Chapter 15 - Differential Reinforcement

Premack Principle

The Premack principle is the opportunity to engage in a high-frequency or preferred behavior (i.e. playing video games) as a reinforcer for a low-frequency behavior (completing homework).

Whole-interval DRO

The problem behavior must be absent for the whole interval for the reinforcer to be delivered.

Interresponse time (IRT)

Is the time between responses The time between the occurrence of consecutive responses (in differential reinforcement procedures).

Stereotypic behavior (often called self-stimulatory behaviors)

Stereotypic behavior is a repetitive behavior that does not serve any social function for the person. They produce some sort of sensory stimulation for the person.

Variations of DRL

-Full-session DRL is where reinforcement is delivered if fewer than a specified number of responses occurs in a period of time. The session might be a class period or some other appropriate period at home, school, work. Contrast this procedure with a DRO procedure, in which the student would have to refrain completely from the behavior during the session to receive the reinforcer. -Spaced-responding DRL is when there must be a specified amount of time between responses for the reinforcer to be delivered. The objective is to pace the behavior.

Intermittent reinforcement vs continuous reinforcement

Continuous reinforcement for the desirable behavior is used in the early stages of DRA. However, once the desirable behavior is occurring consistently and the undesirable behaviors occur rarely, if at all, you should start to thin the schedule of reinforcement and reinforce the desirable behavior intermittently. Intermittent reinforcement maintains the desirable behavior over time by making it more resistant to extinction.

Program for Generalization

In DRA, it is important not only to program for maintenance with an intermittent schedule of reinforcement but also to program for generalization. Generalization means that the target behavior should occur outside the training situation in all relevant stimulus situations.

Cowdery, Iwata, and Pace working with Jerry, a 9-year old boy who engaged in SIB in which he scratched or rubbed his skin until he produced open sores all over his body.

The reinforcer for the scratching SIB was its sensory consequences. DRO was used and reinforcers were applied when the problem behavior was absent for X amount of time, then X + Y amount of time, and X + Y + Z amount of time etc, until Jerry could go a full day without scratching. Although the reinforcer for Jerry's scratching (its sensory consequence) was not eliminated, the reinforcers for the absence of the scratching apparently were stronger (strong enough to produce a decrease in the problem behavior, even though it continued to be reinforced). *Whenever possible, extinction should be a component of the DRO procedure.

Momentary DRO

The problem behavior must be absent at the end of the interval (the short time before the interval is about to end) for the reinforcer to be delivered. *not really effective as a procedure in on itself, but, may be useful in maintaining the behavior manifested from whole-interval DRO.

Steps in implementing the DRO procedure

1. Identifying the reinforcer for the problem behavior -Reinforcing the absence of the problem behavior would not be effective if instances of the problem behavior continued to be reinforced. If it is not possible to use extinction for the problem behavior, it usually will not be possible to use DRO effectively. An exception would be teh case in which the reinforcer for the absence of the problem behavior is more powerful or potent than the reinforcer for the problem behavior itself. Payoff for not engaging in problem behavior is bigger than pay-off for engaging in the problem behavior). 2. Identifying the reinforcer to use in the DRO procedure. -One consequence that is certain to function as a reinforcer for the person is the reinforcer for the problem behavior identified in the functional assessment. If a reinforcing event is maintaing the problem behavior, this reinforcer should be effective in a DRO procedure when it is made contingent on the absence of the problem behavior. 3. Choosing the initial DRO time interval. DRO involves delivering the reinforcer after an interval of time in which the problem behavior does not occur. -The length of the interval for delivering the reinforcer should be tied to the baseline rate of the problem behavior: if the problem behavior occurs frequently, the DRO interval will be short; if the problem behavior occurs infrequently, the DRO interval will be longer. As the frequency of the problem behavior decreases, the DRO intervals can be lengthened gradually. 4. Implementing DRO: elmininate the reinforcer for the problem behavior and deliver the reinforcer for the absence of the problem behavior. The change agent has a stopwatch to time the DRO interval. At the end of each interval, the change agent is cued by the stopwatch to deliver the reinforcer ONLY IF the problem behavior hasn't occurred. For instance, if the DRO interval is 10 minutes, any time the problem behavior occurs, 10 minutes is reset on the stopwatch again. However, if the problem behavior DOESN'T occur in the 10 minutes, the reinforcer is delivered.

