timothy r. vollmer. presentation... · 2019. 4. 22. · eliana pizarro, university of florida kerri...

33
7/19/2018 1 Differential Reinforcement as a Way of Life TIMOTHY R. VOLLMER Co-authors Meghan Deshais, Caldwell University Faris Kronfli, University of Florida Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center

Upload: others

Post on 17-Aug-2021

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

1

Differential Reinforcement as a Way of Life

TIMOTHY R. VOLLMER

Co-authorsMeghan Deshais, Caldwell University

Faris Kronfli, University of Florida

Eliana Pizarro, University of Florida

Kerri Peters, University of Florida

Sarah Freeman, Marcus Center

Page 2: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

2

My backgroudUniversity of Florida with Brian Iwata, self-injurious behavior

Louisiana State University—School psychology

University of Pennsylvania– Medical school

University of Florida– schools, foster care, autism centers

OverviewI will review evidence that severe problem behavior is learned, operant behavior.

I will discuss the concept of “choice” as it relates to operant behavior.

I will describe how differential reinforcement is a logical treatment and should be adopted as a “way of life.”

I will describe ways of transferring behavioral treatment into everyday life via care provider training.

Page 3: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

3

Behavior Disorders/Problem BehaviorSelf-Injurious Behavior (SIB)

Aggression

Property Destruction

Tantrums

Severe Stereotypic Behavior

Classroom Disruptive Behavior

Operant Functions of Behavior DisordersSocially mediated positive reinforcement

Socially mediated negative reinforcement

Automatic positive or negative reinforcement

Page 4: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

4

Examples of socially mediated positive reinforcement maintaining problem behaviorAttention in the form of comfort statements

Attention in the form of proximity

Attention in the form of reprimands

Attention in the form of social interaction

Tangible items such as preferred toys, food items, drinks, videos, computers, etc.

Examples of socially mediated negative reinforcementEscape or avoidance of instructional activity (includes reduced duration of instructional activity)

Escape or avoidance of self-care or daily living routines

Escape or avoidance of medical routines

Escape or avoidance of aversive sounds or situations

Page 5: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

5

Important Note:Sometimes the individual has alternative behavior, such as communication, in their repertoire, but…

Problem behavior produces consequences more reliably and more immediately

Examples of automatic reinforcementThe sensation produced by the behavior functions as positive reinforcement.

Behavior, such as self-scratching, temporarily attenuates aversive stimulation.

Bio-behavioral theories (e.g., endorphin hypothesis).

Page 6: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

6

Functional analysisAn experimental manipulation of independent variables thought to potentially control target behavior (the dependent variable).

The term has a more general meaning, but has come to be used to refer to a specific type of assessment for behavior disorders.

The utility of a FAScientific: To learn more about the nature of and controlling variables for behavior disorders.

Research screening: To identify appropriate subjects for a research question.

Clinical assessment: To isolate variables maintaining or suppressing problem behavior.

Page 7: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

7

Clark

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

0

1

2

3

Play

Lana

5 10 15 20 25 30 35

Res

pons

es p

er m

inut

e

0

1

2

3

4

5

6

7

Tangible

Ignore

George

Sessions

2 4 6 8 10 12 14 16 18 20 22

0

1

2

3

Attention

Escape

Athens & Vollmer, 2010

Slocum & Vollmer, 2015

Res

po

nse

s p

er m

inu

te

Negatively reinforced (escape) behavior

Page 8: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

8

0 10 20 300

20

40

60

80

100

DemandPlay

Attention

No Interaction

Sessions

Perc

enta

ge o

f S

essio

n w

ith S

tere

oty

py

0 10 20 300

20

40

60

80

100

No Interaction

Attention

Play

Demand

Sessions

Perc

enta

ge o

f S

essio

n w

ith S

tere

oty

py

0 10 20 30 400

20

40

60

80

100

DemandPlay

AttentionNo Interaction

Sessions

Perc

enta

ge o

f S

essio

n w

ith S

tere

oty

py

0 10 20 300

20

40

60

80

100

No Interaction

Attention

Play

Demand

Sessions

Perc

enta

ge o

f S

essio

n w

ith S

tere

oty

py

First FA Second FA

Timmy

Abby

Automatically reinforced behavior (Wunderlich & Vollmer, in press)

Revisiting Differential Reinforcement of Alternative BehaviorOld definition: Reinforcement of some alternative response while placing the problem behavior on extinction.

