APPLIED BEHAVIOR

ANALYSIS
Cooper · Heron · Heward — Second Edition

Comprehensive Chapter Notes

29
All 29 Chapters · Definitions · Examples · Non-Examples · Exam Tips
CHAPTERS

QASP-S IBA QBA

BEHAVIOR CANVAS
For personal study use only
Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

TABLE OF CONTENTS

Ch 1 Basic Characteristics of Science & ABA

Ch 2 Basic Concepts — Behavior, Environment, Conditioning

Ch 3 Selecting & Defining Target Behaviors

Ch 4 Measuring Behavior — Dimensions & Recording Methods

Ch 5 Improving & Assessing Measurement Quality (IOA)

Ch 6 Constructing & Interpreting Graphic Displays

Ch 7 Analyzing Behavior Change — Experimental Control

Ch 8 Reversal & Alternating Treatments Designs

Ch 9 Multiple Baseline & Changing Criterion Designs

Ch Planning & Evaluating ABA Research

10

Ch Positive Reinforcement — Types, Schedules, Preference Assessment

11

Ch Negative Reinforcement — Escape, Avoidance

12

Ch Schedules of Reinforcement — FR, VR, FI, VI, Compound

13

Ch Punishment by Stimulus Presentation (Type I)

14

Ch Punishment by Removal of a Stimulus (Type II)

15

Ch Motivating Operations — EO, AO, CMO-R/T/S

16

Ch Stimulus Control — SD, S∆, Discrimination, Transfer

17

Ch Imitation — Definition, Formal Similarity, Training

18

Ch Shaping — Successive Approximations, Guidelines

19

Ch Chaining — Task Analysis, Forward/Backward/Total Task

20

Ch Extinction — Burst, Spontaneous Recovery, Renewal, Resurgence

21

Ch Differential Reinforcement — DRO, DRA, DRI, DRL, DRH

22

Ch Antecedent Interventions — NCR, High-p, FCT, Setting Events

23

Ch Functional Behavior Assessment — Methods, FA, Functions

24

Page 2 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Ch Verbal Behavior — Skinner, Operants, Autoclitic, Extensions

25

Ch Contingency Contracting, Token Economies, Group Contingencies

26

Ch Self-Management — Self-Monitoring, Strategies, Applications

27

Ch Generalization & Maintenance — Strategies, Stimulus Equivalence

28

Ch Ethical Considerations for Applied Behavior Analysts

29

Page 3 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Definition & Characteristics of ABA

Science, Attitudes, Development of ABA
1

Science — 3 Levels of Understanding

Definition
A systematic approach for seeking and organizing knowledge about the natural world. In
ABA, the phenomena of interest are socially important behaviors.
Description (Level 1)
A collection of facts about observed events that can be quantified, classified, and
examined for possible relations with other known facts. Often suggests hypotheses for
further research. No causal claims possible.
Prediction (Level 2)
Repeated observations revealing that two events consistently covary. Demonstrates
correlational relationships — NOT causal. Enables preparation but cannot confirm cause.
Control (Level 3)
Highest level. A specific change in the dependent variable (DV) is reliably produced by
manipulations of the independent variable (IV). Functional relations derived. Extraneous
factors ruled out.

✓ EXAMPLE — Description: Observer notes a child elopes every time the mother holds up a
fresh diaper. This is description — facts recorded, no cause claimed.
✓ EXAMPLE — Prediction: After 3 days of observation, analyst predicts elopement will occur
whenever diaper change is signalled. Correlation established.
✓ EXAMPLE — Control: Token board introduced; elopement decreases reliably. Removed;
elopement returns. Reintroduced; elopement decreases again. Functional relation established.
✗ NON-EXAMPLE: Writing "Client looks sad" is NOT description — "sad" is subjective, not
measurable. Description requires objective, quantifiable language.

★ EXAM TIP: Description → Prediction → Control is a progression. Each builds on the last.
Only CONTROL demonstrates a functional relation. Questions often test whether you can
identify which level is being described.

6 Attitudes of Science

Page 4 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Determinism
The universe is a lawful and orderly place. All phenomena occur as the result of other
events — nothing happens randomly. Behavior has causes. The ABA analyst always
looks for causes, never accepts "it just happens."
Empiricism
The practice of objective observation and measurement of the phenomena of interest. All
scientific knowledge is built on objective observation. Subjective language has no place in
empirical science.
Experimentation
A careful comparison of the DV under two or more conditions where only one factor (IV)
differs at a time. The basic strategy of most sciences.
Replication
Repeating experiments to determine the reliability and usefulness of findings. Method by
which errors are discovered. Includes repetition of IV conditions within a single study AND
across independent studies.
Parsimony
Simple, logical explanations must be ruled out before more complex or abstract
explanations are considered. Prevents over-interpretation and keeps science grounded.
Philosophic Doubt
Continual questioning of the truthfulness and validity of all scientific theory and knowledge.
Requires scientific evidence before implementing any practice, then constant evaluation of
effectiveness.

✓ EXAMPLE — Determinism: A BT reports a client spitting "out of nowhere." Analyst

observes and identifies that spitting occurs after sister wears perfume and walks past.
Behavior has a cause — it was not random.
✓ EXAMPLE — Parsimony: Neighbor walks past without responding to wave. Rather than
concluding anger, you conclude she had headphones in — the simplest, most logical
explanation.
✗ NON-EXAMPLE — Determinism: Analyst says "sometimes these regressions just happen."
This violates determinism — there is always a cause to be found.
✗ NON-EXAMPLE — Philosophic Doubt: New analyst accepts previous agency's FBA
function without question despite failed interventions. Failure to question is a violation of
philosophic doubt.

★ EXAM TIP: DEEP-RP mnemonic — Determinism, Empiricism, Experimentation,

Parsimony, Replication, Philosophic Doubt. Know definitions AND
examples/non-examples for each.

7 Dimensions of ABA (GEAT-CAB / Baer, Wolf & Risley, 1968)

Page 5 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Applied
The behavior chosen for study must be socially significant — it must genuinely matter and
improve lives. Not laboratory curiosity. Targets chosen for their importance to the
individual and society.
Behavioral
Behavior must be observable and measurable. Must pass the Dead Man's Test — if a
dead man can do it, it is not a behavior. Focuses on what the person DOES, not inferred
internal states.
Analytic
The experimenter demonstrates a functional relation: the behavior changes reliably when
and only when the IV is manipulated. Control over the occurrence and non-occurrence of
behavior.
Technological
All procedures are described completely, precisely, and operationally — any trained
practitioner can replicate them exactly from the written description alone.
Conceptually Systematic
Procedures are described in relation to the relevant behavior analytic principles from
which they are derived (e.g., reinforcement, extinction). Not isolated techniques.
Effective
The intervention produces practical, socially significant improvement — not just
statistically significant change. The size of the effect matters, not just its existence.
Generality
Behavior changes are durable across time (maintenance), settings (stimulus
generalization), and behaviors (response generalization). The effect outlasts the
intervention.

✓ EXAMPLE — Behavioral: Staff takes data on out-of-seat behavior defined as: any instance
the client stands up from his seat and moves at least 1 step away from the desk. Observable,
measurable, passes Dead Man's Test.
✓ EXAMPLE — Effective: Aggression reduced from 15x/hour to 0.1x/hour. This is socially
significant change — the client's life is meaningfully better.
✗ NON-EXAMPLE — Behavioral: "Daydreaming" is NOT a valid behavioral target — it is not
directly observable or measurable. Cannot pass Dead Man's Test reliably.
✗ NON-EXAMPLE — Applied: Labeling cards and matching to sample when the client is 16
and has already mastered and generalized these skills. Not socially significant for current
age/level.
✗ NON-EXAMPLE — Technological: Telling staff "run a DTT program" without specifying the
exact SD, response criteria, consequence, inter-trial interval. Not replicable.

★ EXAM TIP: GEAT-CAB mnemonic. The 7 dimensions are examinable in all three formats
— definition, example identification, and non-example identification. Especially know:
Applied (socially significant) vs. Effective (produces change), and Behavioral
(observable/measurable + Dead Man's Test).

Page 6 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Development of ABA

Experimental Analysis of Behavior (EAB)

Basic science studying behavior-environment relations in controlled conditions. Uses
non-human subjects. Not focused on social significance. Precursor to ABA.
Applied Behavior Analysis
Application of behavioral principles to socially significant human behavior. Baer, Wolf &
Risley (1968) defined the 7 dimensions. Science + application combined.
Behaviorism
The philosophical foundation. Skinner (1953) — behavior is determined by its
consequences; all behavior can be studied scientifically. Radical behaviorism includes
private events.
Methodological Behaviorism
Only publicly observable behaviors are studied. Private events are excluded. Watson's
position. More restrictive than radical behaviorism.

Page 7 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Basic Concepts
Behavior, Environment, Respondent & Operant Conditioning
2

Defining Behavior

Behavior (Johnston & Pennypacker, 1993)

The portion of an organism's interaction with the environment that involves detectable
displacement in space through time of some part of the organism, resulting in a
measurable change in at least one aspect of the environment.
Response
A specific instance of behavior. A single occurrence of the broader behavioral class.
Response Class
A group of responses sharing the same function — each produces the same effect on the
environment. Topography may differ; function is the same.
Repertoire
All behaviors a person can emit. The complete set of knowledge, skills, and behaviors
relevant to a given context.
Topography
The physical form of a behavior — what it looks like. Does not tell us why it occurs.
Function
The effect the behavior has on the environment — its purpose. Why it occurs.

✓ EXAMPLE — Response Class: A child screams, hits the table, drops to the floor, or
crumples paper — all to escape homework. Same function, different topographies. These form
one response class.
✗ NON-EXAMPLE — Response Class: A child screams for attention, screams to escape,
screams when in pain. Same topography but different functions — NOT the same response
class.

★ EXAM TIP: Topography = form (what it looks like). Function = purpose (what it does). A
single topography can serve multiple functions. A single function can be served by
multiple topographies.

Environment

Page 8 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Environment
All behavior occurs within an environmental context. Behavior cannot be emitted in an
environmental void. Includes both external (physical) and internal (biological) stimuli.
Stimulus
Any energy change in the environment that affects behavior. Events that have known or
probable functional relations with behavior.
Stimulus Class
A group of stimuli sharing one or more formal, temporal, or functional features that all
evoke the same response.
Response Class
Responses that, though topographically different, all produce the same consequence
(have the same function).

✓ EXAMPLE — Stimulus Class: Pictures of cars, toy cars, real cars on the street, the word
"car" — all may evoke the response "car" in a child learning to label. Formal similarity.
✗ NON-EXAMPLE — Stimulus Class: Random objects at a birthday party scavenger hunt
(blue thing, star-shaped thing, sparkly thing) — no shared formal, temporal, or functional
features.

Respondent Conditioning (Classical Conditioning / Pavlov)

Respondent Behavior
Behavior elicited (drawn out) by antecedent stimuli. Reflexive. The response component
of a reflex. Involuntary.
Unconditioned Stimulus (UCS)
A stimulus that elicits a response without any prior learning. Biologically significant.
Unconditioned Response (UCR)
The response elicited by the UCS without prior learning.
Conditioned Stimulus (CS)
A previously neutral stimulus (NS) that, after repeated pairing with a UCS, acquires the
ability to elicit a conditioned response.
Conditioned Response (CR)
The learned response elicited by the CS alone, after conditioning.
Neutral Stimulus (NS)
A stimulus that does not initially elicit the target response. Becomes a CS after pairing.
Stimulus-Stimulus Pairing
The process of repeatedly presenting an NS with a UCS until the NS becomes a CS and
elicits a CR.

Page 9 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ EXAMPLE: Dog salivates when it sees you put on tennis shoes (paired with going for
walks). Tennis shoes = CS. Salivation to tennis shoes = CR. Walking = UCS. Salivation to
walking = UCR.
✗ NON-EXAMPLE: A child cries whenever candy is given, because crying has been reinforced
by candy. This is operant conditioning (behavior→consequence), not respondent
(stimulus→reflex response).

★ EXAM TIP: Respondent = ELICITED (triggered by antecedent). Operant = EMITTED

(influenced by consequences). Respondent behavior is involuntary. The key test: does an
antecedent CAUSE the behavior reflexively, or does a consequence MAINTAIN it?

Operant Conditioning

Operant Behavior
Behavior selected, shaped, and maintained by its consequences. Emitted — not just
elicited. Voluntary. Influenced by what follows it.
Operant Conditioning
The process of changing the frequency of a behavior by manipulating its consequences.
Behavior → Consequence → change in future behavior.
Three-Term Contingency (ABC)
Antecedent → Behavior → Consequence. The fundamental unit of analysis in ABA.
Behavior occurs in the presence of an antecedent and is followed by a consequence.
Four-Term Contingency
MO → Antecedent (SD) → Behavior → Consequence. The motivating operation sets the
context for the entire ABC sequence.
Phylogeny
Behaviors influenced by the evolutionary history of the species. Inherited/reflexive
behaviors.
Ontogeny
Behaviors acquired through the organism's individual learning history and experience.
Includes all conditioned behaviors.

