Chapter 17 Ratio Schedules PDF

Preview

Summary

Principles of Behavior by Malott (7th) Ch:17...

Description

Chapter 17: Ratio Schedules Continuous Reinforcement (CRF) - a reinforcer follows each response -

Usually best for shaping or maintaining difficult behavior (shaping vocal responses)

Intermittent Reinforcement Schedule - reinforcement schedule in which a reinforcer follows the response only once in awhile Schedule of Reinforcement - the way reinforcement occurs because of the number of responses, time since reinforcement, time between responses, and stimulus conditions Fixed-Ratio (FR) Schedule of Reinforcement - a reinforcer is contingent on the last of a fixed number of responses -

With fairly large ratios (100 responses/reinforcer), there is usually a consistent pattern of responding - high rate of responding until reinforcer is delivered followed by a pause before responding starts again

Fixed-Ratio Responding - after a response is reinforced, no responding occurs for a period of time, then responding occurs at a high, steady rate until the next reinforcer is delivered Post-Reinforcement Pause - the pause after the consumption of the reinforcer and before the next ratio of responses begins -

Characteristic of fixed-ratio maintained behavior; length of pause is proportional to size of ratio

-

If you want to establish a high ratio requirement, you need to do it gradually; otherwise responding will extinguish

General Rule for Establishing Intermittently Reinforced Behavior - first use continuous reinforcement and gradually increase the intermittency of reinforcement as responding stabilizes at a high rate -

The higher the ratio, the more important it is to introduce it gradually through lower initial ratios

-

Extinction due to making the ratio too high too quickly is called straining the ratio

-

By gradually increasing the requirement, we can minimize # of reinforcers needed to maintain behavior

Cumulative Graph - often used to study schedules of reinforcement -

Vertical axis is cumulative frequency of responses (vs. non-cumulative graph where the vertical axis is labeled responses)

-

The post reinforcement pause is where the slope of the line is zero

Variable Ratio(VR) Schedule of Reinforcement - a reinforcer is contingent on the last of a variable number of responses Variable Ratio Responding - variable ratio schedules produce a high rate of responding, with almost no post-reinforcement pausing Example: -

VR 50; 50 designates the average number of responses required for the reinforcer

Intermittent Reinforcement - generic term that includes both fixed and variable ratio schedules 4 Reasons Why Slot Machines Are NOT VR Schedules 1. It’s loaded with learned reinforcers in addition to the silver dollars (fruits that appear in the window, one after the other/you get 2 cherries in a row - a big reinforcer) 2. The variable amount of the reinforcer you get at the end of the so-called VR a. Sometimes you get 1 silver dollar, sometimes 10, 18, etc. 3. Size of the ratio is much smaller than is typical in the Skinner box of the professional research lab (ie. VR 100; people would not play if machines had ratios of 100) 4. The emotional reaction itself is reinforcing Most of our contingencies of everyday life involve more or less continuous reinforcement & continuous punishment Free-Operant Procedure - animal is free to respond at various frequencies (1 lever press/min to 100 lever presses/min) -

If the animal can make more than 1 correct response before the reinforcer, it is probably a free-operant procedure

-

There is no S-delta after each response, so there is no inter-trial interval between each response and the next SD

-

There may or may not be an SD; then there can be several responses, with the responses being reinforced either continuously/intermittently

Discrete Trial Procedure - there is an SD, a single response, and an outcome, followed by a S-delta (intertrial interval); then the next trial starts -

After single response occurs, the SD ends and the subject immediately receives a reinforcer or goes into S-delta...