Extinction Delay and Reinstated Fear Generalization

After classical conditioning and extinction of fear, presentation of an unconditioned stimulus can produce a re-emergence of the extinguished fear response, a reinstatement effect. Previous research has reported that reinstatement of fear of contextual cues after a post-extinction delay could generalize over time to similar contexts.

Two experiments were designed to investigate reinstated fear to a discrete stimulus with either a post-reinstatement delay (Experiment 1) or a post-extinction delay (Experiment 2).

In Experiment 1 rat subjects were randomly assigned to four groups which received fear conditioning and extinction with a 3000 Hz tone CS. Reinstatement conditioning was then conducted with either the same or different CS, and testing for reinstated fear with the original CS was conducted after either a short or long delay following reinstatement. In Experiment 2 all procedures were the same except the long or short delay employed was between extinction and reinstatement instead of after reinstatement.

Reinstated fear increased with time to the same CS, and an increase in generalization of reinstated fear over time was observed with a different reinstatement CS (Experiment 1). When the delay employed was between extinction and reinstatement no increase in fear generalization was observed (Experiment 2).

The results indicated support for the notion that reinstated fear to a discrete stimulus may generalize over time, but that generalization only takes place when reinstatement is administered a short interval after extinction. The results have implications for clinical behavior therapy treatments of anxiety disorders.