✓ EXAMPLE — Operant: Got a speeding ticket ($300) and haven't sped since. The
consequence (ticket) changed future behavior (speeding). Operant conditioning.
✓ EXAMPLE — 4-Term: Hunger (MO) → Mom says "Come eat" (SD) → Child walks to table
(Behavior) → Child receives food (Consequence).
✗ NON-EXAMPLE — Operant: Sweating when it's hot is NOT operant conditioning — it is a
reflexive biological response with no learning history. Respondent behavior.

★ EXAM TIP: If it's because of a CONSEQUENCE → Operant. If it's because of an

ANTECEDENT STIMULUS alone → Respondent. Jumping at a loud bang = respondent.
Avoiding loud places = operant (learned to avoid the consequence).

Page 10 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Page 11 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Selecting & Defining Target Behaviors

Assessment Methods, Operational Definitions, Social Validity
3

5 Phases of Assessment

Phase 1 — Screening
Initial identification of areas of concern. Broad overview. Determines if further assessment
is warranted.
Phase 2 — Defining the Problem / Goals
Specifying the behaviors to be changed and criteria for acceptable performance.
Phase 3 — Pinpointing Target Behaviors
Selecting specific, operationally defined behaviors as intervention targets.
Phase 4 — Monitoring Progress
Ongoing data collection to evaluate whether the intervention is working.
Phase 5 — Follow-Up
Assessing whether gains are maintained after intervention has concluded.

Assessment Methods

Page 12 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Indirect Methods
Information gathered without direct observation: interviews, checklists, questionnaires,
rating scales. Faster but less accurate. Provides impressions and opinions, not hard
behavioral evidence. Impressions may change over time.
Direct Methods (Preferred)
Information gathered through direct observation of the client in the natural environment or
structured settings. Tests, criterion-referenced assessments, structured observations.
Most accurate.
Behavioral Interview
Develops behavioral descriptions using What, When, How questions. Avoids "Why"
questions (they invite mentalism). Moves from general to specific. Used with clients AND
significant others.
Checklists
Descriptions of specific behaviors and conditions under which each should occur. Often
Likert-scale. Typically combined with interview. Examples: CBCL, Vineland Adaptive
Behavior Scales.

✓ EXAMPLE — Direct: Analyst conducts 3 observations across settings, recording ABCs for
each instance of problem behavior. Most accurate picture.
✗ NON-EXAMPLE — Indirect as Sole Method: Analyst diagnoses behavior function from
parent interview alone without any direct observation. Insufficient — indirect methods provide
only opinions.

Operational Definitions

Operational Definition
A clear, concise, complete, and objective description of a behavior. Allows anyone trained
to observe and measure the behavior the same way. Must include onset and offset.
Good Op Def Criteria — COMIC
Clear: unambiguous. Objective: based on observable events, not inferences. Measurable:
can be counted, timed, or scaled. Includes onset + offset. Concise: readable.
Onset
The moment the behavior begins. Clearly specified to enable duration measurement and
ensure consistency across observers.
Offset
The moment the behavior ends. Often defined as the absence of all target behaviors for a
specified minimum duration.

Page 13 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ GOOD EXAMPLE: Tantrum is defined as any instance of crying (with or without tears),
screaming above conversational volume, or dropping to the floor for at least 10 seconds.
OFFSET = absence of all of the above for at least 30 consecutive seconds.
✓ POOR EXAMPLE: "Tantrum is anytime the client throws a fit." — vague, subjective, no
onset/offset, not replicable.
✓ GOOD EXAMPLE: Biting: pressing top and bottom teeth into the skin or clothing of another
person. ONSET = teeth make contact. OFFSET = jaw releases AND head moves at least 1
inch away.

★ EXAM TIP: Operational definitions must include BOTH onset and offset. A good op def
should include examples and non-examples. Common exam pitfall: "when the child is
angry" is NOT an operational definition.

Social Validity (Wolf, 1978)

Social Validity
The degree to which the goals, procedures, and outcomes of an intervention are
acceptable and meaningful to the client, family, and society.
Goal Validity
Are the targets for change genuinely important and socially appropriate for this person?
Procedure Validity
Are the methods used acceptable, ethical, and aligned with best practices?
Outcome Validity
Did the intervention produce meaningful, significant change that improves the person's
actual life?

✓ EXAMPLE: Parents, teachers, and the client agree that learning to request a break
appropriately is a meaningful goal that will reduce challenging behavior across settings.
✗ NON-EXAMPLE: Teaching a 16-year-old to sort colors because it is the next item on a
curriculum checklist, without regard to its relevance to the student's daily life.

★ EXAM TIP: Social validity is assessed at THREE points: goals, procedures, outcomes.
Wolf (1978) introduced this concept. Always ask: does this matter to the client's actual
life?

Page 14 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Measuring Behavior
Dimensions of Behavior, Measurement Types, Recording Methods
4

Why Measure?

Purpose of Measurement
Measurement operationalizes empiricism. Without it, science is guesswork. ABA
practitioners measure to: evaluate current behavior levels, guide decision-making, avoid
continuing ineffective treatment, avoid discontinuing effective treatment.
Benefits
Optimize effectiveness, verify legitimacy of treatments, identify pseudoscience, ensure
accountability, meet ethical standards.

3 Fundamental Dimensions of Behavior

Repeatability (Countability)
Behavior can occur more than once. Can be counted. The basis for frequency and rate
measures.
Temporal Extent (Duration)
Every instance of behavior occupies some amount of time. Measured from onset to offset.
Basis for duration and IRT measures.
Temporal Locus
When behavior occurs in time. Basis for latency and inter-response time measures.

Measures Based on Repeatability

Page 15 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Count / Frequency
The number of times a behavior occurs during an observation period. Absolute number.
Must always report the observation period length for context.
Rate
Count divided by the observation time unit. Responses per minute, per hour, etc. More
meaningful than count alone because it accounts for varying observation lengths.
Celeration
The rate of change in rate over time. Used on the Standard Celeration Chart. Measures
how fast behavior is increasing or decreasing.

✓ EXAMPLE — Rate vs Count: Client eloped twice Monday (3-hr session) and five times
Wednesday (3-hr session). Count differs (2 vs 5). Rate = 0.67/hr vs 1.67/hr. Rate allows fair
comparison.
✗ NON-EXAMPLE: "Client raised hand 5.5 times per day" — this is a RATE (per day), not a
count/frequency. Frequency = raw number.

Measures Based on Temporal Extent

Duration
The total elapsed time from onset to offset of a behavior. Used when how long matters
more than how many times. Example: time on-task, duration of tantrums.
Response Latency
The time from the onset of the SD (stimulus/instruction) to the start of the response.
Measures how quickly the organism begins responding. Example: time from "Come sit
down" to client beginning to move toward chair.
Interresponse Time (IRT)
The time between the OFFSET of one response and the ONSET of the next response of
the same class. NOT from an SD — from the end of the last behavior.

✓ DISTINCTION — Latency vs IRT vs Duration: Duration = how long one response lasts
(onset to offset). Latency = SD to start of behavior. IRT = end of one behavior to start of next
behavior.
✓ EXAMPLE — IRT: Client hits at 2:00pm, then again at 2:05pm. IRT = approximately 5
minutes between responses.
✗ NON-EXAMPLE: "Time from 2pm to 2:05pm tantrum" — if measured from a demand, this is
LATENCY. If measured from end of previous tantrum, this is IRT.

★ EXAM TIP: This triad is a common exam trap. Duration = start to end of ONE behavior.
Latency = SD to start. IRT = end of one to start of next.

Derivative Measures

Page 16 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Percentage
Number of occurrences divided by total opportunities × 100. Used when number of
opportunities varies across sessions. Makes comparison fair.
Trials to Criterion
Number of instructional trials required to reach a mastery criterion. A measure of learning
efficiency.

Direct Measurement Methods

Continuous Measurement
Every single occurrence of the behavior is measured. Most complete and accurate.
Includes: frequency/count, rate, duration, latency, IRT, percentage, trials to criterion.
Permanent Product Recording
Measuring the outcome or product of a behavior after it has occurred rather than during.
Example: counting ripped papers after a session. The behavior itself is not observed
directly.

✓ EXAMPLE — Permanent Product: Teacher counts number of math problems completed

correctly on a worksheet after class. Counts the PRODUCT, not the behavior of writing.
✗ NON-EXAMPLE: Teacher counts how many times student picks up pencil WHILE working.
This is direct, continuous event recording — not permanent product.

Discontinuous Measurement (Time Sampling)

Overview
Records behavior only SOME of the time based on predetermined intervals. More
practical for frequent or continuous behaviors. Three main types: PIR, WIR, MTS.
Partial Interval Recording (PIR)
Observation period divided into equal intervals. Record occurrence if behavior happens
AT ANY TIME during the interval — even for 1 second. OVERESTIMATES actual duration
of behavior.
Whole Interval Recording (WIR)
Record occurrence ONLY if behavior happens throughout the ENTIRE interval.
UNDERESTIMATES actual duration of behavior. Best for sustained, continuous behaviors
you want to increase.
Momentary Time Sampling (MTS)
Observer checks only at the EXACT MOMENT the interval ends — a single snapshot.
Records whether behavior is occurring at that precise instant. Most accurate estimate of
overall prevalence. Best for frequent, stable behaviors.

Page 17 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ PIR EXAMPLE: Client tantrums for 1 minute of a 5-minute interval. PIR records:
OCCURRENCE. (Even though behavior only occurred 20% of interval.)
✓ WIR EXAMPLE: Client tantrums for 4 of 5 minutes. WIR records: NON-OCCURRENCE.
(Did not occur for the FULL 5 minutes.)
✓ MTS EXAMPLE: Teacher glances at student every 5 minutes and records yes/no at that
exact moment. Does not matter what happened the rest of the interval.
✗ NON-EXAMPLE WIR as PIR: If observer marks occurrence whenever behavior happens for
any duration of the interval, they are doing PIR, not WIR.

★ EXAM TIP: PIR = Partial = Permissive = OVER-estimates. WIR = Whole = strict =

UNDER-estimates. MTS = snapshot at end only. Know which to use: PIR for behaviors you
want to decrease (shows higher than actual — conservative). WIR for behaviors you want
to increase.

Page 18 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Improving & Assessing Measurement Quality

Validity, Reliability, Accuracy, IOA, Observer Drift & Reactivity
5

Trustworthy Measurement

Trustworthy Measurement
Measurement that is simultaneously valid, accurate, AND reliable. All three properties
must be present for data to be trusted.
Validity (of measurement)
The measurement actually assesses what it intends to assess. The data are
representative of the behavior of interest.
Accuracy
Observed values match the true value of the behavior as it exists in nature. The data
correctly represent what actually happened.
Reliability
The measurement system produces consistent results across repeated measurement of
the same event. Inter-observer agreement (IOA) is the main tool for assessing reliability.

✓ VALID BUT NOT ACCURATE: Observer consistently misses some instances (always
undercounts). Data are consistent (reliable) but don't match true value (not accurate).
✓ ACCURATE BUT NOT RELIABLE: Observer counts 12 hits, then 10, then 11 watching
same video. Inconsistent — not reliable, even if one count was accurate.

★ EXAM TIP: Trustworthy = Valid + Accurate + Reliable. You need ALL THREE. A measure
can be reliable (consistent) but inaccurate (consistently wrong).

Threats to Measurement Quality

Page 19 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Observer Drift
A gradual, unconscious change in how an observer applies an operational definition over
time. Results in inaccurate, unreliable, inconsistent data. Prevented by: regular
recalibration training, ongoing IOA checks.
Observer Reactivity
The observer changes how they record data BECAUSE they are aware they are being
observed or evaluated. Can inflate OR deflate data. Prevented by: unannounced checks,
habituation to being observed.
Indirect Measurement Artifacts
Using a measure that does not directly capture the behavior of interest. Requires
inferences.

✓ OBSERVER DRIFT EXAMPLE: BT stops recording biting on clothing after weeks — only
records bites that leave marks. Data shows apparent decrease in biting, but this is observer
drift, not real change.
✓ REACTIVITY EXAMPLE: Supervisor arrives and BT begins recording more carefully —
data spikes during observed sessions. This is reactivity, not a real behavior change.
✗ NON-EXAMPLE — DRIFT: Supervisor changes op def and team all update recording. This
is an intentional definition change marked by a phase change line — NOT drift.

