Skip to content
Psych Dictionary

Extinction

  • Concept · Learning theory
  • Strong evidence
  • Published October 19, 2026
  • 12 sources
  • 4 min read

In short

Extinction is the weakening of a learned response when the event that maintained it stops following: a conditioned stimulus is presented without the outcome it predicted, or a behavior no longer produces its reinforcer. It suppresses the response without erasing the original learning.

Key facts

Field
Learning theory
First described
1900s–1930s
Also known as
Extinction (psychology); Experimental extinction; Operant extinction

State of the evidence

Strong evidence

Extinction and spontaneous recovery were described systematically in Pavlov’s laboratory[1], and there is now a broad consensus, resting on renewal, reinstatement, spontaneous recovery and rapid reacquisition, that extinction reduces learned behavior without erasing the original learning[2][3]. In applied research an initial burst of responding occurred in about a quarter of 113 extinction data sets[4]. Making extinction last is less settled: a review of animal and human work describes fear extinction as relatively easy to learn but difficult to remember[5].

Evidence labels summarise the research literature; they are not recommendations. How labels are assigned

Extinction is the weakening of a learned response when the event that maintained it no longer follows. In classical conditioning, the conditioned stimulus is presented repeatedly without the unconditioned stimulus it used to predict; in operant conditioning, the behavior no longer produces its reinforcer[2]. The word names both the procedure and the decline in responding it produces. Pavlov chose it because it implied no hypothesis about the mechanism[1].

Extinction is not erasure. A response that has been extinguished can return in several predictable circumstances, which is the main evidence that extinction adds new learning rather than removing the old[2]. It is also not forgetting. Skinner distinguished the two: in forgetting, the effect of conditioning is lost as time passes, whereas extinction requires the response to occur without reinforcement; pigeons produced sizeable extinction curves six years after a response had last been reinforced[6]. Nor is it habituation, the decline of a response to a stimulus that is simply repeated, which requires no prior conditioning.

Pavlov’s extinction of conditioned reflexes

Pavlov demonstrated extinction with a dog conditioned to the sound of a metronome. Sounded for 30 seconds at two-minute intervals without food, the metronome produced progressively less saliva, and the reflex would eventually have disappeared[1]. Extinction could also continue below zero: repeating the stimulus after the response had ceased deepened the extinction further, a “silent extinction” detectable only in later tests[1].

Extinguished reflexes came back. Pavlov reported that fully established conditioned reflexes “invariably and spontaneously” returned to full strength after an interval of minutes to hours, and that a presentation of the unconditioned stimulus hastened the return[1]. He ruled out fatigue of the salivary gland or of the nerve centers, and concluded that extinction is a special form of inhibition, not the destruction of the learned connection[1]. The procedure itself is described further under classical conditioning.

Operant extinction

Skinner described operant extinction in the same terms: when reinforcement is no longer forthcoming, a response becomes less and less frequent[6]. A pigeon stops lifting its head when food is withheld; a person who gets no answer to telephone calls eventually stops calling[6]. Recorded under suitable conditions, the decline is a smooth curve that may extend over many hours, and it is much slower than conditioning[6].

The curve can be disturbed by emotion. Skinner noted that the failure of reinforcement produces not only extinction but frustration or rage: a pigeon turns away from the key, cooing and flapping its wings, then returns to it, so that extinction curves can oscillate as the emotional response builds up and subsides[6]. How long responding persists depends on the history of reinforcement. Responding reinforced only intermittently can be much more resistant to extinction than responding reinforced every time, and under some schedules of reinforcement a pigeon may emit as many as 10,000 responses before extinction is substantially complete[6].

The extinction burst

In applied work, extinction means terminating the reinforcement contingency that maintains a problem behavior[7], and it is often associated with an initial increase in that behavior, called an extinction burst[4]. Bursts are less universal than once assumed. In an analysis of 113 sets of extinction data from applied research, a burst occurred in 24% of cases: in 12% when extinction was combined with other procedures and in 36% when it was used alone[4]. In 41 data sets from the treatment of self-injurious behavior, bursts or increases in aggression occurred in nearly half of the cases, again much less often when extinction was part of a treatment package[8]. A review of basic and applied research concluded that current findings were not sufficient for a comprehensive applied technology of extinction[7].

Evidence that extinction is new learning

Reviewing the behavioral literature in 2004, Mark Bouton concluded that extinction does not destroy the original learning but generates new learning that is especially dependent on context[3]. In his analysis, extinction gives the conditioned stimulus or the action a second meaning, stored alongside the first, so that its current meaning is ambiguous and the response depends on what the present context retrieves[2]. The evidence comes from four relapse effects, and a fifth found after operant extinction:

Renewal

Extinguished responding returns when the context changes after extinction. In ABA renewal, conditioning occurs in context A, extinction in context B and testing back in A; return also occurs when testing takes place in a third context (ABC) or when conditioning and extinction share a context and testing is elsewhere (AAB)[2].

