11 - Schedules of Reinforcement

Flashcards about Schedules of Reinforcement

Discrete-trial tasks

Behavior is segmented into chunks (trials), with well-defined beginning and end points.

Free-operant task

Behavior is continuous and progresses at a rate according to the subject.

Reinforcement schedule

A rule that determines when subjects receive reinforcement.

Cumulative record

Plots total number of instrumental responses as a function of time. Slope of cumulative record = instantaneous rate of responding. Tick marks might be used to indicate reinforcer deliveries.

Continuous reinforcement ('CR')

Every response is reinforced.

Partial (intermittent) reinforcement

Reinforcement occurs only when particular conditions are met.

Ratio schedules

Reinforcement depends on the number of responses subjects make

Interval schedules

Reinforcement depends on time since the last reinforcement

Fixed

Threshold for reinforcement is constant and unchanging

Variable

Threshold for reinforcement has a consistent average value but varies from trial to trial.

Fixed ratio (FR)

Reinforcement delivered after a fixed number of responses. Results in a pause and run pattern. Example: freelance writer receiving payment after completing a certain number of articles.

Variable ratio (VR)

Ratio requirement has a consistent average value, but varies for each reinforcer. Promotes steadier responding. Example: slot machines paying out after a random number of plays.

Fixed interval (FI)

Reinforcement is delivered for the first response after a fixed inter-reinforcer time interval has elapsed; leads to FI scallop. Example: a worker receiving a paycheck every two weeks.

Variable interval (VI)

Reinforcement is delivered for the first response after a variable inter-reinforcer interval has elapsed; produces very stable, steady responding. Example: checking email and getting a new message on average every X minutes.

Progressive ratio schedule

Amount of responses required progressively increases with each reinforcer that subjects earn; useful for measuring motivation.

Partial Reinforcement Extinction Effect (PREE)

The group that was partially reinforced during learning showed slower extinction. Subjects know they won’t be reinforced every time so when they aren’t reinforced, they aren’t as deterred.

Capaldi’s Sequential model

Memory of the sequence of previous trials becomes part of the stimulus that controls responding.

Amsel’s Frustration model

Emotional response on previous trial becomes part of the stimulus that controls responding, specifically, not getting reward when it is expected causes frustration and therefore strengthens the association.

Breakpoint

Number of reinforcers a subject earns before abandoning the current behavior in a schedule of reinforcement.