Extinction (psychology)¶
Reduction of a conditioned response after its sustaining outcome is withheld, through nonreinforcement or unpaired cue exposure, often as context-sensitive new learning.
Core Idea¶
Extinction modifies a learned contingency through experience. In classical conditioning the predictive cue occurs without its outcome; in operant conditioning the response no longer produces its reinforcer. Response strength usually declines, sometimes after variability or an extinction burst.
Decline does not necessarily erase acquisition. Spontaneous recovery, renewal, reinstatement, and rapid reacquisition indicate persistent old learning whose expression competes with extinction learning. Studies must separate fatigue, habituation, satiation, punishment, and forgetting.
Scope of Application¶
- Learning experiments. Tests acquisition, nonreinforcement, context, and recovery.
- Behavior analysis. Changes operant contingencies and measures response patterns.
- Clinical theory. Informs professionally supervised exposure approaches.
- Animal cognition. Studies prediction and relapse phenomena.
- Adaptive systems. Models updating when expected outcomes cease.
Clarity¶
Specify acquisition history, cue or response, maintaining outcome, extinction schedule, response measure, context, delay, and recovery tests. Clinical applications remain professionally supervised rather than generic advice. Inclusion test: Require prior conditioning, withdrawal of the maintaining contingency, repeated response decline, and controls for fatigue, habituation, satiation, forgetting, punishment, and context change. Exclusion test: Exclude species extinction, forgetting without contingency manipulation, habituation to repeated unconditioned stimulation, and response suppression by punishment. Nearest boundary: Habituation reduces response to repeated stimulation without an acquired cue–outcome or response–reinforcer contingency; extinction changes expression after reinforcement or prediction stops. Exit condition: The identity changes when decline results from added punishment or sensory adaptation rather than omitted outcome. Common misclassifications: It is not biological species extinction. It is not punishment. It is not habituation to an unconditioned stimulus. It is not proof that the original association was erased. Nearest named distinctions: Habituation: Habituation reduces response to repeated stimulation without withdrawing an acquired outcome contingency. Punishment: Punishment adds a consequence; extinction removes reinforcement. Forgetting: Forgetting can occur with time alone; extinction requires nonreinforced experience. Exposure Therapy: Exposure therapy is a professional intervention informed by extinction, not the laboratory process itself.
Manages Complexity¶
The abstraction distinguishes learning history from current performance. It decomposes response loss into contingency removal, new experience, measurement, context, and retrieval, explaining why behavior can disappear in training yet return later.
Abstract Reasoning¶
- Establish acquisition through a defined contingency.
- Withhold the expected outcome while retaining cue or response opportunity.
- Measure across repeated nonreinforced trials.
- Control alternative causes of decline.
- Vary context or delay to test recovery.
- Interpret response loss as changed expression unless erasure is independently supported.
Knowledge Transfer¶
The transferable cargo is contingency-specific updating after expected outcomes cease. It transfers across conditioning systems when histories and outcomes are mapped; it stops at any generic decrease in behavior.
Neighborhood in Abstraction Space¶
Extinction (psychology) sits in a crowded region of the domain-specific corpus (39th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.
Family — Named Cognitive & Behavioral Effects (32 abstractions)
Nearest neighbors
- Behavioral Effect — 0.89
- De Novo Protein Synthesis Theory of Memory Formation — 0.88
- Electroneuronography — 0.88
- Frustration–Aggression Hypothesis — 0.87
- Neglect of Probability — 0.87
Computed from structural-signature embeddings · 2026-10-08