Differential Reinforcement of Alternative Behavior (DRA)

Differential Reinforcement of Alternative Behavior (DRA) is a procedure used to increase the frequency of desirable behavior via reinforcement and to decrease the frequency of undesirable behaviors via extinction. By decreasing an interfering problem behavior through extinction, it creates an opportunity for the desirable behavior to occur and be reinforced.

Differential Reinforcement of Low Rates of Responding (DRL)

In DRL, the reinforcer is delivered when the rate of the problem behavior is decreased to a criterion level. In the DRL procedure, a lower rate of the problem behavior is reinforced. The behavior may not be a problem in itself, but, high levels of it may be (i.e. student raising his hand too often in class).

Differential Reinforcement of Other Behavior (DRO)

In DRO, the reinforcer is contingent (liable to occur but not with certainty; possible) on the absence of the problem behavior. This means that the reinforcer is no longer delivered after the problem behavior (extinction), but the reinforcer is delivered after an interval of time in which the problem behavior does not occur. The name of DRO is confusing. You are not reinforcing other behavior. Rather, you are reinforcing the ABSENCE of the problem behavior.

When/how to use Differential Reinforcement of Alternative Behavior (DRA)

In order to use DRA,the desirable behavior must be occurring at least occasionally if you are to reinforce it. If the behavior is not occurring at all, DRA by itself is not an appropriate procedure. In order to use DRA effectively, several steps must be undertaken: 1. Define the desirable behavior 2. Define the undesirable behavior 3. Identify the reinforcer The same reinforcer as the problem behavior can be used to implement the alternative behavior OR the premack principle may be used 4. Reinforce the desirable behavoir immediately and consistently (continuous reinforcement schedule) 5. Eliminate reinforcement for the undesirable behaviors 6. Use intermittent reinforcement to maintain the target behavior 7. Program for generalization

Differential reinforcement procedures

Involve applying reinforcement and extinction to increase the occurrence of a desirable target behavior or to decrease the occurrence of undesirable behaviors. Three types of differential reinfocement procedures: -Differential reinforcement of alternative behavior (DRA) -Differential reinforcement of other behavior (DRO) -Differential reinforcement of low rates of responding (DRL)

Differential Negative Reinforcement of Alternative Behaviors (DNRA)

The use of negative reinforcement in order to replace a problem behavior with a desirable behavior. Jason (autistic 8-year-old boy) who threw tantrums when he was faced with tough school work. Was allowed to escape. However, after implementation of DNRA, where, he was only allowed to escape AFTER successful completion of the problem, the problem behavior was more or less fixed. -Jason threw tantrum -> ignore and make him complete the problem -Jason completed problem -> give him a few minutes alone at the back of the class (DNRA). Use continuous reinforcement at first. Then DNRA after every two problems, then three problems, etc.

Variations of DRA: -Differential reinfocement of an incompatible behavior (DRI) -Differential reinfocement of communication (DRC); which is called Functional Communication Training

Variations of differential reinforcement of alternative behavior (DRA) exist in which different types of alternative behavior are reinforced to replace the problem behavior. -Differential reinforcement of an incompatible behavior (DRI), in which the alternative behavior is physically incompatible with the problem behavior, and therefore can't occur at the same time (i.e. head-slapping, in which individuals slap themselves on the side of the head with their hands, any alternative behavior involving the use of the hands, like playing with toys, would be an incompatible behavior). -Differential reinforcement of communication (DRC), but called Functional Communication Training, is another variation of DRA where the alternative behavior that is reinforced to replace the problem behavior is a communication response. Individual with the problem behavior learns to make a communication response that is functionally equivalent to the problem behavior (i.e. individual with a problem behavior reinforced by attention would learn to ask for attention).