BOOK OF ABSTRACTS
THE NINTH MEETING OF MASI
RESEARCH NETWORK ON METHODOLOGY FOR THE
ANALYSIS OF SOCIAL INTERACTION
UNIVERSITY OF WÜRZBURG, GERMANY
AUGUST 26-27th 2016
PROGRAMME
August 25th
Location: Zentrum für Mediendidaktik, room 005
15.00-17:00 Theme Workshop – M.S. Magnusson & G.K. Jonsson
19:00 Get together: Restaurant “Alte Mainmühle”, Mainkai 1, 97070 Würzburg
August 26th
Location: Zentrum für Mediendidaktik, room 005
Chair: Michael Brill
GMT +2
09.00-09:15 Welcome address – F. Schwab
09:15-09:30 An overview of recent research involving the t-pattern model and Theme and other
milestones from MASI members – G.K. Jonsson
09:30-10:30 Demonstration of Theme / From stardust and protein cities to swarm intelligence and
structured societies: t-patterns in a self-similar fractal world – M.S. Magnusson
10:30-10:45 Coffee / Tea
10:45-11:25 T-pattern methodology for human behaviour observation and analysis: Some research
applications – I. Terrenghi, B. Diana, M. Elia, V. Zurloni, A. Elia, M.S. & P.C. Rivoltella
11:25-12:05 Detection of T-Patterns as an important tool for studying changes in physiological
interventions from qualitative inputs – M.T. Anguera, M. Portell, S. Chacón-Moscoso & S.
Sanduvete-Chaves
12:05-12:45 Patterns of topic organisation in turn management. A case study of interactions in
dialogues – L. Hunyadi
12:45-14:00 Lunch
Chair: Mariona Portell
14:00-14:40 Complementarity between T-Patterns and polar coordinates analysis in social
interaction: How to integrate the results? – M.T. Anguera, G.K. Jonsson & P. Sánchez-
Algarra
14:40-15:20 LMX and Leadership Behaviour with T-Pattern Analysis – M. Tarim & M.S. Magnusson
15:20-16:00 Detection of Ludic Patterns in Two Motor Triadic Games – M.P. Aguilar, V.N. Adelantado
& G.K. Jonsson
16:00-16:20 KINEMO software: towards automated gesture annotation with MS Kinect – K. Juszczyk
& K. Ciecierski
16:20-16:50 Coffee / Tea
16:50-17:00 Expanding the Study of Internet Gambling Behaviour – G. Jonsson, H. Milkman, H.M.
Gray, D.A. LaPlante, H.J. Shaffer
17:00-17:20 Financial transactions: Comparing T-Pattern analysis with conventional methods – G.K.
Jonsson, M.T. Anguera & P. Sánchez-Algarra
17:20-18:00 Theme analysis of data collected with the ThemeWatch and ThemeSense apps – M.S.
Magnusson & G.K. Jonsson
20:00 Dinner: Restaurant “Backöfele”, Ursulinergasse 2, 97070 Würzburg
August 27th
Location: Zentrum für Mediendidaktik, room 005
Chair: Barbara Diana
GMT +2
09:00-09:40 The Influence of Libero Position in A Volleyball Game – B. Tenreiro, S. Fonseca & A. Lopes
09:40-10:20 Integration of categorical and continuous data in varied workout programs of physical
activity for women – M. Castañer, S. Puigarnau, O. Camerino & R. Benítez
10:20-11:00 T-Patterns of social interaction during Taekwondo combats – F.J.P. Cabrera & A.R.C,
Herrera
11:00-11:15 Coffee / Tea
11:00-11:40 T-pattern detection of motor skills used in goal scoring in soccer by Lionel Messi – M.
Castañer, D. Barreira, O. Camerino, M.T. Anguera, G.K. Jonsson, R. Hileno & A. Canton
11:40-12:40 Methodology for the analysis of social interaction: What’s next? (interactive session) –
M.S. Magnusson
12:40-13:00 Future of MASI: Renewal of Convention
13:00 Farewell lunch
Scientific Committee: M. Teresa Anguera, Alain Blanchet, Harvey Milkman, Frank Schwab and
Magnus S. Magnusson.
The Organizing Committee: Mariona Portell, Barbara Diana, Michael Brill and Gudberg K. Jonsson.
