Week 5 Chapter 11 Cooper.pdf

Chapter 11: Positive Reinforcement Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Precise Operational Defini...

Chapter 11: Positive Reinforcement Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Precise Operational Definition Define in 4 dimensions: 1. Verbal - What is said, how it is said, under what circumstances 2. Physical -Observable physical movements or actions involved in the behavior 3. Spatial - Location or area where the behavior occurs or the position of the body during the behavior 4. Temporal - Refers to the timing of the behavior including when it occurs, how long it lasts, and the frequency or duration of behavior Key Learnings  Types of Validity Social Validity Includes the social significance of the target behavior, the appropriateness of the procedures, and the social importance of the results External Validity Degree to which a functional relation in an experiment will hold under different conditions Internal Validity Extent to which a study or experiment can demonstrate a causal relationship between an independent variable (intervention/treatment) and a dependent variable (the behavior being measured) Key Learnings  Type I error: when researcher concludes that independent variable had effect on dependent variable, when it did not Statistical analysis leads to more Type I and less Type II errors Type II error: when researcher concludes that independent variable did not have effect on dependent variable, when it did Visual analysis leads to less Type I and more Type II errors Key Learnings  Interobserver Agreement Interobserver Agreement is the degree to which two or more independent observers report the same observed values after measuring the same events. IOA can detect observer drift High IOA increases the confidence that the definition of the target behavior was clear and unambiguous, and the measurement code and system were not too difficult Observers must use the same measurement system Observers must measure the same events Observers must observe and record the behavior independent of any influence from one another Key Learnings  Reinforcement Definition Stimulus presented Contingent on a response Which increases the future probability of the response The future increase in the response is a critical feature in defining reinforcement Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Reinforcement is Not a Circular Concept Circular Reasoning ◦ Faulty logic in which the name used describe the effect is also mistaken for the cause of the phenomenon ◦ Example: Johnny has trouble learning to read (effect). Therefore, he has a learning disability (phenomenon). How do I know he has a learning disability? Because he can’t read (effect now translated into cause) Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Reinforcement is Not a Circular Concept Sometimes, people refer to “reinforcement” as a circular concept--it is not! Example: Robbie’s studying behavior increased when he earned points for studying. ◦ Cause (earning points) and effect (increased study behavior) are different ◦ Points can be manipulated as an independent variable to observe effects on studying Copyright © 2007 by Cooper, Heron, and Heward Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  The Role of Antecedent Stimuli Caveat #1: Reinforcement does not increase behavior under all conditions The temporal relation between ◦ Antecedent variables ◦ Responses ◦ Consequences is important! These antecedent variables become discriminitive stimuli (SDs) Thus, the response is more likely to occur in the future in the presence of these stimuli Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  The Discriminated Operant AKA “The Three-term Contingency” SD Response Tap on faucet SR+ marked with Turn tap Cold water blue dot or with blue presented Turning tap marked letter “C” dot or “C” with blue dot or “C” occurs more often in the future This term is referred to as “the reinforcer” Cooper, Heron, and Heward Copyright © 2008 by Pearson Education, Inc. Applied Behavior Analysis, Second Edition All rights reserved  The Role of Antecedent Stimuli Caveat #2: Reinforcement depends on motivation The SD will only signal the response if the individual is motivated to engage in the response Motivating Operations (MOs) ◦ Alter the reinforcing effectiveness of stimuli, and thus ◦ Alter the momentary frequency of responses reinforced by those stimuli Copyright © 2007 by Cooper, Heron, and Heward Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Motivating Operations Establishing Operations (EO) ◦ Increases the effectiveness of a stimulus as a reinforcer ◦ Usually involves decreased access to the stimulus (deprivation) Abolishing Operation (AO) ◦ Decreases the effectiveness of a stimulus as a reinforcer ◦ Usually involves having increased access to the stimulus (satiation) Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  The Four-term Contingency The consideration of MOs are important in relation to the three-term contingency EO SD