight, 98, 348
debrief, 91 enthusiasm, 224
DECIDE model, 443 evaluation, 91, 188-202, 251,
decision-making, 100, 256, 318, 252, 282
328, 430-475 external, 194, 408
operational, 103 formative, 192
decompression, 298, 299 internal, 194
designated medical examiner, summative, 192
293-294 see testing, assessment
Diploma of Applied Science (Air environment, 16, 97, 102, 122,
Traffic Services), 13, 59-63, 65

484
 Index

164, 214, 248, 276, 291 fly-by-wire, 13, 39, 47

simplified, 164, 230 folkways, 216
examination, 69-87, 251, 252 Futura Program (Lufthansa), 323
see assessment, evaluation, test
experience, 5, 123, 432-435, 444, G forces, 267, 292, 299
445 negative, 299
extrinsic motivation, 121-123 positive, 299
extrovert, 263, 264 ‘red out’, 299
‘get-home-itis ’,435
failure, 256, 257, 340 glass cockpit, 120, 350
see attrition rate, remediation global system, 316, 317
fatigue, 300 globalized aviation, 309-322
feedback, 169, 194, 433 GRADE, 399
fidelity, 93, 130, 169 graphics, 174, 175
field of view, 135 groundwork, 240, 327
fitness, 331 group instruction, 100
fixative, 227, 228, 258 guided discussion, 221, 279
flexibility, 122
flight, 123, 181-183, 261 handling techniques, 171
attendants, 371, 391 hazardous thought patterns, 442
manuals, 357 hours to solo, 123, 125
model, 132-135 HSI, 173, 174, 184, 185
Flight Checking Officer, 53 Hull University, 2, 13, 34, 40, 44,
Flight Control Officer, 53 48-50
flight crew competencies, 70-72 human error, 17, 96, 97
flight crew licensing, 69-83 human factors, 17, 91, 95, 351
Flight Deck Team Management, Digest, 98, 101, 104
116 training, 91, 95-120, 312
flight engineer, 98, 348 hyperventilation, 101, 296-298
flight instruction, 122, 171, 323 hypoxia, 101, 295, 296
see flight training
Flight Management Computer ICAO Human Factors Program,
(FMC), 348, 351 3, 92, 95, 112
Flight Service Officer, 52, 54, 55, IFR, 134, 431
64 ILS, 183, 184
flight simulation, 92, 93, 171-187, in-house training, 66
358 in the loop, 165
flight training, 132-135, 138, 244 indirectness, 223
management, 329, 330, 351 individualized instruction, 31
see flight instruction individualization, 3, 17, 97
flexibility, 224 information processing, 102, 348

485
Aviation Instruction and Training

see coding, chunking domain, 226, 277

innovation, 114, 238, 252, 254, environment, 122, 248
337 objective, 198
instruction, 1, 122, 179, 220 pilots’, 92-127
art, 91, 122, 215 prior, 216
craft, 91, 122, 215 rate, 260, 432-435
design, 271-288 lecture, 248, 279
effectiveness, 259-261 brief, 247
efficiency, 122, 201 lesson, 245-248
philosophy, 92, 291, 412, 413 plan, 245
simulator, 177, 178 licensing, 2, 69-87, 92
workshop, 91, 92 lift, 346-354
instructor training, 100, 206, 208, Lilienthal, Otto, 343, 348
259-261 Line Operational Simulations
instructors, 5, 253-269 (LOS), 339, 373, 412
negative, 268, 269 Line Oriented Flight Training
positive, 268, 269 (LOFT), 103-105, 351, 353,
instrumentation, 134, 173-175 394,406-408, 412
International Civil Aviation Link trainer, 133-135
Organization (ICAO), 1, 3, 92, Lockheed Super Constellation,
99, 117 345
interpersonal characteristics, 101 locus of control, 124
intrinsic motivation, 92-127
Manoeuvre Oriented Flying
judgment, 100, 124, 307, 440 Training (MOFT), 339, 426
Massey University, 69-87, 110
Kendell Airlines, 385, 386 media, 280, 281
Kendell Crew M anagement medical certificate, 291
Course, 385-392 medication, 303, 304
KHUFAC, 97, 328, 388 medicine, 97, 123, 206, 291, 308
KLM, 92, 97, 208, 323-333 memory, 73, 75, 452-462
knowledge, 239 working, 257, 453
declarative, 69-87, 452 short-term, 257
procedural, 69-87 mental model, 259, 260, 332
knowledge-process requirements, MICROPAT, 13, 40
72 microteaching, 225
military, 261
Lanchester, 345 flight surgeon, 206, 294
leadership, 103, 328, 401 model, 5, 171-173, 190, 194, 223
learning, 257-269, 276-288 more-is-better approach, 122
approach to, 92, 121-129 motivation, 116, 121-129, 261 -
difficulties, 432-435

