Preventing recovery from extinction and relapse: A product of current retrieval cues and memory strengths

When a cue (X) and an outcome are paired, X comes to elicit responding similar to that elicited by the outcome. This potential is lessened if X is later presented by itself (i.e., experimental extinction). Extinction is interesting both for theoretical reasons and because it is a model for exposure therapy, which is used to treat numerous psychological disorders. Conventionally, extinction has been viewed as an isolated phenomenon unrelated to other response-attenuating treatments, whereas it is actually but one of several instances of associative interference that reduce conditioned responding. Although much has been learned about extinction from direct study, the larger associative-interference literature provides additional insight. Here we view extinction as new learning that is inconsistent with previously acquired knowledge (i.e., retroactive outcome interference). We provide an account of whether the acquisition or extinction memory will be expressed depending on the relative strengths of these two memories and the relative efficacies of their respective retrieval cues.