ABSTRACT
The primary motivation behind a great deal of research and theorizing in Pavlovian conditioning has been to provide an account of certain “failures of contiguity.” That is, to explain why the same degree of CS–US contiguity does not necessarily produce the same degree of performance to the CS (e.g., Gibbon & Balsam, 1981; Kamin, 1969; Mackintosh, 1975; Pearce & Hall, 1980; Rescorla, 1968; Rescorla & Wagner, 1972; Wagner, 1969, 1981). For example, although repeated pairings of a CS, X, and a US typically result in the development of conditioned responding to X, if X is accompanied by another salient cue (A) on the occasion of those pairings, the degree of conditioned responding elicited by X alone will not be as great (A is said to overshadow X; Kamin, 1969; Pavlov, 1927). Moreover, A’s own relation with the US affects its ability to interfere with conditioning to X. Separate reinforced presentations of A will enhance A’s ability to interfere with conditioning to X (blocking), whereas separate nonreinforced presentations will reduce it (Wagner, 1969). Even in the absence of an explicit interfering cue, the same number of CS–US pairings will not necessarily support the same level of conditioned responding. Factors such as the time between CS–US pairings and the rate at which the US occurs in the absence of the CS also seem to have an effect (e.g., Terrace, Gibbon, Farrell, & Baldock, 1975; Rescorla, 1968, 1969).