486
 Index

265 demonstration, 83, 280

achievement, 121, 123, 264 personality, 14, 18, 39, 196
extrinsic, 92, 121, 123, 262, personnel licensing, 100, 238,
263 245
intrinsic, 92, 121, 122, 127, phases of instruction, 205-221
262, 263 philosophy, 92, 291, 412, 413
social, 121, 122 physiology, 98, 330
motor-coordination, 39 pilot
multi crew, 116, 425 abilities, 15-51
multi-task, 261 error, 431-440
multiple choice test, 21-25 learning, 92, 127
see test licence, 2, 69-87, 92, 100,
Myers-Briggs Type Indicator, 391 238, 245
medical certificate, 291
National Transportation and performance, 108, 303, 304
Safety Board (NTSB), 276, 430 progress, 126
navigation, 242-244, 348 selection, 15-51
NDB, 185 stress, 104, 256, 267-269, 330
needs, 275 student, 123, 124
analysis, 273, 274 training, 91-93, 163-187, 253
N eurolinguistic Programming workload, 300, 301
(NLP), 199 PILOT, 389
New Zealand Civil Aviation PLATO, 213
Authority, 13, 70 PLPQ (Pilot Learning Process
noise, 300 Questionnaire), 92, 127
non-technical training, 328-331 practice, 6, 193
blocked, 228
objective, 91 distributed, 228
behavioural, 97, 219, 222 part, 229, 230
performance, 278, 279 whole, 229, 230
on the job training, 66 praise, 225
operational decisions, 96 presage, 205, 221
overload, 300, 301 ‘pressing on’,435
oxygen, 102, 292 pre-testing, 223
problem solving, 124, 294, 462
PA, 390 procedure, 74, 163, 454
part-task training, 165-187, 212 trainer, 13
patter, 3, 254 training, 130-157, 163-187
pedagogy, 122, 215 cognitive, 454-457
perceptual trace, 266 flight, 130-157
performance, 80, 256, 267, 278 process, 205, 221, 450

487
Aviation Instruction and Training

of learning, 257-269, 276-278, see practice

432-435 reinforcement, 265
process-content interaction, 73 see motivation
processing, 348, 453 reliability, 70-73, 189
product, 205, 221 remediation, 257-269, 276-288
Professional Aviation see learning difficulties
Maintenance Association, 275 remembering, 73, 75, 231
professional, 207, 307 resources, 91, 280-281
education, 109 retention, 73, 75, 231, 256
practice, 189, 321 risk, 438
socialization, 5, 320 RMI, 184-186
training, 1, 91, 109, 189, 309- Royal Air Force (RAF), 40
322 Royal Australian Air Force
professionalization, 5, 109, 212 (RAAF), 4
flight instructor, 122, 171, 323 Royal Navy (RN), 40, 44, 45
pilot, 91-93, 108, 303, 304
profile, behavioural, 27 SAAB 340, 48, 391
progress, 256 safety, 96, 331
prototype course, 330-351 levels, 96
psychology, 98, 330 nets, 81, 82
of learning, 91 schema theory, 227, 244
of instruction, 91 School for Air Traffic Controller
psychological screening, 15-33 Training, 52-68
psychomotor, 18, 277, 278 seating plan, 380
co-ordination, 18 selection, 2, 15-33, 34-51, 56-68
purpose, 272-288 process, 11-13, 34-51, 56-68
see aim, objective ratio, 13, 37
self, 124
q-feel, 170 awareness, 221, 329
QANTAS, 4, 43, 393-410 concept, 225
qualities, 35, 36 teaching, 221, 249
self medication, 304, 305
recall, 227 sensory input, 101
recognition, 227, 444 SHL Asia-Pacific, 2, 13, 15, 33,
of prior learning (RPL), 195 43,50
of acquired learning, 195 simulator, 92, 93, 130-162, 225
of current competencies, 195 flight, 93, 131-156, 361
recruitment, 55-56 assessment, 28, 31, 41
see selection transfer, 60, 92
refresher training, 374-378 situational awareness, 102, 104
rehearsal, 6, 193 skill, 104, 207, 239

