Operant Conditioning Notes

Also known as instrumental conditioning.
Law of Effect (Thorndike): The tendency of an organism to produce a behavior depends on the effect the behavior has on the environment.
- Puzzle Box: Thorndike used puzzle boxes to study how animals learn, observing that behaviors leading to escape (and thus reward) became more frequent.

Radical Behaviorism: Skinner believed that behavior is controlled by its consequences.
Skinner Box (Operant Chamber):
- An animal (typically hungry) is placed in the box.
- The animal presses a bar.
- The animal receives a food pellet (reinforcer).
- This increases bar-pressing behavior.

Reinforcement: A consequence of behavior that increases the probability that the behavior will occur in the future.
Punishment: A consequence of behavior that decreases the probability that the behavior will occur in the future.

Problem: One must wait until the desired behavior is exhibited before reinforcing it.
Shaping: Reinforcing closer and closer approximations of the desired response.
Successive Approximations: Responses that are increasingly similar to the desired response are rewarded.

Positive Reinforcement: Presentation of a pleasant stimulus after a behavior, which increases the probability of that behavior.
Negative Reinforcement: Removal of an unpleasant stimulus after a behavior, which increases the probability of that behavior. Note that this is not punishment; it's still reinforcement because it increases the likelihood of the behavior.
Positive Punishment: An unpleasant stimulus follows a behavior, decreasing the probability of that behavior.
Negative Punishment: Removal of a pleasant stimulus after a behavior, decreasing the probability of that behavior.

Generalization: After a behavior is reinforced in one situation, it is performed in different situations.
Discrimination: A behavior that is reinforced in one situation is not performed in a different situation.
Extinction: After the reinforcer is withdrawn, the behavior decreases.
Spontaneous Recovery: After extinction, the behavior reappears.

Continuous Reinforcement: Consequences are the same each time the behavior occurs.
Intermittent (Partial) Reinforcement: Consequences are given only some of the times the behavior occurs. Partial reinforcement is generally more effective in maintaining behavior over the long term.

Ratio Schedules: Reinforcement is given after the behavior is exhibited a certain number of times.
Interval Schedules: Reinforcement is given after a certain amount of time has passed.

Fixed-Ratio: Reinforcement is provided after a fixed number of responses.
Variable-Ratio: Reinforcement is provided after a variable number of responses. This schedule produces high, steady rates of responding because the individual doesn't know when the next reinforcement will come.
Fixed-Interval: Reinforcement is provided for the first response after a fixed amount of time has elapsed.
Variable-Interval: Reinforcement is provided for the first response after a variable amount of time has elapsed.

It may be difficult to identify which behavior is actually being punished.
The individual may come to fear the person administering the harsh punishment, which can damage relationships and create other unintended consequences.
Punishment may not eliminate existing rewards for the behavior; if the behavior is maintained by strong positive reinforcement, the punishment may not be sufficient to suppress it.
Harsh punishment can model aggression, leading the individual to exhibit aggressive behaviors themselves.

Extrinsic Motivation: Pursuit of a goal for external rewards.
Intrinsic Motivation: Pursuit of an activity for its own sake.
Overjustification Effect: Too much reward can undermine intrinsic motivation. If someone is already intrinsically motivated to do something, offering excessive external rewards can decrease their enjoyment and motivation from doing it for its own sake.