Spontaneous recovery

Extinguished responding returns when time passes after extinction, which Bouton treats as renewal outside the temporal context of extinction[2][1].

Reinstatement

Extinguished responding returns when the unconditioned stimulus is presented by itself before the conditioned stimulus is tested again; it depends on the unconditioned stimulus occurring in the test context[2][9].

Rapid reacquisition

Responding returns abruptly when pairings of the conditioned and unconditioned stimulus resume[2].

Resurgence

An extinguished operant behavior returns when an alternative behavior, reinforced while the first was extinguished, is itself put on extinction[2].

The ABC and AAB forms of renewal are essential to the conclusion that extinction is more specific to its context than the original conditioning is[2]. In the experiments that established reinstatement, a presentation of the unconditioned stimulus alone after extinction partially restored the conditioned response, even when the reinstating stimulus differed in kind from the one used in conditioning[9]. Bouton and colleagues found that the same principles apply to the extinction of instrumental behavior: lever pressing extinguished in one context returns in another, and resurgence may itself be a form of renewal[2].

Fear extinction has also become a model for translational neuroscience. Since the early 2000s, coordinated work in rodents and humans began mapping the neural circuits of extinction of conditioned fear[10].

Extinction in treatment

The clinical parallel is direct. Exposure-based treatments for anxiety are generally very effective, but relapse is not uncommon; in the laboratory, conditioned fears are easy to extinguish but recover easily[5]. Vervliet, Craske and Hermans conclude that fear extinction is relatively easy to learn but difficult to remember, and that treatment should focus more on the long-term retrieval of what was learned[5]. Exposure therapy is accordingly described by Craske and colleagues as an application of an inhibitory learning model of extinction, and the inhibitory learning approach rests on evidence that neither the decline of fear during exposure nor its final level predicts outcome[11][12].

Relapse effects also bear on operant treatments. Bouton and colleagues point out that a person reducing food-seeking or drug-seeking while learning a replacement behavior might relapse when the new behavior in turn stops being reinforced, and that contingency management treatments, which reinforce new behaviors in place of target ones, are themselves subject to relapse over time[2].

Sources12

Numbered in order of first citation.

  1. 1.Pavlov, I. P. (1927). Conditioned reflexes. Oxford University Press.
  2. 2.Bouton, M. E., Winterbauer, N. E., & Todd, T. P. (2012). Relapse processes after the extinction of instrumental learning: Renewal, resurgence, and reacquisition. Behavioural Processes, 90(1), 130–141. https://doi.org/10.1016/j.beproc.2012.03.004
  3. 3.Bouton, M. E. (2004). Context and behavioral processes in extinction. Learning & Memory, 11(5), 485–494. https://doi.org/10.1101/lm.78804
  4. 4.Lerman, D. C., & Iwata, B. A. (1995). Prevalence of the extinction burst and its attenuation during treatment. Journal of Applied Behavior Analysis, 28(1), 93–94. https://doi.org/10.1901/jaba.1995.28-93
  5. 5.Vervliet, B., Craske, M. G., & Hermans, D. (2013). Fear extinction and relapse: State of the art. Annual Review of Clinical Psychology, 9(1), 215–248. https://doi.org/10.1146/annurev-clinpsy-050212-185542
  6. 6.Skinner, B. F. (1953). Science and human behavior. Macmillan.
  7. 7.Lerman, D. C., & Iwata, B. A. (1996). Developing a technology for the use of operant extinction in clinical settings: An examination of basic and applied research. Journal of Applied Behavior Analysis, 29(3), 345–382. https://doi.org/10.1901/jaba.1996.29-345
  8. 8.Lerman, D. C., Iwata, B. A., & Wallace, M. D. (1999). Side effects of extinction: Prevalence of bursting and aggression during the treatment of self-injurious behavior. Journal of Applied Behavior Analysis, 32(1), 1–8. https://doi.org/10.1901/jaba.1999.32-1
  9. 9.Rescorla, R. A., & Heth, C. D. (1975). Reinstatement of fear to an extinguished conditioned stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 1(1), 88–96. https://doi.org/10.1037/0097-7403.1.1.88
  10. 10.Milad, M. R., & Quirk, G. J. (2012). Fear extinction as a model for translational neuroscience: Ten years of progress. Annual Review of Psychology, 63(1), 129–151. https://doi.org/10.1146/annurev.psych.121208.131631
  11. 11.Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. https://doi.org/10.1016/j.brat.2014.04.006
  12. 12.Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N., & Baker, A. (2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46(1), 5–27. https://doi.org/10.1016/j.brat.2007.10.003

Further reading

Cite this entry

  • APA 7

    Psych Dictionary. (2026, October 19). Extinction. In Psych Dictionary. https://www.psychdictionary.com/concepts/extinction

  • MLA 9

    “Extinction.” Psych Dictionary, 19 Oct. 2026, www.psychdictionary.com/concepts/extinction.

Written by Editorial team. First published October 19, 2026. Editorial standards

Educational reference material. Nothing on this site is a diagnosis, a treatment recommendation or a substitute for care from a qualified clinician.