Research project: Context and learning

If a response to one stimulus is first learned and then extinguished (i.e. trained so that the stimulus no longer elicits the response) it is frequently possible for the original response to reappear. This response-recovery is of theoretical interest, because it shows that old learning does not get lost during extinction, and of applied interest because it seems to be involved in relapses that occur in a variety of clinical behavioural problems. The current project deals with both of these aspects. In terms of theory, we have obtained evidence that contextual changes mediate the expression of learning, selecting which of the available responses to an ambiguous stimulus will be elicited. Working with Byron Nelson (University of the Basque Country) we have found evidence that inhibitory mechanisms operate during extinction, so that the extinction context may suppress the originally learned responses. In terms of application, we have found that extinction in multiple contexts can reduce response-recovery, and this suggests that cue-exposure therapy for addiction might be improved by ensuring that cue-exposure is carried out in more than one setting.