Interobserver Agreement (IOA)

Page 20 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Purpose of IOA
Assesses the reliability and accuracy of behavioral measurement. Two independent
observers record the same session; their agreement is calculated. Target: ≥ 80%. Collect
in ≥ 20% of sessions.
Total Count IOA
Formula: (Smaller count ÷ Larger count) × 100. Best for event recording. Example:
Observer 1 = 5 hits, Observer 2 = 6 hits. IOA = (5÷6) × 100 = 83.3%.
Mean Count-Per-Interval IOA
Divide observation into intervals. For each interval: (smaller ÷ larger) × 100. Then average
across all intervals. More precise than total count for varying distributions.
Exact Count-Per-Interval IOA
Intervals where both observers recorded the EXACT same count ÷ total intervals × 100.
More conservative than mean count.
Trial-by-Trial IOA
Agreement trials ÷ total trials × 100. Used in DTT settings.
Interval-by-Interval IOA
Agreements (both scored same for each interval) ÷ total intervals × 100. For
occurrence/non-occurrence data.
Scored-Interval IOA
Intervals where BOTH scored occurrence ÷ intervals where AT LEAST ONE scored
occurrence × 100. More conservative — focuses only on when behavior occurred.
Unscored-Interval IOA
Intervals where BOTH scored non-occurrence ÷ intervals where AT LEAST ONE scored
non-occurrence × 100. Focuses on non-occurring intervals.
Total Duration IOA
(Shorter total duration ÷ Longer total duration) × 100.
Mean Duration-Per-Occurrence IOA
For each occurrence: (shorter ÷ longer) × 100. Then average across occurrences.

✓ SCORED vs INTERVAL-BY-INTERVAL: Interval-by-Interval counts agreements when

BOTH score occurrence AND when BOTH score non-occurrence. Scored-Interval counts
ONLY when at least one observer scored an occurrence. Scored-Interval is more sensitive to
rare behaviors.

★ EXAM TIP: Know EACH IOA formula. Exam questions give you data and ask you to
calculate. Most tested: Total Count IOA, Trial-by-Trial IOA, Scored-Interval IOA. IOA ≥ 80%
= acceptable; < 80% = need to revise op def or retrain observers.

Page 21 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Constructing & Interpreting Graphic Displays

Types of Graphs, Visual Analysis, Line Graphs, Celeration Charts
6

Purpose of Graphic Display

Primary Function
Communication — displaying relationships between the dependent variable and
independent variable. Visual display of behavioral data over time.
Benefits of Graphic Display
Immediate access to complete behavioral record. Variations prompt exploration.
Facilitates accurate interpretation. No predetermined significance level. Conservative —
requires clear visual evidence. Encourages independent judgment. Effective feedback
tool.

Types of Graphs in ABA

Line Graph (Most Common)

Based on the Cartesian plane. X-axis = always some measure of TIME (sessions, days,
trials). Y-axis = always some measure of BEHAVIOR. Data points connected by lines
within each phase. Phase change lines separate conditions.
Bar Graph
Rectangular bars/columns represent data. Used for comparing discrete categories (e.g.,
behavior rates across settings, or pre/post comparisons). Does not show change over
time.
Cumulative Record
Never decreasing — each data point adds to the total. Slope represents rate of
responding. Flat line = no responding. Steep line = high rate.
Semilogarithmic Chart / Standard Celeration Chart (SCC)
The y-axis is logarithmic. Displays rate of behavior change (celeration) rather than
absolute levels. The SCC shows counts per minute. Used in Precision Teaching.
Scatterplot
Plots behavior frequency or occurrence against time of day (or another variable). Used to
identify temporal patterns and likely MOs. Common in FBA.

Page 22 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ EXAMPLE — Line Graph vs Bar Graph: Change in aggression across a 20-session ABAB
study = line graph (shows trend over time). Comparison of aggression rates across 4 settings
= bar graph (comparing categories).

Visual Analysis of Graphs — TLV

Trend
The overall direction of the data path WITHIN a phase. Ascending (increasing),
descending (decreasing), or level (no systematic direction). Evaluated across the data
within a condition, not based on single points.
Level
The absolute value — the mean or median of the data within a phase. Described as high,
moderate, or low relative to the overall scale. Changes in level between phases =
immediate treatment effect.
Variability
The degree of scatter in the data within a phase. High variability = data points widely
spread. Low variability = data stable. High variability makes it difficult to determine a trend
or confirm a treatment effect.
Stability
Data are stable when they show consistent trend, level, and variability within a phase.
Stable data in baseline = prerequisite before introducing treatment.

✓ EXAMPLE — Full Visual Analysis: Trend in baseline = increasing (countertherapeutic for a

reduction target). Level = high. Variability = low. After intervention: trend = decreasing. Level =
low. Variability = moderate. The contrast demonstrates treatment effect.
✗ NON-EXAMPLE: "It feels like the behavior is getting better." This is NOT visual analysis.
Must describe trend, level, variability from the actual data points.

★ EXAM TIP: TLV = Trend, Level, Variability. Every visual analysis question should
address all three WITHIN each phase, then compare ACROSS phases. A change in level =
immediate effect. A change in trend = gradual effect.

Phase Change Lines & Data Conventions

Page 23 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Phase Change Line

A vertical line drawn on the graph to separate different experimental conditions. All data to
the left = one condition. All data to the right = new condition.
Data Path
The line connecting data points within a single phase. Never draw a data path across a
phase change line.
Stable Baseline
Required before introducing intervention. Data must show consistent trend, level, and
variability. Allows confident prediction of what would happen without intervention.
Countertherapeutic Trend
Baseline trend moving in the WRONG direction (e.g., aggression increasing when target is
to reduce it). Cannot wait for stability ethically — begin intervention.

Page 24 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Analyzing Behavior Change: Basic Assumptions & Strategie

Experimental Control, Internal & External Validity, Baseline Logic
7

Experimental Control

Experimental Control
A predictable, reliable change in behavior (DV) that can be produced by systematic
manipulation of some aspect of the environment (IV). Demonstrated when behavior
changes ONLY when the IV is introduced.
Functional Relation
A predictable, reliable relationship between a behavior and an environmental variable.
The behavior changes consistently as the IV is manipulated. The goal of experimental
analysis in ABA.
Internal Validity
The extent to which the experiment demonstrates convincingly that observed changes in
behavior were caused by the IV and not by uncontrolled or unknown variables. High
internal validity = strong functional relation.
External Validity
The degree to which results can be generalized beyond the specific conditions of the
experiment — to other subjects, settings, behaviors, or time periods. Increased by
systematic replication.
Confounding Variable
A variable that is NOT the IV but could also be responsible for changes in the DV.
Threatens internal validity. Must be controlled.

✓ EXAMPLE — Functional Relation: A-B-A-B design: behavior increases during baseline,

decreases during intervention, increases again when intervention withdrawn, decreases again
when intervention reintroduced. Functional relation demonstrated.
✗ NON-EXAMPLE: Behavior decreases after intervention begins AND the client also starts
new medication. Confounding variable — cannot attribute change solely to the IV.

Baseline Logic

Page 25 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Baseline Logic
The experimental reasoning inherent in single-subject designs. Uses three components:
prediction, verification, replication.
Prediction
From stable baseline data, the analyst predicts what behavior would do in the ABSENCE
of intervention (if nothing changes). The baseline data path is projected forward.
Verification
Showing that the prediction about baseline behavior is accurate — confirming the pattern
holds. In reversal designs: returning to baseline verifies that the level returns.
Replication
Demonstrating the effect again (a second time) within the same study OR in a separate
study. Confirms functional relation. Increases confidence.
Steady State Responding
A behavior pattern showing minimal systematic variation in its measured quantities over
time. Relatively stable data. Required before changing conditions.

✓ EXAMPLE — Prediction + Verification + Replication in ABAB: A phase provides

prediction (baseline trend). B phase shows change. Return to A verifies that without IV,
behavior returns to baseline levels. Second B phase replicates the effect.

★ EXAM TIP: The three functions of baseline logic — prediction, verification, replication —
are tested directly. Know which phase does which in ABAB: A1=prediction, B1=change,
A2=verification, B2=replication.

Direct & Systematic Replication

Direct Replication
Repeating an experiment as closely as possible to the original — same IV, same DV,
similar participants. Increases reliability of findings within a study or across studies.
Systematic Replication
Repeating an experiment with planned, intentional variations: different participants,
settings, behaviors, or experimenters. Increases external validity and generality of
findings. Failures in systematic replication are as informative as successes.

Page 26 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Reversal & Alternating Treatments Designs

A-B-A-B, ATD, Multiple Treatment Reversal
8

Reversal Design (A-B-A-B)

A-B-A-B Design
The most powerful within-subject experimental design. Baseline (A) → Intervention (B) →
Return to Baseline (A) → Intervention (B). Demonstrates experimental control through
prediction, verification, and replication.
Why A-B-A-B over A-B-A
A-B-A ends on a withdrawn intervention — unethical if effective. A-B-A-B reintroduces
treatment, serves the client AND provides stronger experimental evidence (two
demonstrations of the effect).
Limitations
Cannot be used when: (1) behavior change is irreversible (skill acquisition); (2) withdrawal
is unethical or impractical; (3) target behavior might generalize to non-treatment
conditions.
B-A-B Design
Begins with treatment — appropriate for dangerous behaviors where withholding
treatment is unethical. Does not allow assessment of pre-intervention baseline.
Susceptible to sequence effects.

✓ EXAMPLE: Elopement reduced from 5x/session to 0 with visual schedule. Schedule

removed — elopement returns to 4x/session. Schedule reintroduced — elopement drops to 0
again. Functional relation demonstrated.
✗ NON-EXAMPLE: Teacher introduces reinforcement system but never returns to baseline.
Without reversal, cannot confirm the IV caused the change — could be maturation, placebo
effect, etc.

Variations of Reversal Design

Page 27 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Multiple Treatment Reversal Design

Compares effects of two or more experimental conditions against each other or against
baseline. Pattern: A-B-A-B-C-B-C or A-B-C-B-C-B-C. Vulnerable to sequence effects.
NCR Reversal Technique
Non-contingent reinforcement used as the reversal condition instead of a return to pure
baseline. Tests whether the contingency is responsible for behavior change.
DRO/DRI/DRA Reversal Technique
Uses a differential reinforcement procedure as the reversal condition.

Alternating Treatments Design (ATD)

ATD Definition
Two or more conditions (including possibly a no-treatment control) presented in rapidly
alternating succession — typically alternating sessions or days — independent of the level
of responding. Also called: concurrent schedule design, multielement design, multiple
schedule design.
Purpose
Compares the relative effects of two or more interventions. Shows which treatment
produces better outcomes. Can compare against baseline. Does not require baseline first.
Advantage
Rapid comparison of treatments. Useful when time is limited. Less susceptible to
sequence effects than reversal designs.
Multiple Treatment Interference
The effects of one treatment condition may carry over and influence responding in another
condition. A threat to internal validity in ATD.

✓ EXAMPLE: Token reinforcement condition vs verbal praise condition alternated on

successive sessions. Task completion consistently higher during token reinforcement days →
token reinforcement is the more effective intervention.
✗ NON-EXAMPLE: One treatment tried for 3 weeks then switched to another for 3 weeks. This
is NOT ATD — no rapid alternation. This is a simple phase change.

★ EXAM TIP: ATD = RAPID alternation (session by session or day by day). Multiple
Treatment Reversal = comparison within a reversal framework (A-B-C-B-C). Know the
difference.

Page 28 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Multiple Baseline & Changing Criterion Designs

Multiple Baseline Across Behaviors/Settings/Subjects, Changing Criterion
9

Multiple Baseline Design

Overview
The most widely used experimental design in ABA. Highly flexible. Does NOT require
withdrawal or reversal of treatment. Treatment is introduced at staggered (different) times
across two or more tiers.
When to Use
Alternative to reversal when: behavior change would be irreversible, withdrawal is
unethical or impractical, or target behaviors are unlikely to reverse.
Three Basic Forms
1. Multiple Baseline Across Behaviors — same subject, 2+ concurrent behaviors. 2.
Multiple Baseline Across Settings — same behavior, 2+ settings. 3. Multiple Baseline
Across Subjects — same behavior, same setting, 2+ subjects.
Experimental Logic
The IV is applied to the first tier. The remaining tiers continue in baseline. If behavior
changes ONLY in the tier receiving the IV AND remains stable in other tiers —
experimental control is demonstrated. Each introduction is a new replication.
Minimum Tiers
Most commonly 3–5 tiers. More tiers = more convincing demonstration. Minimum 3 tiers
typically required.

✓ EXAMPLE: Intervention applied to on-task behavior in math first. Math on-task increases.
Reading and writing remain at baseline. Intervention then applied to reading — reading
increases. Writing still stable. Finally applied to writing — writing increases. Each change
occurs only when IV introduced.
✗ NON-EXAMPLE: All three behaviors improve simultaneously after intervention applied to the
first. Cannot determine if the IV caused change or if some other variable (maturation, teacher
change) did.
✗ NON-EXAMPLE: Introducing IV to all tiers simultaneously. No staggered introduction = no
experimental control. Not a multiple baseline design.

Page 29 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

★ EXAM TIP: KEY RULE — behavior should change ONLY when the IV is introduced to
THAT tier. If other tiers change before the IV reaches them (concurrent change),
experimental control is compromised.

Delayed Multiple Baseline Design

Overview
A variation where the baseline for additional tiers is not started at the beginning of the
study — they are added later. Allows replication across additional subjects encountered
after the study has begun.