WELCOME ADDRESS
F. Schwab
University of Würzburg
No abstract
AN OVERVIEW OF RECENT RESEARCH INVOLVING THE T-PATTERN MODEL
AND THEME AND OTHER MILESTONES FROM MASI MEMBERS
G.K. Jonsson
University of Iceland
No abstract
DEMONSTRATION OF THEME / FROM STARDUST AND PROTEIN CITIES TO
SWARM INTELLIGENCE AND STRUCTURED SOCIETIES: T-PATTERNS IN A
SELF-SIMILAR FRACTAL WORLD
M.S. Magnusson
University of Iceland
Abstract
Mostly a 20th century mathematical phenomenon, Fractals, self-similar patterns of patterns,
are pervasive in nature as exemplified by the recently discovered fractal distribution of
matter in the universe out to the largest know structures, that is, clusters of clusters of
galaxies. From atoms to large structures of molecules such as amino acids to RNA and DNA,
most structure in the living world can be described as repetitions of hierarchical self-similar
structured clusters as witnessed, for example, by lung arteries, brains, writing and cities. Also
this general kind of structure is again seen in the real-time behavior and interactions of
organisms from tiny brain neurons to animals and humans.
Culminating in the structured societies of humans and social insects simpler social
phenomena are found in swarms composed by a number of animals that operate in a
coordinated fashion without the specialization of individuals characteristic of structured
societies.
Human and animal bodies are composed of specialized cells – also called Cell Cities or better,
protein cities - making up specialized body parts much as the specialized individuals
themselves make up the different functional parts of cities or hives. Thus social structures
can be seen as self-similar from the atoms making up the bodies to their cities.
T-patterns are self-similar structured repeated hierarchical clusters that can be considered
as a particular kind of statistical pseudo fractals, that is, natural fractals and seem relevant
for the discovery, analysis and description of the structure and functioning of numerous
natural phenomena in time and, when generalized two or three dimensions, also in space.
T-PATTERN METHODOLOGY FOR HUMAN BEHAVIOUR OBSERVATION AND
ANALYSIS: SOME RESEARCH APPLICATIONS
I. Terrenghi, B. Diana, M. Elia, V. Zurloni, A. Elia, M.S. & P.C. Rivoltella
University of Milano-Bicocca – Department of Human Sciences for Education, University
of Salerno – Department of Political, Social and Communication Sciences, University of
Salerno – Department of Human, Philosophical and Formation Sciences & Catholic
University of Milan – Department of Education
Abstract
A basic belief in the field of modern behavioral sciences is that behavior consists of
patterns in time. The T-pattern methodology is based on the assumption that complex
streams of human behavior have a hidden temporal sequential structure that cannot be
fully detected through unaided observation (or with the help of standard statistical and
behavior analysis methods). Recurring sequences of behavioral events, usually hard to
detect, can be thereby unveiled and carefully described. This kind of analysis has been
used in a wide variety of observational studies, including microanalysis of Drosophila
courtship behavior, cooperative behavior between humans and dogs when constructing
an object, complex patterns of neuronal networks’ activation. Human behavior patterns
include a multitude of every day events, routines, and processes of work and play (for
example, greeting rituals, a lunch, a religious ceremony or a sports match are all
patterns).
In this work, we present research results from the application of T-pattern methodology
on two different areas: deception detection and soccer performance. We will discuss the
suitability of this methodology in observing and analyzing both micro (as facial
movements in deceptive interactions) and macro (as ball passages in a soccer match)
components of human behavior.
We will then present an ongoing research application area for T-pattern methodology
that is education and teaching. We will specifically focus on the observation and analysis
of teachers’ nonverbal behavior during their lessons, aiming to explore if and how this
relate to communicative efficacy in terms of different outcomes of students’ learning
and engagement.
We will finally discuss some future research directions, which transversally involve the
possibility to combine t-pattern methodology with automatic feature extraction tools,
motion-capture systems and biofeedback equipment.
DETECTION OF T-PATTERNS AS AN IMPORTANT TOOL FOR STUDYING
CHANGES IN PHYSIOLOGICAL INTERVENTIONS FROM QUALITATIVE INPUTS
M. Teresa Anguera1, Mariona Portell2,
Salvador Chacón-Moscoso3, & Susana Sanduvete-Chaves3
1University of Barcelona, Spain
2Autonomous University of Barcelona, Spain
3University of Seville, Spain
Abstract
The aim of any psychological intervention program is to produce an improvement in the
issues, difficulties, or diseases faced by the users/participants. Here we are referring to
interventions in which the users/participants remain in their usual context (home,
school, work, hobbies, day center, nursing home, etc.) and follow their regular routines
with hardly any changes to their everyday lives.
