Training history affects magnitude of spontaneous recovery from extinctionof appetitive conditioned responding

Three experiments examined spontaneous recovery from extinction of appetiti ve conditioned responding (CR) as a function of training history. Rats firs t received reinforced and non-reinforced conditioning trials. Groups of rat s were equated for total number of reinforced trials but differed in the nu mber of non-reinforced trials, or in the order of reinforced and non-reinfo rced trials. CR was then extinguished in all groups. Subsequently, the exte nt of recovery of CR was assessed in a test session performed either 1 or 1 7 days after the last extinction session. After a 17-day delay, rats that h ad received all reinforced trials immediately prior to the first extinction session showed stronger recovery than did rats having received all reinfor ced trials at the beginning of training, or interspersed among non-reinforc ed trials. No significant spontaneous recovery was observed after a 1-day t est delay. These results, which may be of clinical relevance with respect t o relapse after therapy, are explained in terms of the training schedules g enerating differences in strength of inhibitory associations, and a relativ ely long, but not a short, test delay attenuating these associations. (C) 2 000 Elsevier Science B.V. All rights reserved.