Changing Criterion Design

Overview
The criterion for reinforcement is systematically changed in a stepwise fashion following a
baseline phase. The independent variable is the change in criterion.
Experimental Logic
If behavior tracks the changing criterion — increasing (or decreasing) as the criterion
changes — experimental control is demonstrated. The more closely behavior tracks each
criterion change, the stronger the evidence.
When to Use
Ideal for shaping — gradually building a behavior toward a terminal goal through
successive approximations. Also used for gradually reducing a behavior.
Advantages
Does not require withdrawal of treatment. Ethically appropriate for behaviors where
intervention should not be stopped.

✓ EXAMPLE: Baseline: 2 minutes independent work. Criterion 1: 3 min — client meets it.
Criterion 2: 5 min — client meets it. Criterion 3: 8 min — client meets it. Each step tracked
precisely. Experimental control demonstrated.
✗ NON-EXAMPLE: Behavior reinforced only when it exceeds 5 minutes but criterion never
changes over time. Without stepwise changes, it is not a changing criterion design.

Page 30 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Planning & Evaluating ABA Research

Individual Subject Designs, Group Designs, Planning Considerations
10

Individual Subject Designs

Rationale
Each person serves as their own control. Within-subject comparison rather than
between-group. The individual's behavior under one condition is compared to their own
behavior under another condition.
Advantages over Group Designs
Group data masks individual variability. A treatment that helps some and hurts others can
show average improvement. Individual designs reveal effects on each person. More
applicable to clinical practice.
Intrasubject Replication
Replicating the effect within the same individual (e.g., ABAB). Fundamental to
demonstrating functional relations in single-subject research.

✓ EXAMPLE: A group study shows intervention improved reading by 10% on average. But 3
students improved 30% while 2 students got worse. Group average hides this critical
information.

Concerns in Research Planning

Page 31 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Internal Validity
Controlled by: holding extraneous variables constant, randomization (in group designs), or
manipulating them as IVs. Single-subject designs use baseline logic and replication.
External Validity
Increased through systematic replication — same effects demonstrated across different
subjects, settings, behaviors, therapists, and time periods.
Social Validity
Assessed for goals, procedures, and outcomes. Are the changes meaningful to the
individual?
Flexibility in Design
Good design is any IV manipulation that produces convincing evidence of a functional
relation AND serves the client. Designs are adapted to the situation — there is no single
"right" design.

Page 32 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Positive Reinforcement
Definitions, Types, Schedules, Stimulus Preference Assessments
11

Definition & Core Concepts

Positive Reinforcement
A stimulus change FOLLOWS a response (stimulus is ADDED or increased in intensity)
AND the FUTURE FREQUENCY of that response INCREASES in similar conditions. Both
elements required for definition.
Reinforcer
A stimulus, object, or event that INCREASES the frequency of a behavior when presented
contingently. Defined by its EFFECT, not by its nature. Something is only a reinforcer if it
actually increases behavior.
Circular Reasoning Trap
NOT circular: "Robbie's studying increased when he earned points → points function as a
reinforcer." IS circular: "Johnny can't read → he has a learning disability → how do we
know? → because he can't read." Reinforcement is not circular because cause (points)
and effect (studying) are separable and measurable.
Automaticity
Behavior is modified by its consequences regardless of whether the person is aware of
the contingency. Reinforcement can work without conscious understanding.

✓ EXAMPLE: A parent gives a child candy every time the child says "please." Over time, the
child says "please" more often. Candy is a positive reinforcer.
✗ NON-EXAMPLE: A parent gives a child candy every time the child cries. The child cries
more. While candy is added (positive), this is still reinforcement — but of crying, not of a
desired behavior.
✗ NON-EXAMPLE — Not reinforcement: A parent gives a child stickers for completing
homework. Despite the stickers, homework completion doesn't increase. Stickers are NOT
functioning as a reinforcer for this behavior with this child.

★ EXAM TIP: Three requirements for positive reinforcement: (1) stimulus ADDED, (2)
CONTINGENT on behavior, (3) FUTURE behavior INCREASES. If behavior does not
increase, reinforcement has not occurred — regardless of intent.

Types of Reinforcers

Page 33 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Unconditioned (Primary) Reinforcers

Stimuli that are reinforcing without any prior learning. Biologically significant. Examples:
food, water, warmth, sleep, sensory stimulation, sexual contact. Value does not depend
on learning history.
Conditioned (Secondary) Reinforcers
Stimuli that have acquired reinforcing properties through pairing with other reinforcers.
Learned. Examples: praise, money, tokens, grades, approval. Value depends on learning
history.
Generalized Conditioned Reinforcers
Conditioned reinforcers paired with MANY other reinforcers. Not dependent on any single
motivating operation. Examples: money, tokens, praise (after extensive history). Maintains
value across a wide range of conditions.

✓ PRIMARY EXAMPLE: Food: child works for crackers during session. Biologically necessary
— reinforcing without training.
✓ CONDITIONED EXAMPLE: Teacher says "great job!" — this becomes reinforcing only after
pairing with preferred activities, smiles, and tangible rewards over time.
✓ GENERALIZED EXAMPLE: Tokens in a token economy. Initially neutral; paired with many
backup reinforcers. Eventually the token itself has reinforcing value.

Principles of Reinforcement

Page 34 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Rate of Reinforcement
How often reinforcement is delivered. Higher rates of reinforcement → stronger
maintenance of behavior. Rate must be sufficient to maintain the desired behavior.
Magnitude
The size, amount, or intensity of the reinforcer. Larger magnitudes generally more
effective, but subject to satiation.
Value (Current)
How reinforcing a stimulus is at this moment — influenced by motivating operations
(satiation/deprivation).
Contingency
Reinforcement must be contingent on (dependent on) the target behavior. Non-contingent
delivery weakens the behavior-reinforcer relationship.
Immediacy
Reinforcement delivered immediately after the behavior is more effective than delayed
reinforcement. Delays weaken the contingency.
Deprivation
Not having had access to a reinforcer increases its value (EO). The client should not have
unlimited free access to reinforcers used during sessions.
Satiation
Having had too much of a reinforcer decreases its value (AO). Vary reinforcers to prevent
satiation.

✓ IMMEDIACY EXAMPLE: Token given immediately when client says "break" = effective.
Token given 5 minutes later = less effective — client may not associate it with the behavior.

Stimulus Preference Assessments

Page 35 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Purpose
Identify potential reinforcers BEFORE using them. Preference ≠ reinforcement — must
test empirically. Four main types: Free Operant, Forced/Paired Choice, MSW, MSWO.
Free Operant
Items placed in room; client has free access for observation period. Time engaged with
each item recorded. Hierarchy by time spent. Less accurate but no challenging behavior
risk.
Forced/Paired Choice
Two items at a time; client chooses one; rotation until all pairs presented. Hierarchy by
most chosen. Clear hierarchy; side bias risk.
MSW (Multiple Stimulus With Replacement)
Array of 3+ items; client selects one; chosen item REPLACED for next trial. Less accurate
than MSWO.
MSWO (Multiple Stimulus Without Replacement)
Array of 3+ items; client selects one; chosen item REMOVED (not replaced). Most
accurate hierarchy. Not for clients who react aggressively to removal.

✓ CHOICE EXAMPLE: Client engages with puzzle for 2 min, ball for 5 min, video for 8 min.
Free operant hierarchy: video > ball > puzzle.

★ EXAM TIP: Preference assessment ≠ Reinforcer assessment. Preference tells you what
is LIKED. Reinforcer assessment tests whether the preferred item actually INCREASES
behavior when used as a consequence.

Page 36 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Negative Reinforcement
Escape & Avoidance, EO, Comparison with Punishment
12

Definition

Negative Reinforcement
A stimulus change FOLLOWS a response (an aversive or unwanted stimulus is
REMOVED, reduced, or postponed) AND the FUTURE FREQUENCY of that response
INCREASES.
Escape Contingency
Responding TERMINATES an ongoing aversive stimulus. Four-term: EO (aversive
present) → SD → response → removal of aversive. Example: turning off a loud alarm by
pressing snooze.
Avoidance Contingency
Responding PREVENTS or POSTPONES the onset of an aversive stimulus. The aversive
never fully occurs. Example: studying to avoid failing an exam.

✓ ESCAPE EXAMPLE: Child completes demands to terminate the teacher's nagging.

Nagging removed = negative reinforcement of compliance.
✓ AVOIDANCE EXAMPLE: Employee finishes project before deadline to avoid manager's
disappointment (which has never occurred because they always meet deadlines). The
aversive is prevented.
✗ NON-EXAMPLE: A child stops hitting when a teacher says "stop" — the teacher's command
is an SD, not a reinforcement procedure.

★ EXAM TIP: Negative reinforcement is NOT punishment. Both involve aversive events.
Key difference: Negative reinforcement = aversive REMOVED → behavior INCREASES.
Punishment = behavior DECREASES. The + and − refer to adding/removing, not
pleasant/unpleasant.

Comparison: Positive vs Negative Reinforcement

Page 37 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

How They Are Similar

Both produce an INCREASE in responding. Both involve a stimulus change following
behavior. Both require contingency and affect future behavior.
How They Differ
Positive: stimulus ADDED. Negative: stimulus REMOVED. The type of stimulus change —
not the pleasantness — defines the category.
Michael's Suggestion (1975)
The distinction between positive and negative reinforcement is often unnecessary and
potentially misleading. Better to identify the key stimulus features before and after the
behavior. More complete functional understanding.

✓ AMBIGUOUS EXAMPLE: Client completes work → gets free time. Is this positive (free time
added) or negative (demand removed)? Both are true — Michael argues the distinction is not
always clinically meaningful.

Page 38 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Schedules of Reinforcement
CRF, FR, VR, FI, VI, Compound Schedules, Ratio Strain
13

Introduction

Schedule of Reinforcement
A rule that specifies which occurrences of a response (if any) will be followed by a
reinforcer. The schedule governs the temporal and numeric relationship between behavior
and reinforcement.
CRF — Continuous Reinforcement
Every response is reinforced. Produces fastest acquisition. Produces fastest extinction
when discontinued. Most sensitive to satiation.
INT — Intermittent Reinforcement
Only some responses are reinforced. Necessary for progression to natural reinforcement
contingencies. Produces greater resistance to extinction than CRF. Foundation for all
maintenance.

Ratio Schedules

Fixed Ratio (FR)

A set number of responses must occur before reinforcement is delivered.
Post-reinforcement pause (PRP): the organism stops responding briefly after each
reinforcement. Large FR requirements → long PRPs. Generally produces high rates of
responding.
Variable Ratio (VR)
An average number of responses required; varies unpredictably around that average.
Produces the highest, most consistent rates of responding. Most resistant to extinction of
all schedules. No predictable PRP.

✓ FR EXAMPLE: FR5: every 5th correct response earns a token. Long PRPs visible after
each token delivery.
✓ VR EXAMPLE: Slot machines: VR schedule. Variable — sometimes wins after 1 pull,
sometimes after 50. High, steady rate. Extremely resistant to extinction — the hallmark of
gambling behavior.

Page 39 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Interval Schedules

Fixed Interval (FI)

First response AFTER a set period of time is reinforced. Produces a scallop pattern: low
responding at start of interval, accelerating responding as interval ends. PRP after each
reinforcement.
Variable Interval (VI)
First response after an average time interval is reinforced. Average varies unpredictably.
Produces steady, moderate rates of responding. More resistant to extinction than FI.

✓ FI EXAMPLE: FI 15-minute schedule: first correct response after 15 minutes earns

reinforcement. Produces scallop — little responding at minute 1, rapid responding at minute
14.
✓ VI EXAMPLE: Checking email: sometimes new email arrives after 2 minutes, sometimes 20
minutes. VI schedule → steady checking behavior.

★ EXAM TIP: Resistance to extinction order (most to least): VR > VI > FR > FI > CRF.
Variable schedules are more resistant than fixed. Ratio schedules produce higher rates
than interval schedules.

Compound & Complex Schedules

Page 40 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Concurrent Schedule (conc)

Two or more schedules available simultaneously for two or more behaviors. The organism
can choose which to respond to. Basis of matching law.
Multiple Schedule (mult)
Two or more schedules alternate, each correlated with a distinct SD. Only one component
in effect at a time.
Mixed Schedule (mix)
Two or more schedules alternate without distinctive SDs.
Tandem Schedule
Two or more schedules must be completed in sequence before reinforcement is delivered,
without distinctive SDs.
Chained Schedule
Two or more schedules completed in sequence, each correlated with a distinctive SD.
Each completed component produces the SD for the next.
Alternative Schedule
Reinforcement when either a ratio OR interval requirement is met first — whichever
comes first.
Progressive Schedule
The response requirement systematically increases over time or across responses.
Lag Reinforcement Schedule
Requires that the response differ from a specified number of previous responses. Used to
increase behavioral variability.

Ratio Strain

Ratio Strain
The breakdown of responding when a ratio requirement is increased too rapidly. Behavior
becomes irregular, pausing more, and may cease entirely. Prevented by: thinning
schedules gradually, monitoring data closely.

✓ EXAMPLE: Client working on FR1 → suddenly placed on FR20 with no intermediate steps.
Client stops responding, engages in escape behavior. Ratio strain.
✓ PREVENTION EXAMPLE: Gradually move from FR1 → FR3 → FR5 → FR8 → FR12 →
FR20 with data confirming stability at each step.