Because of its flexibility, observational methodology can be used in a diverse range of
natural contexts in which it is not always possible to apply other methodologies. With
direct observation, records of behaviors (which can be more or less systematized) are
obtained from video recordings. These records can take various formats that, by itself,
are insufficient to evaluate the efficacy of a treatment.
Therefore, researchers need a system that allows them to objectively analyze the
changes brought about by a low-intensity intervention. If record data are processed
appropriately, it is possible to carry out a rigorous quantitative analysis in order to
evaluate the efficacy of an intervention. Nevertheless, previously it is necessary to decide
how to organize the heterogeneous information available.
The first part of the process is to correctly record and code the data, through some
decisions. As a consequence, we obtain a matrix of codes. The columns correspond to
the dimensions of observation instrument, while the rows correspond to the successive
units observed over time. The matrix is thus created by assigning each of the
observations or textual units to the corresponding category within the corresponding
dimension to ensure correct coding. Each row shows the string of codes for categories
that occur simultaneously. In this methodological paper we have worked with simulated
data that corresponds to a program intervention in clinical psychology, considering the
specific scientific literature.
The second part of the process consists of the adaptation of data matrix for each period
of intervention, in order to get the THEME vs. Edu software requirements. Previously, it
is very important to establish the criteria to differentiate groups of treatment sessions,
and periods, in order to aggregate and/or disaggregate data matrixes.
And The third and last part of the process consists on of the analysis, in order to detect T-
Patterns for each period, and considering several aggregations of sessions, with the aim
of distinguishing between them.
We expect that the changes of T-Patterns structures could be an important evidence for
the evaluation of low intervention programs.
PATTERNS OF TOPIC ORGANISATION IN TURN MANAGEMENT. A CASE
STUDY OF INTERACTIONS IN DIALOGUES
L. Hunyadi
University of Debrecen, Hungary
Abstract
The HuComTech corpus has been under development since 2009 with the aim to better
understand the multimodal nature of human-human interactions based on dialogues in
Hungarian. The total duration of recordings is more than 50 hours of dialogues between
an agent and 110 speakers both in formal and informal conditions.
The task to arrive at reasonable generalizations about such interactions appears to be
rather complex due to several factors, including the following: (a) a given communicative
function can be expressed by more than one modality, (b) the simultaneous participation
of these modalities in such expressions can be optional, (c) the sequences of multimodal
events leading to the interpretation of the given function need not be contiguous, (d)
behavioral patterns vary according to contexts, and (e) there is a considerable variation
in behavioral patterns among individuals across all these conditions.
In order to capture multimodality in a possibly wide range of events, both audio and
video were annotated for several features, some of them more than once for different
combinations of audio and video. Accordingly, unimodal video (video only) was
annotated for gaze, handshape, head shift, posture, perceived emotions, turn
management and pragmatic functions. Unimodal audio (audio only) was annotated for
transcription, prosody (pitch, intensity, pause, tempo), perceived emotions, turn
management and pragmatic functions. Multimodal video+ and audio was annotated for
turn management and pragmatic functions. The total number of annotated items
exceeded 1.5 million.
Although descriptive statistics reveals a lot about the frequency of the occurrences of
various items, and it is possible to quantify the alignment and sequences of such items,
these traditional methods are less suitable for making generalizations about behavior
beyond surface observations in order to possibly meet the challenges mentioned above.
Our goal is to understand the hidden patterns of multimodal behavior in dialogues as
represented in our data using Theme’s framework in the hope that the findings will be
useful for further research in several fields including linguistics, psychology, cognitive
science and robotics.
As a case study, we are presenting multimodal behavioral patterns of topic organization
in turn management involving the functions of topic initiation, topic elaboration and
topic change. All 110 available recordings of both formal and informal dialogues will be
considered offering and demonstrating both generalizations across speakers and
individual variations between them.
COMPLEMENTARITY BETWEEN T-PATTERNS AND POLAR COORDINATES
ANALYSIS IN SOCIAL INTERACTION: HOW TO INTEGRATE THE RESULTS?