New definition: Providing greater reinforcement, along at least one dimension, for alternative behavior in comparison to reinforcement for problem behavior.

Page 9: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

9

Differential ReinforcementMinimize reinforcement for problem behavior (preferably via extinction-- withholding reinforcement*)

Maximize reinforcement for appropriate alternative behavior

6 12 18

0

1

2 BL DRA BL DRA

Mand

Problem Behavior

Session

Resp

on

es P

er

Min

ute

Page 10: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

10

Differential Reinforcement of Alternative

Behavior (DRA)• DRA is essentially a concurrent schedule.

• Baseline circumstances (reinforcement schedules) usually favor

problematic behavior.

• Treatment circumstances represent schedules that favor appropriate

behavior.

• Ideally, Extinction vs. Reinforcement.

• However, there are circumstances when extinction is not possible or

practical.

Examples of factors influencing the

application of extinction schedule

• Treatment integrity failures.

• Legal or ethical requirement to block attention-maintained self-injury

or aggression.

• Extinction burst is too dangerous.

• Automatic reinforcement.

• Large and/or fast individuals may produce escape even if we attempt

escape extinction.

Page 11: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

11

The Matching Law

In a concurrent arrangement, the relative rate of one

response alternative will essentially “match” the relative

rate of reinforcement available for that response

alternative.

Herrnstein, 1961

Pro

port

ion

of

Res

pon

ses

on

Key

A

Proportion of Reinforcement for Key A

Matching in Pigeon Key Pecks

Page 12: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

12

Pro

port

ion

of

3 p

tsh

ots

Proportion of Reinforcement for 3 pt shots

Bourret & Vollmer, 2000

Matching in College Basketball:

2 and 3 point shots

Matching in Child Problem Behavior

Borrero & Vollmer, 2002

Page 13: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

13

Averages

Borrero et al., 2010

What do Behavior Analysts mean by “choice”?

• Allocation of responding on two or more response alternatives

• Each alternative is associated with some schedule of reinforcement,

punishment, or both

• Allocation of responding is governed by the outcome of responding

(consequences to behavior)

Page 14: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

14

Other factors influencing response allocation

• Quality of reinforcement

• Magnitude/duration of reinforcement

• Delay to reinforcement

• Response effort

• Punishment

Borrero et al. (2005)