Page 41 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Punishment by Stimulus Presentation

Positive Punishment, Type I, Considerations & Limitations
14

Definition & Core Concepts

Punishment (operational definition)

Punishment has occurred when a response is followed IMMEDIATELY by a stimulus
change that DECREASES the FUTURE FREQUENCY of similar responses. Defined by
its EFFECT on future behavior — not by its nature.
Positive Punishment (Type I)
A stimulus is ADDED (or increased in intensity) following a behavior, resulting in a
DECREASE in the future frequency of that behavior. Also called punishment by stimulus
presentation.
Common Misuse of Term
Punishment does NOT mean: hurting someone, being mean, or acting punitively. A
decrease in future behavior MUST be observed before a consequence qualifies as
punishment.

✓ EXAMPLE: Every time a child runs in the hallway, the teacher makes them walk back to the
start and walk slowly. The child runs less over time. Positive punishment — an aversive
response requirement was added, and running decreased.
✗ NON-EXAMPLE: Teacher reprimands a child for calling out. The child continues calling out.
The reprimand is NOT functioning as punishment — behavior did not decrease.

★ EXAM TIP: If behavior does NOT decrease, it is NOT punishment — regardless of intent.
Always check: did future behavior actually decrease?

Types of Positive Punishment

Page 42 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Reprimands
A verbal statement expressing disapproval contingent on problem behavior. Effective
when used sparingly and delivered immediately. Overuse reduces effectiveness.
Overcorrection
After engaging in problem behavior, client must: (1) restore the environment to its original
or better state (Restitutional Overcorrection), AND/OR (2) repeatedly practice the correct
behavior (Positive Practice). Consequence must be logically related to the problem
behavior.
Positive Practice Overcorrection
Client repeatedly practices the correct or incompatible behavior contingent on engaging in
the problem behavior. Example: child runs in hallway → must walk back and walk slowly 5
times.
Contingent Exercise
Client performs a specific physical exercise contingent on the problem behavior. Less
commonly used. Must be used cautiously.
Response Blocking
Physically preventing the completion of a response. Does NOT withhold reinforcement —
only prevents the motor response. Often combined with other procedures.

✓ OVERCORRECTION EXAMPLE: Child throws food → must clean up the food AND clean
all tables in the cafeteria (restitutional) + practice proper eating 10 times (positive practice).
✗ NON-EXAMPLE: Child spits → put in timeout. This is NOT overcorrection — consequence is
not logically related to the behavior.

Ethical Considerations for Punishment

Least Restrictive Alternative

Punishment should only be used after less restrictive procedures (reinforcement-based)
have been tried and documented as ineffective. Ethical hierarchy: reinforce →
differentially reinforce → extinction → mild punishment → more aversive punishment.
Side Effects
Punishment can produce: aggression, escape/avoidance (client avoids the punishing
agent), emotional responses, suppression of desirable behavior, and modeling of
aggressive behavior.
Documentation Required
All punishment procedures require: thorough documentation, informed consent from
guardian/client, review by a treatment team, regular data review, a concurrent
reinforcement component.

Page 43 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Punishment by Removal of a Stimulus

Negative Punishment, Time-Out, Response Cost
15

Definition

Negative Punishment (Type II)

A stimulus is REMOVED following a behavior, resulting in a DECREASE in the future
frequency of that behavior. Also called punishment by removal.
Two Main Types
Time-out from positive reinforcement (removal from reinforcing environment) and
Response cost (removal of a specified amount of a reinforcer already possessed).

✓ EXAMPLE: Child hits sibling → TV turned off for 5 minutes. TV (preferred activity) removed.
If hitting decreases over time, TV removal was negative punishment.
✗ NON-EXAMPLE: Child loses tokens but tokens weren't earned yet. Not response cost —
nothing was removed (it was never given). Response cost requires removing something
already in possession.

Time-Out from Positive Reinforcement

Page 44 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
The withdrawal of the opportunity to earn positive reinforcement, or the loss of access to
positive reinforcers, contingent on the occurrence of a target behavior, resulting in a
decrease in future frequency.
Critical Requirement: Time-In Must Be Reinforcing
The environment from which the client is removed MUST be reinforcing. If time-in is
aversive (difficult demands, disliked setting), time-out will NOT function as punishment —
it may actually reinforce escape.
Exclusionary Time-Out
Client is physically removed from the reinforcing environment. Example: sent to a
separate room. Most restrictive form.
Non-Exclusionary (Inclusionary) Time-Out
Client remains in the setting but cannot access reinforcement. Examples: contingent
observation (watches others from periphery), time-out ribbon (ribbon removed = no
reinforcement available), partition time-out.
Contingent Observation
Client moved to edge of activity area to observe others receiving reinforcement. Less
restrictive than exclusionary. Used in group settings.
Time-Out Ribbon
Client wears a ribbon/bracelet — when worn, reinforcement available. When removed
contingent on problem behavior, no reinforcement available. Can be used in natural
settings.

✓ EXAMPLE: Child hits at a birthday party → sent to room while party continues. Exclusionary
time-out. Effective if the party was reinforcing.
✗ NON-EXAMPLE: Child is in a difficult class they dislike. Sent to hallway as time-out. Time-in
was NOT reinforcing — time-out may reinforce escape rather than punish the behavior.

★ EXAM TIP: Time-out ONLY works if time-in is reinforcing. If a child prefers being in the
hallway, time-out in the hallway is negative reinforcement (escapes demands), not
punishment.

Response Cost

Page 45 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
The removal of a specified amount of a reinforcer (typically tokens, money, points,
privileges) contingent on the occurrence of a target behavior. The reinforcer must be in the
individual's possession before it can be removed.
Token Economy with Response Cost
Tokens earned for desired behavior; tokens removed for problem behavior. Combined
system. Effective when time-in (token-earning environment) is reinforcing.
Advantages
Efficient, easy to implement, no physical contact, can be combined with positive
reinforcement. Less stigmatizing than exclusionary time-out.

✓ EXAMPLE: Client has 10 tokens. Elopes from work area → loses 2 tokens. If elopement
decreases, response cost was effective.
✗ NON-EXAMPLE TRAP: "If Lucy doesn't earn 5 tokens, she can't access iPad time." This is
NOT response cost — nothing is being TAKEN AWAY. This is simply a reinforcement criterion
not met.

Page 46 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Motivating Operations
EOs, AOs, Unconditioned & Conditioned MOs, CMO-R/T/S
16

Definition & Core Concepts

Motivating Operation (MO)

An environmental variable that (1) alters the current REINFORCING or PUNISHING
EFFECTIVENESS of some stimulus, object, or event (value-altering effect), AND (2)
alters the current FREQUENCY of all behavior that has been reinforced by that stimulus
(behavior-altering effect). BOTH effects occur together.
Value-Altering Effect
Changes how effective a stimulus is as a reinforcer or punisher. EO = increases
reinforcing effectiveness (makes reinforcer more valuable). AO = decreases reinforcing
effectiveness (makes reinforcer less valuable).
Behavior-Altering Effect
EO → Evocative effect: INCREASES current frequency of behavior reinforced by that
stimulus. AO → Abative effect: DECREASES current frequency of behavior reinforced by
that stimulus.
Establishing Operation (EO)
An MO that INCREASES the reinforcing effectiveness of a stimulus AND has an evocative
effect on behaviors reinforced by that stimulus.
Abolishing Operation (AO)
An MO that DECREASES the reinforcing effectiveness of a stimulus AND has an abative
effect on behaviors reinforced by that stimulus.

✓ EO EXAMPLE: Food deprivation (hunger): increases the reinforcing value of food AND
increases food-seeking behavior. Heat on a marathon day increases value of water AND
increases drinking behavior.
✓ AO EXAMPLE: Eating a full meal: decreases the reinforcing value of food AND decreases
food-seeking behavior. Cold temperature as AO: decreases value of avoiding warmth.
✗ NON-EXAMPLE: Teacher standing at front of class with marker in hand → students raise
hands. The teacher with marker is an SD (signals reinforcement available), not an MO — it
doesn't change the VALUE of reinforcement.

Page 47 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

★ EXAM TIP: MOs alter VALUE of reinforcers. SDs signal AVAILABILITY of reinforcement.
MOs are contingency-INDEPENDENT (don't depend on consequences). SDs are
contingency-DEPENDENT. Know this distinction — it is heavily tested.

Unconditioned MOs (UMOs)

UMOs
Motivating operations whose value-altering effects are unlearned — biologically
determined. Do not depend on prior learning.
Examples of UMOs
Food deprivation/satiation. Water deprivation/satiation. Temperature (hot/cold). Pain.
Sleep deprivation. Oxygen deprivation. Activity deprivation.

Conditioned MOs (CMOs)

CMOs
Motivating operations whose value-altering effects depend on prior learning. Three types:
CMO-R, CMO-T, CMO-S.
CMO-R (Reflexive)
A stimulus that has reliably preceded a WORSENING of conditions. Signals that
something aversive is coming. Increases the value of escape/avoidance behavior.
Example: teacher approaching with worksheet signals demand — increases escape
value.
CMO-T (Transitive)
A stimulus that MAKES SOMETHING ELSE needed to access reinforcement. Makes
another stimulus necessary as an instrument. Example: locked cabinet increases the
value of the key that opens it.
CMO-S (Surrogate)
A stimulus that has been paired with an unconditioned or conditioned MO and takes on
the same motivating effect. Example: fireplace paired with cold → fireplace alone
increases value of blanket even when not cold.

✓ CMO-R EXAMPLE: Child sees teacher pick up a worksheet (CMO-R) → escape motivation
increases → child elopes. The worksheet signals aversive demand coming.
✓ CMO-T EXAMPLE: Locked cabinet (CMO-T) → increases the value of the key →
key-seeking behavior increases.
✓ CMO-S EXAMPLE: Fireplace (paired with being cold) becomes a CMO-S → seeing the
fireplace increases the child's desire for a blanket even in a warm room.
✗ NON-EXAMPLE CMO-T: Key is valuable because the child likes keys. This is just a
reinforcer — not CMO-T. CMO-T requires that the key be valuable BECAUSE it provides
access to something else.

Page 48 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Page 49 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Stimulus Control
Discriminative Stimuli, Generalization, Discrimination Training
17

Definition

Stimulus Control
A behavior is said to be under stimulus control when the rate, latency, duration, or
amplitude of a response is altered by the PRESENCE of an antecedent stimulus as a
result of a history of differential reinforcement in the presence of that stimulus.
SD (Discriminative Stimulus)
A stimulus that has a history of signaling that a particular behavior has been or will be
reinforced. Its presence EVOKES the operant behavior. Not a trigger — an evocator. Prior
learning history required.
S-Delta (S∆)
A stimulus correlated with the non-availability of reinforcement. Behavior occurs at
LOWER rates in its presence. Learned through a history of non-reinforcement in its
presence.
SD-p (Discriminative Stimulus for Punishment)
A stimulus that signals that punishment is available contingent on a behavior. Its presence
suppresses the behavior.

✓ SD EXAMPLE: Dog runs to the gate when the mailman arrives — because past mailman
arrivals have been followed by treats. Mailman = SD for gate-running behavior.
✓ S∆ EXAMPLE: "Do Not Disturb" sign on professor's door — students don't knock because
past knocking with sign present led to no response (non-reinforcement). Sign = S∆.
✓ SD-p EXAMPLE: "Do Not Disturb" sign AND professor shouts "-10 points!" when disturbed
with sign on door. Sign = SD-p for knocking behavior. Suppresses knocking.
✗ NON-EXAMPLE SD: A child says "dog" when seeing cats, birds, rabbits, and many other
animals. Seeing a dog is NOT an SD for "dog" behavior — the behavior has not come under
stimulus control. Too broad.

★ EXAM TIP: SD ≠ CS. SD evokes operant behavior (history of reinforcement). CS elicits

respondent behavior (paired with UCS). SD ≠ MO. SD signals availability of reinforcement.
MO changes the VALUE of reinforcement.

Stimulus Generalization

Page 50 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Stimulus Generalization
The occurrence of a behavior in the presence of stimuli that are similar to (but not identical
to) the SD, without specific training in those conditions.
Stimulus Generalization Gradient
A systematic decrease in response strength as stimuli become less similar to the training
SD. Quantifies the spread of stimulus control.

✓ EXAMPLE: Child trained to say "dog" to a golden retriever photo also says "dog" to photos
of other dog breeds, a dog on TV, and a dog at the park. Stimulus generalization across
instances.

Stimulus Discrimination Training

Stimulus Discrimination Training

A procedure that produces different responses under different stimulus conditions.
Reinforcement delivered in the presence of SD; reinforcement withheld in the presence of
S∆.
Result
The behavior comes under tight stimulus control — occurring in the presence of the SD
and not in the presence of the S∆.

✓ EXAMPLE: Green light (SD) = tokens available for appropriate behavior. Red light (S∆) = no
tokens available. Over time, appropriate behavior increases in presence of green light,
decreases with red light.

