Operant conditioning
Operant conditioning is learning in which the consequences of an action affect how likely it is to be repeated. It is also called instrumental conditioning. A response that changes the environment can lead to an outcome, and experience of that relationship influences later behaviour. The approach is associated with Edward Thorndike and B. F. Skinner and is an important part of the experimental study of learning.[1]
Consequences and learning
A reinforcer increases the subsequent occurrence of a response; a punisher reduces it. These terms describe effects on behaviour, rather than whether an outcome seems pleasant to an observer. Positive means that something is presented, while negative means that something is removed. Negative reinforcement therefore strengthens a response through removal of a stimulus; it is distinct from punishment.[2]
Researchers can arrange these consequences in an operant conditioning chamber, recording responses such as lever presses or key pecks. Shaping establishes a response by reinforcing successive approximations to it. A new action can thus be acquired gradually, without requiring the complete response at the beginning of training.[2]
Schedules and context
Reinforcement may follow every response or only some responses. Ratio schedules depend on the number of responses; interval schedules depend on elapsed time. Either requirement can be fixed or variable. The resulting response patterns differ, and intermittent reinforcement can produce behaviour that persists when reinforcement is discontinued.[2]
The surrounding conditions also matter. A discriminative stimulus signals when a response can produce its outcome. It sets the occasion for action without necessarily eliciting a reflex. Choices between actions depend on the reinforcement available for alternatives as well as the consequences of the particular response being studied.[1]
Stopping reinforcement can produce extinction, but a response may return after time has passed or the context changes. This is one reason that extinction is not simply equivalent to erasing earlier learning. Outside laboratory experiments, operant and classical conditioning commonly operate together: an outcome can become associated both with an action and with the circumstances in which it occurred.[1]
References
- ↑ 1.0 1.1 1.2 Mark E. Bouton. Conditioning and Learning. Noba Project.
- ↑ 2.0 2.1 2.2 Rose M. Spielman, William J. Jenkins and Marilyn D. Lovett. Operant Conditioning. Psychology 2e, OpenStax, 2020.
Attribution
Adapted with reference to English Wikipedia, revision 1374182721 (contributors and history), under CC BY-SA 4.0. The prose has been rewritten and checked against the cited sources; changes have been made.