488
 Index

acquisition, 11, 257-259, 451 techniques, instruction, 91, 122,

communication, 104 215, 259-261, 271-288
complex, 228-230 technological change, 114, 238,
continuous, 229 252
learning, 91, 207 tertiary course, aviation, 3, 13,
sequential, 229 59-63, 206, 310-311
sleep, 101 test, 21-27, 281-284
smoking, 303 adaptive, 41
solo, 122 battery, 18
Spiritus audio-visual system, 359, construction, 11, 34-51, 69-87,
360 282-284
stable personality, 263, 264 data, 79
standard operating procedures, 74, design, 42, 63, 64
99, 115, 130-157, 163-187, multiple choice, 69-87, 282
292, 368 true-false, 79, 282
standardization, 3, 99, 115, 201, testing, 15-33, 34-51, 63, 69-87,
292, 368 188-202, 251, 252
standardized test, 34-51, 69-87 see assessment, evaluation, test
stress, 256, 267-269 The Ohio State University, 309-
management, 104, 330 322
structuring, 223 theory, 2, 6, 193, 210
student pilot, 3, 19, 41, 136, 188, behavioural, 219
203, 207, 238, 411 cognitive, 219
Study Process Questionnaire, 126 thrust, 346-354
supplementary airline, 385-386 Tiger Moth, 344
surface approach, 121-123 total quality management (TQM),
SWISSAIR, 92 196, 201
syllabus, 61, 100, 178, 240-242, toxicology, 303, 304
271-288, 325 trainee-centred instruction, 220,
see curriculum 337
synergy, 368-410 training, 1, 34, 55, 59-63, 96,
Systems Approach to Training 114-120, 122, 318, 368, 404
(SAT), 93, 188-202 analysis, 93, 165, 166
synthetic training, 130-157, 361 basic flying, 132-135, 138
cycle, 93, 190, 191, 199
task, 77 design, 92, 191
analysis, 219 evaluation, 188-202, 352
oriented, 223 model, 114-120, 352, 411-429
TASKOMAT, 40 requirements, 39, 166
teaching, 91, 122, 279 synthetic, 92, 130-157, 163-
teamwork, 328, 369, 403 187

489
Aviation Instruction and Training

system, 190
technology, 130-157, 163-187
transfer, 92, 131-156
twin engine, 184
Trans Australia Airlines, 4
transfer of training, 92, 131-156
Trinity College, Dublin, 5
true-false test, 79
twin engine training, 184

ultra long haul, 120

University Aviation Association,
275
University of Illinois, 3, 5
University of Newcastle, 3, 110
University of Tasmania, 59
unstable personality, 263, 264

validity, 11-13, 36, 37, 42-49, 70-

73, 189
variability, 223, 229
VARIG, 40
verbalization, 22
VFR, 431-467
vibration, 300
vision, 394
visual, 133
system, 130, 132, 137, 163,
175-177, 394
teaching, 132, 179, 180, 223
Flight Rules, 431-467

weather, 434, 435, 450

weight, 346-354
weighting, 42, 43
workload, 101, 180, 256
workshop, 91, 92

490