Transfer of Stimulus Control

Transfer of Stimulus Control

Shifting stimulus control from one stimulus (often a prompt) to another stimulus (the
natural SD). The goal of prompt fading.
Method
Gradually fading the prompt while simultaneously strengthening the natural SD through
differential reinforcement.
Stimulus Fading
Systematically altering a physical dimension (size, color, position) of a discriminative
stimulus to shift stimulus control.

✓ EXAMPLE: Therapist prompts waving by waving first (prompt = therapist wave). Over time,
therapist wave faded out. Now client waves when they hear "hello" from any person. Control
transferred from prompt to natural greeting.

Page 51 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Page 52 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Imitation
Defining Imitation, Formal Similarity, Imitation Training
18

Definition

Imitation
A response that: (1) immediately follows a MODEL stimulus, (2) has FORMAL
SIMILARITY to the model, and (3) the model functions as a CONTROLLING VARIABLE
(SD) for the imitative response.
Formal Similarity
The model stimulus and the imitative response: (a) belong to the same sense modality,
AND (b) physically resemble each other. Example: vocal model → vocal response (both
involve sound, and both produce similar sound patterns).
Immediacy
The imitative response must occur very soon after the model. Delayed imitation
(hours/days later, outside the model context) is not controlled by the model as an SD —
different discriminative features are controlling.
Controlled Relation
The model must be demonstrated to be the controlling variable. Best evidenced when a
novel model still evokes an imitative response.

✓ EXAMPLE: Therapist claps hands (model) and says "do this" → client immediately claps
hands (response). Formal similarity: both motor, both hand-clapping. Therapist clapping is the
SD.
✗ NON-EXAMPLE — Not Formal Similarity: Therapist says "ball" (vocal) → client points to
ball (motor). Different sense modalities. Not formal similarity. This is a tact under instruction —
not imitation.
✗ NON-EXAMPLE — Not Imitation: Child sees older sibling ride a bike, then two days later
attempts to ride. This is observational learning/delayed imitation, but the model is not the
immediate SD — different context controls the behavior.

Types of Models

Page 53 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Planned Models
Pre-arranged antecedent stimuli used deliberately to teach new skills. Therapist
demonstrates exactly what to do. Controlled. Used in imitation training.
Unplanned Models
Occur in everyday social interactions — at school, home, play. Often produce new
adaptive behaviors through naturalistic observation. Basis of much incidental learning.

Imitation Training

Purpose
Build a generalized imitation repertoire — the ability to imitate novel models. Foundational
for language learning, social skills, and academic skills. Children with ASD often lack or
have limited imitation.
Procedure
Present model SD → prompt imitative response (if needed) → reinforce correct imitation
→ fade prompts over trials → test with novel, untrained models.
Generalized Imitation
After training with multiple models, the client begins imitating novel models that were
never specifically trained. This generalization is a hallmark of genuine imitation repertoire.

Page 54 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Shaping
Successive Approximations, Differential Reinforcement, Terminal Behavior
19

Definition

Shaping
A process of systematically and differentially reinforcing SUCCESSIVE
APPROXIMATIONS to a terminal (target) behavior. Used to develop new behaviors that
are not in the individual's current repertoire.
Terminal Behavior
The final behavior goal. The destination of the shaping process.
Successive Approximations
Intermediate behaviors that are members of the same response class or precursors to the
terminal behavior. Each step is reinforced in turn.
Differential Reinforcement in Shaping
Some members of a response class are reinforced (those that are closer to the terminal
behavior). Other members (not closer) are placed on extinction. This combination drives
behavior toward the terminal.

✓ EXAMPLE: Teaching "ball": (1) Reinforce any vocalization → (2) Reinforce "ba" sound →
(3) Reinforce "ba-" → (4) Reinforce "bal" → (5) Reinforce "ball." Each step, closer
approximations are reinforced; previous steps are extinguished.
✗ NON-EXAMPLE: Therapist says "Try again, say ball" every time client says "ba." This is
ECHOIC training, not shaping — the correct response is simply modeled and prompted, not
built through successive approximations.

★ EXAM TIP: Shaping requires BOTH (1) reinforcing closer approximations AND (2)
extinguishing less close approximations. Without extinction of previous steps, shaping
stalls.

Dimensions That Can Be Shaped

Page 55 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Topography
The physical form or shape of the behavior can be shaped (e.g., from babbling to words to
sentences).
Frequency
Number of responses per time period can be shaped upward or downward.
Duration
How long the behavior lasts can be shaped (e.g., increasing time on-task).
Latency
Time between SD and response can be shaped shorter.
Amplitude/Force
Intensity or force of behavior can be shaped (e.g., louder vocal response, softer touch).

Guidelines for Effective Shaping

Starting Point
Begin with a response the client already emits. Cannot shape a behavior that never
occurs.
Step Size
Steps should be small enough that each is achievable but large enough to make progress.
Steps too large → client fails and loses reinforcement. Too small → progress is
unnecessarily slow.
Movement Rule
Move to the next step ONLY when the current step is stable (occurring at a consistent
rate).
Regression Rule
If progress stalls, return to the previous step. Rebuild momentum before attempting the
difficult step again.

Page 56 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Chaining
Behavior Chains, Task Analysis, Forward/Backward/Total Task Chaining
20

Behavior Chains

Behavior Chain
A specific sequence of responses in which each response produces a stimulus change
that: (a) functions as conditioned reinforcement for the response that produced it, AND (b)
functions as the SD for the next response in the chain.
Dual Function of Each Link
Each response in the chain (except first and last): (1) a conditioned reinforcer for the
PREVIOUS response, and (2) an SD for the NEXT response.
Terminal Reinforcer
The reinforcer at the END of the chain. This is the unconditioned or primary reinforcer that
drives the entire chain.
Behavior Chain with Limited Hold
A contingency requiring the chain to be completed WITHIN a specified time limit to
produce reinforcement. Promotes both accuracy and fluency.

✓ EXAMPLE: Making a sandwich: Get bread (response 1) → bread on plate (SR+ and SD for
next) → get meat (response 2) → etc. → eat sandwich (terminal reinforcer).

Task Analysis

Task Analysis
Breaking a complex skill or chain of behaviors into smaller, discrete, teachable steps.
Each step is one link in the chain.
How to Create
(1) Identify the terminal behavior. (2) Identify all component steps. (3) Determine the
order. (4) Write each step as an observable, measurable response. (5) Create a data
sheet.

Page 57 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ EXAMPLE — Handwashing TA: 1. Turn on water. 2. Wet hands. 3. Apply soap. 4. Scrub
20 seconds. 5. Rinse soap off. 6. Turn off water. 7. Dry hands.
✗ NON-EXAMPLE: "Wash hands with soap and water. Dry hands." — too vague. No discrete
steps specified. Cannot measure or teach individual components.

Chaining Methods

Forward Chaining
Teaching begins with the FIRST step in the chain. Trainer completes all remaining steps.
Client receives reinforcement at the end (via trainer completion). Once first step is
mastered, client learns second step, and so on.
Backward Chaining
Teaching begins with the LAST step. Trainer completes all PREVIOUS steps. Client
performs only the last step. Reinforcement is earned immediately after performing the last
step. Once last step mastered, client learns second-to-last, etc.
Total Task Chaining (Total Task Presentation)
Client attempts ALL steps in the chain on every trial, receiving prompts as needed at each
step. Reinforcement at end of complete chain. Used when all steps are in repertoire but
need sequencing.

✓ BACKWARD ADVANTAGE: Client always finishes the chain → always contacts the
terminal reinforcer. High motivation maintained throughout training. Good for complex chains.
✓ FORWARD ADVANTAGE: More natural — matches real-world sequence. Good when first
steps are already in repertoire.
✗ NON-EXAMPLE — Not Chaining: Teaching a single behavior (high-five) is NOT chaining
because there are no sequential steps linking separate behaviors.

★ EXAM TIP: Know the order of teaching for each method. Forward = 1st step first.
Backward = last step first. Total task = all steps every trial.

Page 58 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Extinction
Procedural & Functional Extinction, Extinction Burst, Recovery, Renewal, Resurge
21

Definition

Extinction
A procedure in which reinforcement of a previously reinforced behavior is
DISCONTINUED; as a result, occurrences of that behavior DECREASE in the future. Zero
probability of reinforcement for the target behavior.
Effectiveness Depends On
(1) Correctly identifying and consistently withholding the MAINTAINING reinforcer. (2)
Consistent application across ALL people and contexts.
Procedural vs Functional Extinction
Procedural: ignoring the behavior (only works if the function is attention). Functional:
withholding the ACTUAL maintaining reinforcer regardless of topography. Functional
extinction is necessary for all functions.

✓ EXAMPLE — Escape Function: Child tantrums to escape homework. Extinction = not

allowing escape from homework when tantrum occurs. NOT ignoring. Escape-maintained
behaviors must continue to receive the demand during extinction.
✗ NON-EXAMPLE: Therapist ignores tantrums during session, but child still escapes to get
preferred items by tantruming. Extinction not in effect — the reinforcer (escape/access) is still
accessible.

Effects of Extinction

Page 59 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Extinction Burst
An IMMEDIATE INCREASE in the rate, duration, or intensity of the behavior when
extinction is first implemented. Expected and typical. Signal that the correct reinforcer was
identified. Gets worse before getting better.
Spontaneous Recovery
Reappearance of the behavior after it has diminished to zero (or near-zero). In the SAME
context with the SAME function. Typically weaker than pre-extinction levels. Will
extinguish again if extinction is maintained.
Renewal
Reappearance of the extinguished behavior in a NEW context or setting — even though
extinction was maintained in the original context. Different context = different stimulus
control.
Resurgence
A previously reinforced behavior (different from the one placed on extinction) reappears
when the CURRENT behavior is placed on extinction. The old behavior resurges because
it previously produced the same reinforcer.

✓ BURST EXAMPLE: Client screamed to get items. Extinction implemented. First session:
screaming increases dramatically. Second session: slightly less. Week 2: mostly resolved.
✓ SPONTANEOUS RECOVERY EXAMPLE: Screaming reduced to zero. Two months later,
parent reports screaming returned — same function, same context. Not relapse; normal
spontaneous recovery. Re-implement extinction.
✓ RENEWAL EXAMPLE: Screaming extinguished at school. Same child begins screaming at
home (new context) to get items. Renewal — behavior maintained stimulus control in new
setting.
✓ RESURGENCE EXAMPLE: Client previously bit for access to toys, then was taught to
request. Requesting placed on extinction → biting returns. Biting resurges because it
previously produced the same reinforcer.
✓ DISTINGUISH — Renewal vs Spontaneous Recovery: Spontaneous Recovery = same
context, same function, time has passed. Renewal = new/different context, same function.

★ EXAM TIP: Four extinction phenomena: Burst (temporary increase), Spontaneous

Recovery (returns same context), Renewal (returns new context), Resurgence
(DIFFERENT behavior returns). Exam questions often confuse these.

When NOT to Use Extinction Alone

Page 60 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Dangerous Behaviors
Extinction is NOT recommended as the sole procedure for SIB (self-injury) or severe
aggression. The extinction burst could cause significant harm. Use most rapid, humane
procedure available.
Always Pair with Reinforcement
Extinction alone is rarely sufficient. Always combine with DRA, DRI, or FCT to teach an
appropriate replacement behavior.

Page 61 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Differential Reinforcement
DRO, DRA, DRI, DRL, DRH, DRD — Definitions, Procedures, Examples
22

Overview

Differential Reinforcement
Reinforcing one response class while withholding reinforcement for another response
class. Produces behavior change through selective reinforcement. Five main types in
ABA.

DRO — Differential Reinforcement of Other Behavior

Definition
Reinforcement is delivered ONLY if the target behavior has NOT occurred during a
specified time interval. Any other behavior is acceptable. The reinforcer is time-based.
Two Formats
Whole-session DRO: reinforcement at end of session only if behavior did not occur at all.
Interval DRO (most common): session divided into intervals; reinforcement at end of each
interval during which behavior did not occur. Interval resets if behavior occurs.

✓ EXAMPLE: DRO 15 minutes: if aggression does not occur during the 15-minute interval,
client earns 2 minutes break. If aggression occurs, interval resets.
✗ NON-EXAMPLE — DRO vs NCR: Delivering reinforcement every 15 minutes regardless of
behavior is NCR, not DRO. DRO requires that behavior NOT have occurred.

★ EXAM TIP: DRO does NOT specify WHICH behavior gets reinforced — just that the
target behavior did NOT occur. Compare: DRA/DRI specify a REPLACEMENT behavior;
DRO does not.

DRA — Differential Reinforcement of Alternative Behavior

Page 62 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
A specific ALTERNATIVE behavior is reinforced. Reinforcement is withheld for the
problem behavior. The alternative and the problem behavior CAN both occur
simultaneously (they are not physically incompatible).
FCT as DRA
Functional Communication Training is a specific application of DRA — the alternative is a
communicative response that serves the same function as the problem behavior.

✓ EXAMPLE: Raising hand instead of calling out. Both behaviors CAN occur simultaneously
(student could raise hand AND call out). Reinforcement for raising hand; extinction for calling
out. DRA.
✗ NON-EXAMPLE — DRA vs DRI: If the alternative behavior makes the problem behavior
PHYSICALLY IMPOSSIBLE, it is DRI, not DRA.