M. Teresa Anguera1, Gudberg K. Jonsson2, & Pedro Sánchez-Algarra1
1University of Barcelona, Barcelona, Spain
2University of Iceland, Reykjavík, Iceland
Abstract
In the study of social interaction, we consider that observational methodology is the best
approach to study the hidden structures underlying an interactive situation, and from a
previous observation an instrument was build ad hoc. This implies some dimensions and a
category system for each dimension, and later a systematic record of episodes in the same
situation. Two techniques of data analysis, detection of T-Patterns and polar coordinates
analysis imply a possible convergence in the results of the study on social behavior,
independent of subject or field of study.
Detection of T-Patterns implies discovering the hidden relations that should reveal those
aspects of social interaction that are not immediately observable. The recorded episodes of
behavior are governed by structures of varying stability, and can be visualized by obtaining
T-Patterns, that have proven to be an exceptional analytical tool. These temporal patterns
can be detected with THEME 6.0. Polar coordinates analysis searches for a vectorial image of
the complex network of interrelations between categories that make up the different
dimensions of the observation instrument. The values of length and angle of vectors, and
their graphical representation is achieved with HOISAN, a free software.
The aim of this paper is to propose some guides to integrate the results from both data
analyses.
Method: We worked with two databases of anonymized data from interactive behavior
(clinical psychology and sport behavior), in order to analyze comparatively two techniques of
data analysis, the T-Patterns detection and polar coordinates analysis. Both techniques of
data analysis have the common aim, that is goal to discover some hidden relations between
observed behaviors, but each one has a different algorithm and goal.
Results: We compared the degree of similarity between the results of codes relations
obtained from both techniques. Also, we propose a guide to facilitate researchers the
integration of results.
Keywords: T-Patterns, polar coordinates, parameters, variability of relations
LMX AND LEADERSHIP BEHAVIOR WITH T-PATTERN ANALYSIS
Merve Tarim1, & Magnus S. Magnusson2
1Department of Management, Istanbul Commerce University, Turkey
2Human Behavior Laboratory, University of Iceland, Reykjavik, Iceland
Abstract
In a changing world, organizations need to establish strong ties between the managers and
the employees to survive because only these kinds of organizations can live. To establish
strong ties, both sides have to give labor because these ties can exist in an exchange process.
If the manager is a leader at the same time, he/she is the person who can start this process.
In the literature, this exchange process called LMX (leader-member exchange theory).
According to LMX, leader and the member have a dyadic relationship. To build it, both of
them should give emotional labor.
This is my doctoral thesis research and it still continues. In this research, interviews were
conducted with 27 people which were recorded with camera. Six of them are managers and
21 of them are employees. The data were collected from three different universities’
academic and administrative staff. The employees evaluated the relationship with their
managers, their managers’ behaviors and also the managers’ personalities. The managers
evaluated the employees’ behaviors and the relationship with them.
Data were coded with the behavior coder according to the specified categories (non-verbal).
Then the codes were then transferred to Theme to determine the relationship. The main aim
is to understand the relationships between the managers and the employees determine the
strength of their ties. We hypothesize that the leaders with strong ties have different
behaviors from the leaders with weak or no ties. We also want to determine if there are
common behaviors in leaders. Is it possible to say that the leaders’ personalities effect the
ties between the leaders and their employees? There are many questions and the research
has not finished yet. Data were coded only with non-verbal categories so the categories will
increase like verbal ones. Patterns were found but not enough to answer for these
questions. However, the research will be different to see leadership behavior in patterns and
try to discover the other things with Theme.
DETECTION OF LUDIC PATTERNS IN TWO MOTOR TRIADIC GAMES
Miguel Pic Aguilar1, Vicente Navarro Adelantado1 & Gudberg K. Jonsson2
1University of La Laguna (Spain)
2University of Iceland
Abstract
The triad is a particular structure in which an ambivalent social relationship takes place.
A chasing games model was followed, with rules, and in two different structures
(A↔B↔C↔A, y and? A→B→C→A) on four class groups (two for each structure), for a
total of 84, 12 and 13 year olds secondary school students, 37 girls (44%) and 47 boys
(66%). The aim was to examine if the players’ behavior, in relation to the triad structure,
matches with any ludic behavior patterns.
