Avoidance Learning and Escape Learning

Abstract

The procedures used to study instrumental avoidance and escape learning are defined in two forms: active and passive, and are of two kinds: discrete-trial and free-operant. The general characteristics of the escape and avoidance responses have been found to differ and need not even be topographically similar. Because of this, avoidance learning is not simply a Pavlovian process involving stimulus substitution, so many theoretical accounts of avoidance learning have involved mediation by conditioned or learned fear. The most robustly supported account is the two-process theory, which postulates Pavlovian aversive conditioning of the warning signal during escape trials, and instrumental learning of the escape response (reinforced by shock termination and reduction in pain) and of the avoidance response (reinforced by termination of the warning signal and/or prevention of the impending painful stimulus). Strong support for this model comes from experiments using a transfer paradigm in which Pavlovian conditioned stimuli are superimposed on operant baselines of responding, although cognitive theories dispute that model. Recent work has utilized selective breeding for extreme differences in avoidance learning to document the genetic contribution to individual differences in this form of learning, which is now being studied at the molecular genetic level.