Although extinction in Pavlovian learning is highly context-dependent, less research has investigated the role of context in the extinction of operant learning. This talk will fill this gap and explore a number of parallels between Pavlovian and operant extinction. Recent research has studied the renewal effect after operant extinction, in which extinguished responding returns when the context is changed. We have produced clear evidence of ABA, ABC, and AAB renewal (where the letters correspond to the contexts of conditioning, extinction, and testing, respectively), even when the learning history of the contexts is controlled. We also have demonstrated renewal in nondeprived rats working for sucrose or sweet/fatty food pellets--the rodent equivalent of junk food. The ABC and AAB renewal effects suggest that operant extinction is more context-dependent than operant conditioning. Other experiments have studied resurgence, in which a behavior that is extinguished while a second is reinforced recovers when the second behavior is extinguished. Resurgence can be viewed as another renewal effect. Contextual cues have a general role in the control of operant behavior. But the extinction of operant behavior, like the extinction of respondent behavior, is especially sensitive to the context, with a number of interesting implications for understanding behavioral inhibition, lapse, and relapse.
Review Mark E. Bouton’s biographical statement.