An observational methodology was applied, with a nomothetic, punctual and
multidimensional design (Anguera, Blanco & Losada, 2001; Anguera, 2003). The intra
(0,973) and inter-evaluative (0,964) correlation coefficients and the generalizability
theory ensured the quality of the data. A mixed behavioral role system was used (4
criteria and 17 categories), and the pattern detection tool Theme (Magnusson, 2000)
was applied.
The results show that time location of motor responses in triad games was not random.
The ‘labyrinth’ game got a dendrogram with more ludic patterns than the ‘three fields’
game, which might be explained by means of structural determinants such as circulation.
This research warns about the decisional complexity in motor games, and it confirms the
differences among triads from the point of view of motor communication.
KINEMO SOFTWARE: TOWARDS AUTOMATED GESTURE ANNOTATION WITH
MS KINECT
Konrad Juszczyk and Kamil Ciecierski
Adam Mickiewicz University, Poznań, Poland
Abstract
The aim of the presentation is to introduce the architecture of the initial version of the
software application for automated gesture annotation called KINEMO and discuss
possibilities of its development. Research on gestures relies mainly on manual
annotation. Reliability of manual annotation depends on researchers, trained raters and
on the adequate coding systems. However, manual annotation of spontaneous gestures
used in conversation is known to be very time-consuming and laborious, reaching times
100 longer than the length of the annotated media [1]. In order to achieve
intersubjectivity of studies on gesture, high level of consistency in coding system and
annotation process is needed. Disadvantages of manual annotation lead gesture studies
to limited amount of data annotated with low or not even reported interrater agreement
[2]. This directly threatens transparency and replicability of many studies. However,
measurements used in research on gestures, such as gesture rate (per minute or per
word) and amounts of certain gesture types or gestures with certain features, such as
gesture space, hand-shape and orientation or gesture phase can be supported with
automatic gesture recognition software.
The input data comes from MS Kinect v2 motion sensor. During recognition phase the
program relies on LASG [3] and NEUROGES [4] coding systems as the theoretical
frameworks. The output data consists of labelled hand movements and can be exported
to the MS EXCEL, MPI ELAN or THEME software. Evaluation of the software performance
was conducted on dialogues and monologues (20 participants and 400 minutes in total)
recorded with MS Kinect and KINEMO software. Selected dialogues and monologues
were annotated manually by three trained raters using NEUROGES coding system [4].
Following suggestion given in NEUROGES coding system [5], to avoid circularity in
gesture classification, mute videos and kinesic criteria were used for annotation; to
establish reliability one-fourth of the material was coded by all three raters and to avoid
habituation samples were given in pseudo-randomized fashion. Average interrater
agreement calculated using modified Cohen’s KAPPA [6] was substantial (0.6). Results of
manual annotation were compared with results of automatic annotation and F-score
values were obtained for movement (0,80), phasic (0,62), in space (0,70). Other types of
hand movements based on NEUROGES, such as on body or in space from focus step or
left/right hand dominance from Formal Relation step and baton or palm-out from type
step are also possible to be recognised and this recognition in is the process of
development.
The application software presented here is still in its development phase. Nevertheless,
initial results suggest that automatic annotation based on material recorded by MS
Kinect v2 is possible. We hope to simplify and speed-up the mundane manual annotation
process and raise the transparency and replicability of research in social interaction.
Demo version of the software will be released in Autumn 2016.
The project is co-financed by the grant from the Foundation for Polish Science under the
contract nr 126/UD/SKILLS/2013 on making use of the award granted in the popularizing
competition within the framework of the SKILLS project co-financed by the European
Social Fund.
[1] Lenkiewicz, P., Auer, E., Schreer, O., Masneri, S., Schneider, D., & Tschöpe, S. AVATecH ―
automated annotation through audio and video analysis. In N. Calzolari (Ed.),
Proceedings of LREC 2012: 8th International Conference on Language Resources and
Evaluation (pp. 209-214). European Language Resources Association. 2012.
[2] Stec, K. in prep. Reporting practices in Multimodal Viewpoint Research. A methodological
meta-analysis. in GESTURE.
