Mazur Chapter 6: Reinforcement Schedules Flashcards
Reinforcement Schedule
The reinforcement schedule is simply a rule that states under what conditions a reinforcer will be delivered
Continuous Reinforcement (CRF)
Every occurrence of the operant response is followed by a reinforcer
Most behaviors outside the laboratory receive only intermittent reinforcement
Skinner devoted considerable effort to the investigation of how different schedules of reinforcement have different effects on behavior
Cumulative Recorder
A simple mechanical device which records responses in a way that allows any observer to see the moment-to-moment patterns of the subjects behavior
Y-axis represents the cumulative number of responses
X-axis represents time
The greater the slope, the faster the response rate
Fixed-Ratio (FR)
A reinforcer is delivered after every N responses, where N is the size of the ratio
CRF = FR1 [reinforcer for each response]
After subject has performed on an FR schedule for some time, a distinctive pattern of stop-and-go develops
After each reinforcer, there is a pause in responding that is called a post reinforcement pause
Eventually, this pause gives ways to an abrupt continuation of responding
*With FR schedules, the average size of the post-reinforcement pause increases as the size of the ratio increases
Variable-Ratio (VR)
A subject will receive, on average, one reinforcer for every N responses
The exact number responses required at any moment may vary widely
Results in a rapid and fairly steady pattern of responding
*The major difference between FR and VR is that post reinforcement pauses are typically quite brief on VR schedules
Logic: after each reinforcer, there’s always the possibility that another reinforcer will be delivered after only a few additional responses= shorter PRP
e.g. gambling
Random Ratio (RR) is a type of VR schedule in which each response has an equal probability of reinforcement
Fixed-Interval (FI)
The first response after a fixed amount of time has elapsed is reinforced
As on FR schedules, there is a PRP, but after this pause the subject usually starts by responding quite slowly (unlike the abrupt switch to rapid responding on an FR schedule)
FI Scallop: as the interval progresses, the subject responds more and more rapidly, and just before reinforcement, the response rate is quite rapid
e.g amount of time studying for midterms and finals
Variable-Interval (VI)
The amount of time that must pass before reinforcer is stored varies unpredictably from reinforcer to reinforcer
VI schedules typically produce a steady, moderate response rate
e.g. checking for mail: delivery of mail is unpredictable; if the mail has been delivered, only one response is required to collective; the mail has not been delivered, no amount of responding will bring it forth – – resulting behavior is moderate and study: most people check the mail every day, but usually only once a day
Partial Reinforcement Effect
Extinction is more rapid after CRF than after a schedule of intermittent reinforcement
With CRF, every response has been reinforced, so the change to extinction is easy to discriminate, so does not take long for responding to disappear
Differential Reinforcement of Low Rates (DRL)
A response is reinforced if and only if a certain amount of time has elapsed since the previous response
DRL schedules produce very low rates of responding
E.g. under a DRL 10 second schedule, every response that occurs after a pause of at least 10 seconds is reinforced
If a response occurs after 9.5 seconds, this not only fails to produce reinforcement but it resets the 10 second clock to zero, so that now 10 more seconds must elapse before response can be reinforced
Factors Affecting Performance On Reinforcement Schedules
Quality
Rate of presentation
Delay
Response effort
Amount of reinforcement
Level of motivation
Behavioral Momentum
Behaviors associated with higher rates of reinforcement are harder to disrupt
Rule-Governed Behavior
Because people have language, they can be given verbal instructions or rules to follow
These rules may not have anything to do with the prevailing reinforcement contingencies
The discrepancy between animal and human performance on reinforcement schedules may occur because people are capable of both contingency-shaped behavior and rule-governed behavior
Reinforcement History
Prior experience with a particular reinforcement schedule can alter how subjects, both animal and human, later perform on another schedule
Behavior analysts seek to obtain a better understanding of reinforcement history effects because of the important influence they may have on treatment programs
Comparisons of VR and VI Response Rates:
Inter-Response Time Reinforcement Theory (IRT)
Subjects usually respond faster on the VR schedule
The molecular theory of inter-response time reinforcement theory (IRT) suggests that response rates are slower on VI schedules because long IRT’s (long pauses between responses) are more frequently reinforced on VI schedules
Short IRT’s are selectively strengthened on a VR schedule simply because they were reinforced more frequently
Response – Reinforcer Correlation Theory
Molar theory that emphasizes a more global relationship between responses and reinforcement