Page 15: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

15

Differential Attention: Baseline

ExampleAggressive Behavior Appropriate Behavior

Probability of Attention 1.0 0.2

Delay to Attention

Quality of Attention

Duration of Attention

Differential Attention: Baseline

ExampleAggressive Behavior Appropriate Behavior

Probability of Attention 1.0 0.2

Delay to Attention < 3 sec on average > 20 sec

Quality of Attention

Duration of Attention

Page 16: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

16

Differential Attention: Baseline

ExampleAggressive Behavior Appropriate Behavior

Probability of Attention 1.0 0.2

Delay to Attention < 3 sec on average > 20 sec

Quality of AttentionVerbal and Physical

AttentionBrief Verbal Attention

Duration of Attention

Differential Attention: Baseline

ExampleAggressive Behavior Appropriate Behavior

Probability of Attention 1.0 0.2

Delay to Attention < 3 sec on average > 20 sec

Quality of AttentionVerbal and Physical

AttentionBrief Verbal Attention

Duration of Attention > 20 sec < 3 sec

Page 17: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

17

Differential Attention: Solution

Aggressive Behavior Appropriate Behavior

Probability of Attention 1.0 1.0

Delay to Attention

Quality of Attention

Duration of Attention

Differential Attention: Solution

Aggressive Behavior Appropriate Behavior

Probability of Attention 1.0 1.0

Delay to Attention < 3 sec < 3 sec

Quality of Attention

Duration of Attention

Page 18: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

18

Differential Attention: Solution

Aggressive Behavior Appropriate Behavior

Probability of Attention 1.0 1.0

Delay to Attention < 3 sec < 3 sec

Quality of Attention Physical AttentionVerbal and Physical

Attention

Duration of Attention

Differential Attention: Solution

Aggressive Behavior Appropriate Behavior

Probability of Attention 1.0 1.0

Delay to Attention < 3 sec < 3 sec

Quality of Attention Physical AttentionVerbal and Physical

Attention

Duration of Attention < 10 sec > 20 sec

Page 19: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

19

Clark

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

0

1

2

3

Play

Lana

5 10 15 20 25 30 35

Res

pons

es p

er m

inut

e

0

1

2

3

4

5

6

7

Tangible

Ignore

George

Sessions

2 4 6 8 10 12 14 16 18 20 22

0

1

2

3

Attention

Escape

Athens & Vollmer, 2010

Re

spo

nse

s p

er

Min

Sessions

Greg

Res

po

nse

s p

er

Min

Sessions

Athens & Vollmer, 2010

Greg

Page 20: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

20

Vollmer et al., 1994

Perc

enta

ge o

f In

terv

als

Page 21: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

21

Using Differential ReinforcementMaximize reinforcement for appropriate behavior

Present only the minimal amount of reinforcement necessary for inappropriate behavior; when possible, this would be none at all

Just remember this rule of thumb: Maximize/Minimize

Translating to everyday life

Page 22: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

22

Care providers make errors

•Errors are made by care providers of all sorts.

•These errors can be interpreted as resulting from contingencies of reinforcement and punishment.

•As behavior analysts, we should be exploring contingencies on care provider behavior, rather than complaining when they do not follow our rules or instructions.

Care Provider Errors

•Delivery of potential reinforcers for problematic behavior

•Failure to promote independence

Page 23: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

23

Attention as an Example

Vollmer et al., 2001

Attention given disruptive behavior

Page 24: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

24

Why?

Contingencies Influencing Care Provider Reprimanding

• It is likely that the child’s behavior is an aversive stimulus• If the child’s behavior temporarily goes away when the reinforcer is delivered, a possible negative reinforcement contingency maintains the care provider’s behavior.• If care provider behavior produces problem behavior, the care provider behavior is likely punished•Rules provided by behavior analysts relate to delayed and probabilistic contingencies that the care provider may have never contacted•We should not be surprised these rules do not work; they cannot compete with immediate reinforcement

Page 25: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

25

Sloman et al., 2005

Sloman et al., 2005

Page 26: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

26

Sloman et al., 2005

Sloman et al., 2005

Page 27: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

27

Miller, Lerman, & Fritz, 2010

Escape/Avoidance as an Example

Page 28: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

28

Three Cases (FAs Showing Escape Behavior)1. Probability of problem behavior given instruction:

Child 1: .93Child 2: .89Child 3: .67

2. Probability of problem behavior given casual social interaction:Child 1: .12Child 2: .15Child 3: .04

3. Probability of problem behavior given no social interaction:Child 1: 0Child 2: 0Child 3: .02

One Possible Solution: Competency-Based Training

Page 29: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

29

Competency-Based Parent Training Accomplishes Two Aims

1. Brings the parent’s behavior into contact with the reinforcer of reduced child problem behavior and increased appropriate behavior, and…

2. Results in correspondence between the behavior analysts instructions and longer-term outcomes

Marcus et al, 2001

Page 30: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

30

0 10 20

0

1

2

Ignore

Attention

Tangible

Escape

Control

Session

Resp

on

es P

er

Min

ute

Preferred toys

Attention

6 12 18

0

1

2 BL DRA BL DRA

Mand

Problem Behavior

Session

Resp

on

es P

er

Min

ute

Page 31: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

31

Initial

Marcus, Swanson, & Vollmer, 2001

Behavioral Skills Parent Training

Identify effective treatment in highly controlled circumstances, then:

Step 1: Didactic interaction

Step 2: Role play A—parent as kid

Step 3: Role play B– therapist as kid

Step 4: Immediate feedback

Step 5: Delayed feedback

Step 6: Monitoring and follow up

Booster training as necessary

Page 32: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

32

Kronfli & Vollmer, in progress

Kronfli & Vollmer, in progress

Page 33: TIMOTHY R. VOLLMER. Presentation... · 2019. 4. 22. · Eliana Pizarro, University of Florida Kerri Peters, University of Florida Sarah Freeman, Marcus Center. 7/19/2018 2 My backgroud

7/19/2018

33

Post

Conclusions

•Problem behavior is often predictable and lawful

•Differential reinforcement is an ideal treatment because it involves minimizing reinforcement for problem behavior and maximizing reinforcement for appropriate alternative behavior

•The matching law is useful in developing interventions, especially when extinction is not likely or even not possible

•Care-provider behavior is sensitive to contingencies of reinforcement and punishment

• It is our job to understand those contingencies in order to implement successful interventions

•Differential reinforcement is not just a treatment, it is a way of life!