[3] Bressem, J. "A linguistic perspective on the notation of form features in gestures". In:
Cornelia Müller, Alan Cienki, Ellen Fricke, Silva H. Ladewig, David McNeill & Sedinha
Teßendorf (Eds.) (forthcoming). Body-Language-Communication: An International
Handbook on Multimodality in Human Interaction. (Handbooks of Linguistics and
Communication Science 38.1.) Berlin, Boston: De Gruyter: Mouton, 2013.
[4] Lausberg, H., & Sloetjes, H. Coding gestural behavior with the NEUROGES-ELAN system.
Behavior Research Methods, 41(3), 841–849, 2009.
[5] Lausberg, H. Understanding Body Movement. A Guide to Empirical Research on
Nonverbal Behaviour With an Introduction to the NEUROGES Coding System. Frankfurt
am Main: Peter Lang, 2013.
[6] H. Holle and R. Rein, “EasyDIAg: A tool for easy determination of interrater agreement,”
Behav Res Methods, vol. 47, no. 3, pp. 837–847, Sep. 2015.
EXPANDING THE STUDY OF INTERNET GAMBLING BEHAVIOR
Gudberg K. Jonsson1, Harvey Milkman2, Heather M. Gray3,
Debi A. LaPlante3, & Howard J. Shaffer3
1University of Iceland, 2Metropolitan State University of Denver, 3Harvard Medical School
Abstract
As rates of Internet gambling participation increase worldwide, so too does the need to
understand how people engage in this form of gambling. This study represents the first
examination of actual Internet gambling records within Iceland, a Nordic country with an
active Internet lottery market that imposes strict regulations on gambling operator licenses.
We summarized electronic betting records of a cohort of subscribers to the Internet betting
service provider Íslensk Getspá. In addition, we searched for temporal patterns in individual
gambling records. We observed that the typical subscriber bet approximately 3 days per
month and made fewer than two bets per gambling day, each worth approximately the
equivalent of $4 US. Subscribers lost the bulk (96 %) of the amount they wagered, for a total
loss of approximately $40 across the 2-year window of observation. Although these
observations do not support the view of Internet gambling as an activity that is inherently
risky for the typical subscriber, we did observe discontinuity across the distributions of
gambling behavior, with the top 1 % of subscribers making more than three bets per day.
Certain betting patterns were exclusively detected with frequent gamblers and different
pattern types between gender and age groups.
THEME ANALYSIS OF DATA COLLECTED WITH THE THEMEWATCH AND
THEMESENSE APPS
Magnusson, M.S. & Jonsson, G.K.
Human Behaviour Laboratory, University of Iceland
Abstract
The tech world is creating a future of wearable devices that promises to entertain
consumers and help them live healthier lives. Technology companies’ interests in health
and wellness have sparked the creation of a myriad of wearable devices, from fitness
bands that monitor activity and sleep patterns to flexible patches that can detect body
temperature, heart rate, hydration level and more. These devices produce data that,
often enabled with analytics, can be used by consumers to manage their health and
potentially increase well-being and reduce costs through systems such as remote
monitoring. As wearable technology becomes cheaper and more sophisticated, and data
quality improves, these devices and their associated apps will become a part of
consumers’ lives and the health ecosystem.
The current study focuses on pilot data obtained with the apps ThemeWatch and
ThemeSense, installed on iOS and android wearable devices. The data was collected over
a three 3-month period by 3 three different individuals and analyzed with the Theme 6.0
software. Results for these 3 three pilot projects and future directions will be discussed.
THE INFLUENCE OF LIBERO POSITION IN A VOLLEYBALL GAME
Tenreiro, B., Fonseca, S., & Lopes, A.
Faculty of Physical Education and Sports, Lusófona University, Portugal
Abstract
Volleyball is a complex and unpredictable game, which due to regulatory constraints
does not permit a close relationship between teams, making vital the technical and
tactical aspects between elements of the same team for defensive and offensive success.
The FIVB introduced in 1998 the libero position, making it later a very important element
in defense.
This research project seeks to identify and understand the variables of libero action and
influence in a high-performance volleyball team, in the phase of the game complex 1
(reception service) and complex 2 (service itself). Thus, the study aims to use the
dynamical systems theory in order to explain the influence of this player in the game,
through the study of the combination of technical and tactical variables and spatial
measures that can predict the defensive and offensive success, before, during and after
his contact with the ball.
The samples used for this study will be in complex 1 and 2, when the libero has direct
contact with the ball, in the 2014 Men’s Volleyball World League covering six teams with
a total of eleven 11 games spanning forty and three 43 sets played.