DRI — Differential Reinforcement of Incompatible Behavior

Definition
A specific behavior that is PHYSICALLY INCOMPATIBLE with the problem behavior is
reinforced. The two behaviors CANNOT occur simultaneously by definition.

✓ EXAMPLE: Reinforcing sitting at desk (incompatible with running around the room). Cannot
simultaneously sit at desk AND run. DRI.
✓ KEY TEST: Can the person do BOTH behaviors at the same time? YES = DRA. NO = DRI.

DRL — Differential Reinforcement of Low Rates

Definition
Reinforcement is delivered when the behavior occurs at or BELOW a specified criterion
rate. Used to REDUCE but not eliminate a behavior. Criterion systematically tightened
over time.
Two Formats
Interval DRL: behavior must occur no more than N times in an interval for reinforcement.
Spaced-responding DRL: each response must be separated by a minimum amount of
time (IRT criterion).

✓ EXAMPLE: Student asks to use bathroom more than 5 times per period. Intervention: earn
token if bathroom requests ≤ 1 per period. Criterion gradually reduced: 3 per period → 1 per
period.

DRH — Differential Reinforcement of High Rates

Page 63 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
Reinforcement is delivered when the behavior occurs at or ABOVE a specified criterion
rate. Used to INCREASE the frequency of a desirable behavior that occurs too rarely.

✓ EXAMPLE: Student taking too long to eat. Baseline: 10 bites in 20 minutes. DRH: reinforce
if ≥ 12 bites, then ≥ 15, then ≥ 18, progressing toward completing meal in time.

DRD — Differential Reinforcement of Diminishing Rates

Definition
A variation of DRL. Reinforcement if behavior count in current session is LESS than the
previous session. The criterion automatically decreases with each session.

★ EXAM TIP: DR procedure selection guide: Want to eliminate behavior → DRI or DRO.
Want to reduce but keep some → DRL or DRD. Want to increase a replacement → DRA or
DRI. Want to increase a rare behavior → DRH. DRA vs DRI → can they do both at once?

Page 64 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Antecedent Interventions
NCR, High-p Sequence, FCT, Environmental Modification, MO Manipulation
23

Antecedent Control vs Antecedent Intervention

Antecedent Control
Manipulating CONTINGENCY-DEPENDENT antecedent stimuli (SDs, S∆s) to affect
stimulus control over behavior. The consequence still determines the stimulus control.
Antecedent Intervention
Manipulating CONTINGENCY-INDEPENDENT antecedent stimuli (motivating operations)
to change the value of reinforcers. Not dependent on the consequence contingency.

✓ KEY DISTINCTION: SD manipulation = antecedent control (contingency-dependent). MO

manipulation = antecedent intervention (contingency-independent).

Non-Contingent Reinforcement (NCR)

Definition
Delivering a reinforcer on a fixed-time (FT) or variable-time (VT) schedule, REGARDLESS
of behavior. Abolishes the MO for the problem behavior by providing the reinforcer freely.
Mechanism
Functions as an abolishing operation — reduces the establishing operation for the
reinforcer maintaining problem behavior. Value of the reinforcer decreases because it is
no longer scarce.
Fixed Time (FT) Schedule
Reinforcer delivered after a fixed time period regardless of behavior. Not contingent on
any response.
Variable Time (VT) Schedule
Reinforcer delivered after a variable time interval regardless of behavior.

Page 65 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ EXAMPLE: Client elopes to gain access to car rides. Team schedules car rides every 30
minutes regardless of behavior (NCR). Elopement decreases because car ride is freely
available — no longer need to elope to access it.
✗ NON-EXAMPLE: Teacher gives student a sticker every time they raise their hand. This is
contingent — it IS reinforcement of a specific behavior, not NCR.

High-Probability Request Sequence (High-p)

Definition
Two to five easy, mastered requests presented before a low-probability (low-p) demand.
Each high-p request is followed by acknowledgment before the next. Also called behavior
momentum.
Requirements for High-p Instructions
Must be: in the learner's repertoire, have a history of compliance, occur frequently, have
short duration. All must be met before presenting the low-p instruction.
Mechanism
High-p compliance builds behavioral momentum — the pattern of compliance is
established, making subsequent low-p compliance more likely.

✓ EXAMPLE: "Touch your head." (complies) "Clap hands." (complies) "Stomp feet."
(complies) "Go downstairs." (complies — low-p instruction). Behavioral momentum from three
easy responses carries into the difficult one.
✗ NON-EXAMPLE: High-p instruction that is NOT in repertoire (e.g., "What's your address?"
when client doesn't know it). Third instruction fails → momentum broken → low-p instruction
fails.

Environmental Modification

Overview
Modifying the physical or social environment proactively to reduce antecedents for
problem behavior and increase antecedents for appropriate behavior.
Examples
Arranging furniture to limit elopement routes. Positioning preferred materials within reach.
Reducing sensory stimulation (lower noise, lighting). Seating disruptive students near
teacher. Visual boundaries.

Setting Events

Page 66 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Setting Events
Variables (biological, physiological, social, or environmental) that alter the probability of
behavior occurring in the presence of typical antecedent stimuli. Not immediate
discriminative stimuli but CONTEXTUAL factors.
How They Work
Setting events can function as EOs or alter the effectiveness of SDs and reinforcers.
Example: headache makes demands more aversive — increases escape motivation.

✓ EXAMPLE: A student with a headache is more likely to engage in aggression when given
academic demands. The headache is a setting event — it changes the stimulus-response
relationship for demands.
✗ NON-EXAMPLE: Bobby's mother enters the room and Bobby stops singing (because
mother has punished singing in the past). This is NOT a setting event — the mother is
functioning as an SD-p (discriminative stimulus for punishment).

Page 67 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Functional Behavior Assessment

FBA Methods, Functional Analysis, Functions of Behavior (SEAT)
24

Functions of Behavior — SEAT

Positive Reinforcement — Social

Behavior maintained by attention from others or access to tangible items/activities
(socially mediated). Examples: hitting to get attention, grabbing to get a toy.
Positive Reinforcement — Automatic
Behavior produces its own sensory reinforcement directly — not mediated by another
person. Examples: hand-flapping, rocking, repetitive vocalizations.
Negative Reinforcement — Social
Behavior maintained by escape from interactions, demands, or aversive tasks (socially
mediated). Example: tantrum to terminate homework demand.
Negative Reinforcement — Automatic
Behavior removes an internal aversive stimulus automatically. Example: scratching to
relieve itch, covering ears to reduce sensory pain.
Function vs Topography
Same topography can serve different functions. Same function can be served by different
topographies. ALWAYS assess function. Never infer function from topography alone.

✓ TOPOGRAPHY TRAP: Two children hit. Child A hits to get attention. Child B hits to escape
demands. Same topography, different functions. Completely different interventions needed.

★ EXAM TIP: SEAT = Sensory (automatic), Escape, Attention, Tangible. Know all 4. Exam
will present scenarios and ask you to identify function from the description.

FBA Methods

Page 68 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Indirect Assessment
Information gathered WITHOUT direct observation: interviews, checklists, rating scales
(FAST, MAS, QABF). Provides a hypothesis of function. Less accurate. Faster. Provides
historical context.
Direct Assessment
Information gathered through DIRECT OBSERVATION: ABC continuous recording,
scatterplot analysis, structured descriptive assessment. More accurate. Shows full
antecedent-consequence patterns.
Functional Analysis (FA)
Experimental manipulation of antecedent and consequence conditions to DIRECTLY
DEMONSTRATE functional relations. Gold standard. Controlled conditions. CONFIRMS
function rather than merely hypothesizing it.

✓ INDIRECT EXAMPLE: Parent interview + FAST questionnaire → hypothesis: behavior is

escape-maintained. Sufficient to generate a BIP hypothesis.
✓ FA EXAMPLE: Behavior analyst sets up ATTENTION condition, ESCAPE condition,
ALONE condition, PLAY (control) condition. Behavior highest in ESCAPE condition → escape
function confirmed experimentally.
✗ NON-EXAMPLE FA: Analyst watches client in natural environment and determines function
from ABC data. This is DIRECT OBSERVATION (descriptive FBA), NOT a functional analysis.
No controlled conditions set up.

Functional Analysis Conditions

Attention Condition
SD: therapist gives demand or ignores. Consequence: social attention contingent on
behavior. Tests attention function.
Escape Condition
SD: therapist presents demands. Consequence: termination of demands contingent on
behavior. Tests escape function.
Alone Condition
Client alone with no stimulation. Tests automatic/sensory function.
Tangible Condition
Preferred item present, then removed. Behavior produces access to preferred item. Tests
tangible function.
Control/Play Condition
High levels of noncontingent attention, preferred items freely available, no demands.
Behavior should be lowest here — this is the control.

Conditional Probability Analysis

Page 69 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
A formula that calculates the likelihood of behavior occurring given specific antecedent or
consequence conditions. Probability = # occurrences following/preceded by specific event
÷ total occurrences. Results 0.0–1.0. Closer to 1.0 = more likely.

✓ EXAMPLE: 5 instances of pushing. 4 occurred when waiting for peer's preferred toy.
Conditional probability = 4÷5 = 0.80. Pushing is very likely (80%) when waiting for a toy in
peer's possession.

Page 70 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Verbal Behavior
Skinner's Analysis, Verbal Operants, Extensions, Autoclitic
25

Overview

Skinner's Verbal Behavior (1957)

A functional analysis of language — categorizes verbal behavior by its CONTROLLING
VARIABLES (what causes it), not its formal/structural properties. Skinner did NOT reject
formal descriptions — he found fault with failure to account for causes.
Verbal Behavior
Behavior reinforced through the mediation of another person's behavior. The listener
provides the reinforcement. Includes vocal speech, sign language, written language,
pointing, PECS.

Primary Verbal Operants

Page 71 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Mand
Verbal behavior EVOKED by a MOTIVATING OPERATION (EO/AO) and REINFORCED
by a SPECIFIC CONSEQUENCE specified by the mand itself. The reinforcer is named in
the response. Function: requesting.
Tact
Verbal behavior evoked by a NONVERBAL STIMULUS (object, event, property) and
reinforced by GENERALIZED SOCIAL REINFORCEMENT. Contact with the environment
controls the response. Function: labeling, describing, commenting.
Echoic
Verbal behavior evoked by a VERBAL STIMULUS that has POINT-TO-POINT
CORRESPONDENCE with the response. The response immediately follows and
physically matches the model. Function: repeating exactly.
Intraverbal
Verbal behavior evoked by a VERBAL STIMULUS but WITHOUT point-to-point
correspondence. The response is different from the verbal stimulus. Function: answering
questions, fill-in-the-blank, conversation.
Textual
Reading written text aloud. Point-to-point correspondence between written stimulus and
vocal response. Reading without necessarily understanding meaning.
Transcription (Dictation-Writing)
Writing or typing a response to a vocal stimulus. Point-to-point correspondence between
vocal stimulus and written response.

✓ MAND EXAMPLE: Child is thirsty (MO = water deprivation) and says "juice" → receives
juice. Mand: MO present, reinforcer specified and received.
✓ TACT EXAMPLE: Child sees a dog and says "dog!" → parent smiles. No MO for the dog
itself. Nonverbal stimulus (dog) controls response. Social reinforcement.
✓ MAND vs TACT TRAP: Child sees a doll in a store and reaches saying "doll!" → mom
smiles. If child has been asking for a doll all week (MO present), this is a MAND. If the child
just happens to see a doll with no particular wanting, it could be a tact. Context determines
classification.
✓ INTRAVERBAL EXAMPLE: "Have you seen my keys?" → "They're on the table." Verbal
stimulus controlled the response, but the response (table location) differs completely from the
question — no point-to-point correspondence.
✓ ECHOIC vs INTRAVERBAL: "Say thank you" → "thank you" = ECHOIC (point-to-point
match). "What do you say?" → "thank you" = INTRAVERBAL (no match to verbal stimulus).

★ EXAM TIP: Classification depends on controlling variables, not topography. Key

questions: (1) Is there an MO? → Mand. (2) Is the control nonverbal environmental
stimulus? → Tact. (3) Is there a verbal stimulus with a matching response? → Echoic. (4)
Verbal stimulus, different response? → Intraverbal.

Autoclitic

Page 72 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Autoclitic
Verbal behavior in which some aspect of the speaker's own verbal behavior functions as
an SD or as an MO. Modifies, qualifies, or adds meaning to another verbal operant. "Talk
about talk."
Examples
"I THINK it will rain" — "I think" qualifies the certainty of the tact. "I WANT some cookies"
— "I want" marks the operant as a mand. "Can you PLEASE pass the salt" — "please" is
an autoclitic strengthening the mand.

Tact Extensions

Generic Extension
Tact emitted to a new stimulus sharing formal features with the original SD. Example:
calling a horse "dog" because it has four legs like a dog.
Metaphorical Extension
Tact used figuratively. Example: calling a tall person "a giant."
Metonymical Extension
A stimulus closely associated with the original SD evokes the tact. Example: saying "the
crown" to refer to the king (crown is associated with royalty).
Solistic Extension
An awkward, grammatically incorrect extension — still understood in context. Example:
"The dog is bigly" instead of "big."

