Operant Learning
Operant Learning
Animal Intelligence and the Birth of Behaviorism
Edward Thorndike: Graduate student mentored by William James, known for "Functionalism".
Aim: Study ordinary animal intelligence, in contrast to extraordinary cases like "Clever Hans".
Key Point: Thorndike published his dissertation in 1898; Pavlov's work was published in English in 1927, leading to a lack of awareness between them during their early careers.
Thorndike’s Law of Effect
Definition: Organisms will repeat behaviors that lead to a “satisfying state of affairs” and resist behaviors that lead to an “annoying state of affairs”.
Key Insight: Behavior occurs as a function of its consequences, termed The Law of Effect.
Four Key Elements of The Law of Effect:
Context: The setting in which the behavior occurs.
Behavior: The specific action taken by the organism.
Change in Environment: The alteration in the environment following the behavior.
Change in Behavior: The change in behavior as a result of this consequence.
Thorndike discovered that behavior systematically strengthens or weakens due to its consequences.
Experiment: Thorndike’s “Puzzle Box” illustrated these principles.
B.F. Skinner
Beliefs: Considered free will an illusion; human action is dependent on consequences of prior actions.
Defined: The Principle of Reinforcement articulates how consequences affect behavior.
Operant Learning: Described as behavior "operating on the environment".
Also referred to as Instrumental Learning because behavior is instrumental in producing resulting events.
Other Terms:
Response Learning
R-S Learning
Reward Learning (with a clarification that people are rewarded, while behavior is reinforced).
The Skinner Box
Developed as a research tool to study Operant Learning in a controlled environment.
Components:
Operant Conditioning Chamber (Skinner Box): Used to study behaviors in response to stimuli.
Cumulative Recorder: Not originally used by Pavlov, but essential in Skinner's research.
Types of Operant Learning Studied:
Positive Reinforcement
Negative Reinforcement
Positive Punishment
Negative Punishment
The Contingency Square
Strength of Behavior
Increases (Strengthened) under:
Positive Reinforcement (Stimulus is Presented)
Negative Reinforcement (Stimulus is Removed)
Decreases (Weakened) under:
Positive Punishment (Stimulus is Presented)
Negative Punishment (Stimulus is Removed)
Summary of Operant Conditioning:
Positive Reinforcement: Giving something rewarding; increases behavior.
Negative Reinforcement: Removing something negative; increases behavior.
Positive Punishment: Adding something unpleasant; decreases behavior.
Negative Punishment: Removing something pleasant; decreases behavior.
Examples of Operant Conditioning
Positive Reinforcement: Giving a reward for good behavior (e.g., treats).
Negative Reinforcement: Lifting a curfew for good behavior.
Positive Punishment: Inflicting a spanking for poor behavior.
Negative Punishment: Taking away a favorite toy for poor behavior.
Distinction between Operant and Classical Conditioning:
Nature of response:
Classical: Involuntary (reflexive).
Operant: Voluntary (but can also be involuntary).
Timing of Stimulus:
Classical: Precedes response.
Operant: Follows desired response.
Role of learner:
Classical: Passive.
Operant: Active.
Forms of Negative Reinforcement
Positive Reinforcement: Increases behavior by adding a reward after successful behavior (e.g., treat for sitting).
Positive Punishment: Decreases behavior by presenting aversive stimuli (e.g., spanking for cursing).
Negative Reinforcement: Decreases behavior by removing pleasant stimuli (e.g., sending child to room for cursing).
Escape: Removes noxious stimuli following correct behavior (e.g., turning off an alarm).
Active Avoidance: Engaging in behaviors to avoid negative outcomes (e.g., studying to avoid poor grades).
Operant Learning and Reinforcement
Definition: Operant (or Instrumental) Learning occurs when consequences strengthen behavior (reinforcers) or weaken behavior (punishers).
Three Characteristics of Reinforcement:
A behavior must have a consequence (SR or SP; or S+ / S- for clarity).
The behavior must strengthen after the consequence.
The increase in strength must arise from the consequence.
Measuring Strengthening of Behavior
Operant Learning measured as the Frequency or Probability of a behavior.
Reinforcement effects beyond mere frequency:
Persistence of behavior despite cessation of reinforcement.
Resistance to adverse consequences.
Increased effort required to obtain reinforcer.
Ability to engage in competing reinforced behaviors.
Four Kinds of Reinforcers
Primary or Unconditioned Reinforcers: Reinforcers that fulfill biological needs without prior conditioning (e.g., food).
Secondary or Conditioned Reinforcers: Reinforcers that derive power from association with primary reinforcers (e.g., money).
Natural or Automatic Reinforcers: Reinforcing events that follow behavior naturally (e.g., the satisfaction from exercise).
Contrived Reinforcers: Reinforcers arranged to modify behavior (e.g., rewards for tasks).
Variables Affecting Operant Learning
Contingency: Correlation between behavior and consequence (stronger impact with reliable reinforcement).
Contiguity: Delay between behavior and reinforcing consequence (shorter intervals generally yield faster learning).
Reinforcer Characteristics:
Size of the reinforcer: More significant rewards increase motivation.
Quality of the reinforcer: Fresh vs. stale rewards influence desire.
Characteristics of Behavior and Motivating Operations:
Behavior's characteristics can affect reinforcement efficiency.
Motivating Operations influence the effectiveness of reinforcers.
Establishing Operation: Increases reinforcer effectiveness.
Abolishing Operation: Decreases reinforcer effectiveness.
Other Miscellaneous Variables
Previous learning experiences impact the effectiveness of current behavioral reinforcers or punishers.
Competing Contingencies: Various reinforcement contingencies can exist simultaneously, giving individuals multiple choices.
Next Time…
Theoretical and Neurobiological Explanations of Operant Learning.