Firstly, With the help of statistical data, the study will begin by examining the libero’s
actions in its team. Secondly, Data with the libero’s in game positioning using nonlinear
techniques will then be examined to pinpoint movement patterns related to the
success/failure of the team. This will enable a more complete analysis of the libero
position in a highly efficient teams.
Keywords: volleyball, libero, dynamic systems theory, spatial measures, performance
analysis.
INTEGRATION OF CATEGORICAL AND CONTINUOUS DATA IN VARIED
WORKOUT PROGRAMS OF PHYSICAL ACTIVITY FOR WOMEN
Marta Castañer1, Sílvia Puigarnau1, Oleguer Camerino1, & Raúl Benítez2
1National Institute of Physical Education, University of Lleida, Spain.
2Politecnic University of Barcelona, Spain
Abstract
The aim of this study was to use an embedded mixed methods design to investigate
whether a workout program incorporating greater work intensity levels and a wider
variety of motor skill patterns led to changes in heart rate, perceived exertion, and
subjective experiences among four groups of women (n=73; mean ± SD age of 65.1±11.7
years) taking a different type of exercise class at a public gym.
Data was collected during and after the women’s routine class and during and after the
newly designed class. Heart rate data was collected by continuously monitoring with
Polar Team 2 (version 1.4.5), perceived exertion was analyzed using the Borg Rating of
Perceived Exertion Scale, and subjective experiences were evaluated using a modified
version of the Intrinsic Motivation Inventory containing several open-ended questions.
Exercise patterns were analyzed and coded using an ad hoc observation instrument
within an observational methodology design.
The methodological challenge was to employ an analytical method that would allow the
integration of categorical and continuous data. To this end, we prepared a MATLAB code
to extract the information for automatic analysis by means of a hidden Markov model.
The results show that the more intense, varied workout was more effective than the four
more routine, repetitive workouts in terms of achieving the moderate- and vigorous-
intensity physical activity levels recommended by the American College of Sports
Medicine. The questionnaire and Borg Scale results also showed that following initial
resistance to changes to workout routine, the women felt that the varied program was
more enriching and suited to their exercise needs. Better designed exercise programs
incorporating a greater variety of motor skill patterns and intensity levels can bring about
actual and perceived physical and psychological benefits.
Keywords: Adult women, physical activity, exercise patterns, perceived benefits, mixed
methods, heart rate.
T-Patterns of Social Interaction During Taekwondo Combats
Francisco Javier Pedroza Cabrera & Ana del Refugio Cervantes Herrera
Universidad Autónoma de Aguascalientes, México
Abstract
The main role of a psychologist in sport environment is to find the variables involved in
the success and failure of sport people, in order to explain and predict the trajectory of
athletes. To get reach this goal direct observation seems to be the right way to follow
because it provides reliable information about the topography, frequency, durations, and
rate about the conduct. Nevertheless, most of the work has been developed with
indirect tools like questionnaires or interviews, leaving aside the direct data of the
performance in competition.
Taking this information into account, the goal of this work was to find T-patterns in
Taekwondo fights that show the differences in the behavior between the athletes that
wins and those that lose. Eighty-four (84) black belt Taekwondo athletes between 13 and
19 years old (mean age of 16.33 years old) were contacted, in order to obtain the signed
informed consent form required to participate in the research. These athletes belong to
categories, sub13, sub15, sub17 y and sub 20.
Afterwards, 54 Taekwondo fights were videotaped during official tournaments in
Aguascalientes, México. Those fights were registered by trained observers (with a
Kappa=0.84) in the software Observer TX using the Direct Observational System for the
Performance of the Athlete During Combat. It consists of 52 behaviors that include
offensive moves, defensive behaviors, unsportsmanlike conduct and use of tactical
resources. Once finished the register, the information was analyzed in the THEME V.
Taking into account the patterns involving one athlete or between fighters, differences
were found in the T-Patterns in all fighting categories in terms of the type and order of
behaviors shown by the winners and those that lost the competition. The implications of
these findings on the design of training strategies are discussed.