Page 73 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Contingency Contracting, Token Economies & Group Contin

Behavioral Contracts, Token Economy Design, Group Contingency Types
26

Contingency Contract (Behavioral Contract)

Definition
A document specifying a contingent relationship between the completion of a specified
behavior and access to a specified reward. An agreement between the individual and
practitioners/stakeholders.
Required Components
WHO performs the task. WHAT the task is. WHEN it must be completed. HOW WELL
(criterion). WHAT the reward is. WHEN the reward is delivered. HOW MUCH of the
reward is given. TASK RECORD for monitoring.
How They Work
Package including reinforcement, rules, and response prompting. Can be a self-contract.

✓ EXAMPLE: "When Juan completes his homework (30+ minutes, ≥ 80% correct), he earns
20 minutes of video games before bedtime." Clear who, what, when, how well, and reward.
✗ NON-EXAMPLE: "Be good and you can watch TV." No operational definition, no criterion, no
specified reward amount or schedule.

Token Economy

Page 74 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Definition
A behavior change system where tokens are earned contingent on engaging in desired
behaviors and then exchanged for backup reinforcers.
Components
(1) Target behaviors — operationally defined. (2) Tokens — generalized conditioned
reinforcers. (3) Backup reinforcers — what tokens are exchanged for. (4) Exchange
schedule.
Generalized Conditioned Reinforcer
Tokens have been paired with so many backup reinforcers that they maintain value across
a wide range of motivating operations. Not dependent on any single MO.
Response Cost in Token Economy
Tokens REMOVED contingent on problem behavior. Form of negative punishment.
Reduces problem behavior.
Ratio Strain
If the number of tokens required for backup reinforcers is increased too quickly,
responding breaks down. Thin gradually.
Advantages
Bridges delay between behavior and backup reinforcer. Easy to administer in groups.
Highly motivating. Flexible.

✗ NON-EXAMPLE Tokens: Praise alone is not a token economy. Token economies require
tangible tokens that have exchange value.

Group Contingencies

Overview
A behavior management protocol where a single consequence (reward or aversive) is
given based on the behavior of an individual, a subgroup, or the entire group.
Independent Group Contingency
Each individual in the group receives the consequence based SOLELY on their OWN
behavior. No individual affects others. Example: each student completing their worksheet
earns 5 minutes free time individually.
Dependent Group Contingency
The consequence for the ENTIRE group depends on the behavior of ONE or a FEW
individuals. Example: if one designated student completes their work, everyone earns
recess. Can create peer support but risks bullying if selected student fails.
Interdependent Group Contingency
The consequence is delivered based on the COMBINED or AVERAGE performance of
ALL group members. Everyone must meet criterion. Example: all students must complete
worksheets for the class to earn free time.

Page 75 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ INDEPENDENT EXAMPLE: Everyone in the class who completes their assignment earns a
sticker independently.
✓ DEPENDENT EXAMPLE: A randomly selected student — if they complete their work,
everyone earns extra recess.
✓ INTERDEPENDENT EXAMPLE: Average class score ≥ 80% → everyone earns a pizza
party.

★ EXAM TIP: Know the three types by WHO determines the consequence: Independent =
each individual for themselves. Dependent = one person determines outcome for all.
Interdependent = ALL members' combined behavior determines outcome for all.

Page 76 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Self-Management
Self-Control, Self-Monitoring, Self-Evaluation, Applications
27

Conceptual Foundations

Self-Management
The personal application of behavior change tactics that produces a desired change in
one's own behavior. Intentionally broad. A relative concept. Occurs on a continuum.
Skinner's Conceptualization
Self-control as a TWO-RESPONSE PHENOMENON: (1) Controlling Response — the
self-management behavior itself (the behavior used to change behavior). (2) Controlled
Response — the target behavior to be changed.
Radical Behaviorism & Self-Control
Causal variables for behavior are found in the environment, including the person's history.
When variables are not readily apparent, tendency is to attribute control to the "self" —
Skinner applied behaviorism to this systematically.

✓ EXAMPLE: Setting an alarm (controlling response) to wake up on time (controlled

response). Leaving a sticky note on the fridge (controlling response) to remember to take
medication (controlled response).

Self-Management Behaviors

Page 77 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Self-Monitoring
The individual observes and records their own behavior. Increases awareness of
behavior. Can function as intervention itself (reactivity to self-monitoring — behavior
changes simply from being observed).
Self-Evaluation
The individual compares their own behavior against a criterion or standard. Often
combined with self-monitoring.
Self-Reinforcement
The individual delivers a reinforcer to themselves contingent on achieving a behavioral
goal.
Self-Instructions
Verbal self-statements that prompt or guide behavior. Example: talking through steps of a
task.
Goal Setting
Identifying specific behavioral goals. Increases commitment to performance standards.
Stimulus Control Techniques
Arranging antecedent stimuli to increase or decrease likelihood of target behavior.
Example: removing junk food from the house to reduce snacking.

✓ SELF-MONITORING EXAMPLE: Student uses a wrist counter to count how many times
they blurt out in class. Simply recording the behavior leads to decrease.
✗ NON-EXAMPLE: Parent orders food for child because child is "scared to say the wrong
thing." This does NOT promote self-management — it maintains dependence.

Advantages of Self-Management

Key Benefits
Can influence behaviors not accessible to external intervention (private behaviors).
Promotes generalization across settings. Fades external contingencies. Promotes
independence. Can be applied to an infinite range of behaviors. Relatively low cost.

Page 78 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Generalization & Maintenance of Behavior Change

Stimulus Generalization, Response Generalization, Maintenance, Programming Strat
28

Definitions

Generalized Behavior Change

One of the 7 dimensions of ABA (Generality). Behavior changes that are durable over
time, occur across settings/situations, and spread to related behaviors. Not limited to the
training conditions.
Response Maintenance
The extent to which behavior continues to occur after the intervention has been reduced
or removed. Durability over time.
Setting/Situation Generalization
The behavior occurs in settings or stimulus situations that DIFFER from the training
setting, without specific training in those settings.
Response Generalization
The occurrence of untrained responses that are functionally equivalent to the trained
target behavior. The spread of effect to related behaviors.

✓ MAINTENANCE EXAMPLE: Client stops engaging in aggression after 6 months of

intervention. Follow-up 6 months after intervention ended shows aggression still at zero.
Response maintenance demonstrated.
✓ SETTING GENERALIZATION EXAMPLE: Client learned to request breaks at home during
ABA sessions. Teacher reports client is now requesting breaks at school too (untrained
setting).
✓ RESPONSE GENERALIZATION EXAMPLE: Client trained to say "I want a break" to
request breaks. Begins also saying "Stop please," "One minute," and walking to a calming
area — functionally equivalent, untrained responses.

Why Generalization Must Be PROGRAMMED

Page 79 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Generalization Is Not Automatic

Behavior change does not automatically generalize. Training in one context typically
produces strong stimulus control in that context only. Must proactively plan for
generalization from the start.
Stokes & Baer (1977)
Identified "train and hope" as insufficient. Described multiple strategies for programming
generalization.

Strategies for Programming Generalization

Teach Multiple Exemplars

Train across enough examples that the learner extracts the relevant concept and applies it
to untrained examples. Don't rely on a single training stimulus.
General Case Analysis
Systematically select training examples that sample the full range of relevant stimulus
variations and response variations in the generalization setting.
Train in the Natural Environment
Conduct training in the actual settings where behavior is needed. Natural discriminative
stimuli control the behavior directly.
Programming Common Stimuli
Include in the training setting stimuli that are common in the generalization setting.
Stimulus control transfers automatically.
Indiscriminable Contingencies
Make it difficult for the learner to tell when they are being reinforced vs not. Use
intermittent schedules. Prevents the learner from only performing when the trainer is
present.
Mediate Generalization
Teach the learner a mediating behavior (self-instruction, self-monitoring) that can be used
across settings.
Use Natural Reinforcers
Bring behavior into contact with naturally occurring reinforcers in the generalization
setting. Ensures maintenance after training ends.
Train Loosely
Vary training conditions (different trainers, materials, settings, times) to promote
generalization. Avoid rigid uniformity.
Sequential Modification
After training in one setting, systematically apply the intervention to additional settings one
at a time until generalization is achieved.

Page 80 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ MULTIPLE EXEMPLAR EXAMPLE: Teaching "dog" using golden retrievers, labs, poodles,
chihuahuas, and dog cartoons — the variety of examples ensures the label generalizes to
novel dog breeds.
✓ INDISCRIMINABLE CONTINGENCY EXAMPLE: Thinning from CRF to VR schedule —
learner cannot predict when reinforcement will come, so they keep responding. Prevents "only
perform when token board is visible."

★ EXAM TIP: Generalization must be actively programmed — never assumed. Stokes &
Baer (1977) identified multiple strategies. Know the strategies by name and with examples.

Stimulus Equivalence

Stimulus Equivalence
When an individual demonstrates untrained (derived) relations among stimuli. Three
defining properties must all be present: reflexivity, symmetry, and transitivity.
Reflexivity
A stimulus is matched to itself (A = A). Generalized identity matching.
Symmetry
If A = B is trained, then B = A emerges without training.
Transitivity
If A = B and B = C are trained, then A = C emerges without training.

✓ EXAMPLE: Trained: written word "dog" (A) matches picture of dog (B); spoken word "dog"
(C) matches picture (B). Untrained: written word → spoken word, and all reverse relations
emerge. Stimulus equivalence class formed.

Page 81 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

CHAPTER

Ethical Considerations for Applied Behavior Analysts

Professional Ethics, Social Validity, Rights of Clients, Best Practice
29

Foundations of ABA Ethics

Three Fundamental Questions

(1) What is the RIGHT thing to do? (2) What is WORTH doing? (3) What does it mean to
be a GOOD behavior analyst? End goal: FURTHER THE WELFARE OF THE CLIENT.
Personal Influences on Decisions
Personal experiences, cultural/religious upbringing, and professional training all influence
decision-making. When bias may interfere: consult supervisors, review literature, examine
case studies, or recuse yourself.

Social Validity (Wolf, 1978) — Three Levels

Goal Validity
Are the goals of the intervention acceptable and meaningful for the client and society?
Would the client's natural community support these goals?
Procedure Validity
Are the methods used acceptable, ethical, and consistent with best treatment practices?
Outcome Validity
Do the results show meaningful, significant, and SUSTAINABLE change in the person's
real life?

✓ EXAMPLE: Client and family agree that learning to communicate instead of hitting is a goal
that matters, the DRA procedure is acceptable to them, and the change meaningfully reduced
family stress. Social validity addressed at all three levels.

Cost-Benefit Analysis

Page 82 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

Principle
The potential benefit to the individual must justify the short- and long-term cost of
providing the service. Decisions should be made by committee. The person with the
highest stake in the outcome should be given greatest consideration.
Existing Exigencies
More serious behaviors warrant intervention consideration before less problematic
behaviors. Must consider long-term ramifications even for procedures that produce quick
change.

Ethical Obligations of the Behavior Analyst

Follow Professional Code

Adhere to the BACB's (or QABA's) professional and ethical compliance code. This is not
optional — it is a professional obligation.
Evidence-Based Practice
Only implement interventions with demonstrated scientific support. Avoid pseudoscience
and fads (facilitated communication, sensory integration therapy without evidence, etc.).
Informed Consent
Obtain informed consent before implementing procedures. Client/guardian must
understand what will happen, why, and their right to refuse.
Least Restrictive Alternative
Always use the most positive, least intrusive effective intervention first. Escalate restriction
only with documentation that less restrictive options were tried and failed.
Competence
Only practice within areas of competence. Refer out or seek supervision for cases outside
expertise.
Confidentiality
Client data is confidential. Follow HIPAA and state/country law. Minimum 7-year record
retention. No identifying information in public spaces.
Dual Relationships
Avoid multiple relationships that could impair objectivity, effectiveness, or competence.
Small cultural tokens may be appropriate exceptions.
Advocacy
Speak WITH the client — understand their goals first. Speak ON BEHALF OF those goals.
Never speak instead of the client without their input.

Page 83 © Behavior Canvas — For personal study use only

Applied Behavior Analysis — Cooper, Heron & Heward (2nd Ed.) Behavior Canvas — Comprehensive Chapter Notes

✓ ETHICAL EXAMPLE: Analyst encounters a behavior pattern outside their expertise (feeding
disorder). Refers client to an SLP and pediatrician. Does not attempt to treat without
competence.
✓ UNETHICAL EXAMPLE: Company continues services after client has met all goals and has
no remaining clinical needs — because they do not want to lose billable hours. Continuing
unnecessary services violates the client's welfare.

★ EXAM TIP: Ethics questions test both KNOWLEDGE of rules AND APPLICATION to
scenarios. Key principle underlying all ethics: the welfare of the CLIENT is the primary
obligation — above convenience, financial considerations, or institutional pressures.

Page 84 © Behavior Canvas — For personal study use only