Response SR+ Deprived of Tap on faucet Turn tap Cold water water for a marked with with blue presented long period of blue dot or dot or “C” Turning tap marked with blue dot or time letter “C” “C” occurs more often in the future when the individual has been deprived of water for We only expect blue tap-turning behavior when the periods of time person “wants” water (i.e., is thirsty) Cooper, Heron, and Heward Copyright © 2008 by Pearson Education, Inc. Applied Behavior Analysis, Second Edition All rights reserved  Questions About Reinforcement Does a person have to be aware that a response is being reinforced for it to increase? ◦ NO! The effect is automatic. Are certain behaviors susceptible to reinforcement and others are not? ◦ NO! The only relevant relevant property is the temporal relation between the response and the consequence. Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Immediacy of Reinforcement It is critical that the consequence is delivered immediately following the target response Problems with delays to reinforcement ◦ Other behaviors occur during the delay ◦ The behavior temporarily closest to the presentation of the reinforcer will be strengthened Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Delayed Reinforcement Does not necessarily reinforce the target behavior; rather influences it Instructional Control/Rule Following ◦ Rule: verbal description of a behavioral contingency ◦ Can allow delayed consequences to influence behavior Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  “Rule-governed Behavior” Indicators ◦ No immediate consequence apparent ◦ Response-consequence delay > 30 s ◦ Large increase in frequency of the behavior occurs following one instance of reinforcement ◦ No consequence for the behavior exists (including no automatic reinforcement), but rule does Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Superstitious Behavior Occurs when reinforcement “accidentally” follows a behavior that did not produce the reinforcement ◦ Sports players who equate putting on a certain pair of socks with winning a game (leading to the “lucky socks” idea) ◦ A teacher consoling a child who hurt himself may reinforce crying and/or hurting oneself Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Automatic Reinforcement Reinforcement that occurs independent of another person delivering it The response, itself, produces the reinforcement Examples ◦ Wiggling your leg during a boring lecture to stimulate yourself and stay awake Note: This does not mean the behaviors are automatic (i.e., “reflexive”); rather that the consequences are delivered automatically Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Classifying Reinforcers Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Reinforcers by Origin Unconditioned Reinforcers (AKA primary or unlearned reinforcers) ◦ Function as reinforcers due to heredity/evolution ◦ Do not require any learning history to become reinforcers ◦ Examples: Food, water, oxygen, warmth, sexual stimulation, human touch Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Reinforcers by Origin Conditioned Reinforcers (AKA secondary or learned reinforcers) ◦ Neutral stimuli that begin to function as reinforcers as a result of being paired with other reinforcers (either conditioned or unconditioned) ◦ Can also condition reinforcers through verbal analog conditioning ◦ Examples: Yellow paper, stickers, tokens Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Generalized Conditioned Reinforcers A type of conditioned reinforcer that has been paired with many conditioned and unconditioned reinforcers Do not depend on a specific EO to be effective Examples: tokens, money, points Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Reinforcers by Formal Properties Edible reinforcers (food) Sensory reinforcers (massage, tickles) Tangible reinforcers (trinkets, toys) Activity reinforcers (playing a game, recess) Social reinforcers (physical proximity, social interaction) Bear this in mind: Items that function as reinforcers are idiosyncratic across people! Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Identifying Potential Reinforcers It is important to identify reinforcers empirically ◦ Staff, parents, teachers, and even children themselves who report what they believe to be reinforcers are often wrong Two strategies to use in tandem 1. Stimulus Preference Assessments 2. Reinforcer Assessments Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Caveats Regarding Preference/Reinf. Assess. Preference changes over time ◦ Evaluate frequently Preference assessments do not identify the reinforcing effects of stimuli ◦ Just because people prefer paper towels to hot-air hand dryers in public restrooms doesn’t mean they’ll work to earn paper towels! Copyright © 2007 by Cooper, Heron, and Heward Pearson Education, Inc. Applied Behavior Analysis, Second Edition All rights reserved  Stimulus Preference Assessments Identify ◦ Stimuli a person prefers ◦ Relevant preference values ◦ Conditions under which these preferences hold true Three Categories ◦ Asking about stimulus preferences ◦ Observing the target person under free-operant conditions ◦ Presenting various stimuli in a series of trial-based observation Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Asking About Stimulus Preferences Ask the Target Person ◦ Open-ended questions ◦ What would you like to work for? ◦ Asking about specific items ◦ How would you like to work for stickers? ◦ Choice format ◦ Would you rather work for things to eat or things to do? ◦ Rank order format ◦ Put these items/activities in order from which you’d like to work for most to which you’d like to work for least. Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Asking About Stimulus Preferences Offering Pre-task Choices ◦ When you are finished working, you can play with Battleship, checkers, or the computer Asking Significant Others ◦ Ask caregivers to identify preferred stimuli Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Asking About Stimulus Preferences A relatively uncomplicated procedure Problems ◦ Verbal reports may not correspond to actual behavior ◦ High number of false positives and low number of false negatives Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Free-Operant Observation Observing and recording what activities the target person engages in when he/she has unrestricted choice of activities No response requirements All stimuli available within sight and reach Items are never removed Can be contrived or naturalistic Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Contrived Free-Operant Observation Just prior to observation, provide learner with noncontingent exposure to each item (for sampling purposes) Place all items in view and within reach Observe for a set period of time and record the duration of time target person engages with each stimulus item Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Naturalistic Free-Operant Observation Conducted in everyday environments as unobtrusively as possible (e.g., during recess) Observe for a set period of time and record the duration of time target person engages with each stimulus item/activity Copyright © 2007 by Cooper, Heron, and Heward Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Advantages of Free-Operant Assessments Less time consuming than some trial-based methods of preference assessment. Less likely to produce problem behavior because preferred stimuli are never removed. Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved Free Operant Preference Assessment  Trial-Based Methods General Procedure ◦ Present selected stimuli to children in a series of trials ◦ Measure approach (e.g., eye gaze, hand reach), contact (e.g., touch/hold), and/or engagement (e.g., interacting with stimulus) ◦ Can categorize as high, medium, and low preference Many variations for procedure Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Trial-based Method 1: Single Stimulus Presentation Present stimuli, one at a time, in random order and record target person’s reaction to it Well suited for individuals who have difficulty selecting among two or more stimuli Copyright © 2007 by Cooper, Heron, and Heward Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved Single Stimulus Preference Assessment  Trial-based Method 2: Paired Stimuli Presentation Sometimes called “forced-choice” method Present two stimuli simultaneously and ask the target person to choose one Each stimulus is matched to every other stimulus in the set Rank order from high, medium, and low preference Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Edition Education, Inc. Paired Stimulus Preference Assessment  Trial-based Method 3: Multiple Stimulus Presentation Extension of the paired-stimuli presentation Present an array of 3 or more stimuli together Two major variations: ◦ With replacement ◦ Stimulus selected remains in array in subsequent trials ◦ Without replacement ◦ Selected stimulus is removed from the array in subsequent trials (takes about half the time to complete the procedure, and it is still fairly accurate) Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Trial-based Method 3: Multiple Stimulus Presentation Begin trial with: Which one do you want the most? Repeat several times Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved Preference Assessment with Food: Multiple Stimulus with Replacement (MSW) Preference Assessment with Toys: Multiple Stimulus without Replacement (MSWO)  Guidelines for Selecting and Using Stimulus Preference Assessments Monitor target person’s activities prior to assessment to be aware of EOs that may affect results Balance cost-benefits of procedures (time to do vs. level of confidence) Balance rankings vs. no rankings with shifts of preference When time is limited, use fewer stimuli in array When possible, combine data from multiple assessment procedures Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Reinforcer Assessment A direct, data-based method in which ◦ One or more stimuli are presented ◦ Contingent on a target response, and ◦ Observing whether an increase in responding occurs Allows you to verify/confirm whether a stimulus functions as a reinforcer Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Concurrent Schedule Reinforcer Assessment Determines effectiveness of different reinforcers Two or more reinforcement schedules are presented at the same time Effectiveness is measured by how often the individual engages in the behavior that leads to the specific reinforcer Example: Press one button and receive a token for extra playtime and press another and receive a piece of candy Schedules of Reinforcement define WHEN and HOW reinforcement will be delivered. Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved  Multiple Schedule Reinforcer Assessment Unlike a concurrent schedule reinforcers are presented on at a time Measure how frequently the individual engages in a target behavior under each reinforcement schedule Example: Receive tokens for completing math problems during one schedule (green card SD) and receive a break from work during another schedule (red card SD) Copyright © 2007 by Cooper, Heron, and Heward Pearson Education, Inc. Applied Behavior Analysis, Second Edition All rights reserved  Progressive-Ratio Schedule Reinforcer Assessment Determine the strength or effectiveness of a reinforcer by systematically increasing the response requirement needed to obtain the reinforcer Determines the “Breaking Point” or when the individual stops responding due to the increased effort required Example: must complete a math problem to earn a token, and after each token the number of problems required increases (1 problem = 1 token; 3 problems for second token, 6 problems for next token) Determines the strength of the reinforcer Cooper, Heron, and Heward Copyright © 2007 by Pearson Applied Behavior Analysis, Second Education, Inc. Edition All rights reserved Differential Reinforcement Reinforcement that help shape more appropriate behavior by reinforcing either the occurrence or non-occurrence of target behaviors under certain conditions  Types of Differential Schedules of Reinforcement DRA –reinforces a behavior that serves as an alternative to an undesired behavior DRI – reinforces a behavior that is incompatible with (i.e., cannot occur at the same time as) the undesired behavior DRO – reinforces the absence of a target undesired behavior for a specified amount of time DRL- reinforces lower rates of an undesired behavior, rather than trying to eliminate it entirely DRH – reinforces higher rates of a desirable behavior to increase its frequency  Control Procedures for Positive Reinforcement When evaluating the effects of reinforcement in an ABAB reversal design: ◦ “the ideal control procedure…eliminates the contingent relation between the occurrence of the target response and the presentation of the stimulus while controlling for the effects of stimulus presentation alone” (Thompson & Iwata, 2003, p. 259). Perhaps a noncontingent schedule of reinforcement is the appropriate control (A) condition as a comparison for the positive reinforcement (B) condition. Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  Control Procedures for Positive Reinforcement DRO may be another appropriate control procedure ◦ May produce a reversal more quickly than the NCR schedule DRA could be used as a control procedure to reinforce another alternative response Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition  Control Procedures for Positive Reinforcement Limitations of DRO/DRA as controls ◦ Introduce new contingencies that were not present in original experimental arrangement ◦ Reversals may be due to ◦ Termination of a contingency between target response and reinforcer ◦ Introduction of reinforcement for absence of the target response or for the occurrence of a competing response Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  12 Guidelines for Using Reinforcement Effectively 1. Choose reinforcers relevant to current or creatable establishing operations 2. Maintain establishing operations 3. Use high-quality reinforcers of sufficient magnitude 4. Set an easily achieved initial criterion for reinforcement -criterion should be less than or equal to best performance during baseline Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  12 Guidelines for Using Reinforcement Effectively 5. Explain the contingency and provide prompts to respond 6. Deliver the reinforcer immediately following behavior 7. Reinforce each occurrence of the behavior initially 8. Use direct rather than indirect reinforcement contingencies Cooper, Heron, and Heward Copyright © 2007 by Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved  12 Guidelines for Using Reinforcement Effectively 9. Gradually increase response-to-reinforcement delay 10. Use varied reinforcers 11. Use contingent praise and attention 12. Shift from contrived to naturally occurring reinforcers Copyright © 2007 by Cooper, Heron, and Heward Applied Behavior Analysis, Second Pearson Education, Inc. Edition All rights reserved