T-PATTERN DETECTION OF MOTOR SKILLS USED BY LIONEL MESSI
IN SCORING GOALS SCORING IN THE GAME OF SOCCER
Marta Castañer1, Daniel Barreira2, Oleguer Camerino1, M. Teresa Anguera3,
Gudberg K. Jonsson4, Raul Hileno1, & Albert Canton1
1 INEFC,
University of Lleida, Spain
2Faculty
of Sport, University of Porto, Portugal
3 Research Institute IR3C, University of Barcelona, Spain
4 Human Behaviour Laboratory, University of Iceland, Reykjavik, Iceland
Abstract
Soccer research has traditionally focused on technical and tactical aspects of team play,
but few studies have analyzed motor skills in individual actions, such as goal scoring.
Besides, the study of skills in elite players is still largely based on subjective judgements.
Versatility is closely linked to motor skills, as players with a mastery of these skills are
able to process information faster and react to changing game situations. Strikers need
to be multi-skilled and acquire techniques and physical qualities specific to their
positional roles.
We consider that the individual skills of one of the world’s top players, Lionel Messi, who
holds numerous goal records and individual awards, such as five FIFA Ballons d’Or, three
European Golden Shoes, and FIFA World Player of the Year, are worthy of objective,
scientific analysis. The objective of this study was to investigate how Lionel Messi uses
his motor skills and laterality in individual attacking actions resulting in a goal. We
analyzed 103 goals scored by Messi between 2004 and 2014 in three competitions: La
Liga (n = 74), Copa del Rey (n = 8), and the UEFA Champions League (n = 21). The goal
inclusion criteria were a) clear observability of each sequence and b) availability of at
least two recordings of each sequence from a different angle. The exclusion criteria were
a) recordings that, while clear, did not allow for coding of the sequences and b) goals
scored directly following receipt of the ball.
We chose an observational methodology design for the purpose of our study, as this
method has proven effective in the analysis of attacking play in soccer. We used an ad
hoc observation instrument (OSMOS-soccer player) from the Motor Skills Observation
Instrument OSMOS, comprising 12 criteria and 53 categories and THEME v.6 to obtain t-
pattern T-Pattern detection. We obtained several t-patterns T-Patterns that confirms
significant associations between different aspects of motor skill used by Messi
immediately before scoring, namely use of lower limbs, foot contact zones, turn
direction, use of wings, and orientation of body to move towards the goal. As it is
expected, he uses his left foot significantly more than his right foot to score goals but this
condition aids his postural support that enables stasis and blocks movement concerning
of turns and pivot directions.
T-patterns Detection for a Triad Motor Game: The Maze
Miguel Pic-Aguilar1, Gudberg K. Jonsson2, Carmen R. Sánchez-López1, and Vicente
Navarro-Adelantado1
1
La Laguna University & 2Iceland University
Abstract
In physical education, triad motor games are a clear minority amongst their traditional
dual versions. This preference for dual sports, rather than other structures such as the
triad (Caplow, 1956), transcends the field of motor games. To investigate this subject and
organize the registration of motor behavior, we have been helped by the praxiology
(Parlebas, 1981, 1986).
Motor behavior records were compiled using the Lince software (Gabín, Camerino,
Anguera and Castañer, 2012). The sample was composed of thirteen boys and eight girls.
All twenty-one players, aged between twelve and thirteen, enrolled in the 1º course of
ESO (Spanish Secondary Education), practiced a modified version of “the maze” (Navarro,
1995). An ad hoc category system was built. This was based on the original game rules
and role-play system: ‘chaser’, ‘runaway’, ‘prisoner’, ‘liberating’. Motion behaviors for
each of the roles resulted in a combination of seventeen categories following the team
A↔B↔C↔A as a capture formula. The procedure to play the game was to first explain
it and then practice before being recorded, so we could resolve any doubts. After that,
we made the record for three minutes 'from beginning to the end' (Anguera, 1990).
Observational methodology allowed the analysis of the players’ motion behavior, that
after having reached the expected reliability (Pic, Navarro, 2014) the study figures could
be extrapolated to other samples. Through the Theme software, (Magnusson, 1996,
2000, 2006) T-patterns occurring during the game were detected. This showed stronger
(game) structures and interactions, helping us to explain the particularities of the team
interactions associated to the triad.
Keywords: triad motor game, t-patterns, methodology, praxeology
METHODOLOGY FOR THE ANALYSIS OF SOCIAL INTERACTION:
WHAT’S NEXT?
Interactive session
No abstract