Dissociative effects in different prime domains.

These experiments were performed to examine the effects of different types of primes across variations in prime set size, prime strength, and prior study in verification and lexical decision tasks. The primes consisted of taxonomic category names, associates, or rhymes that defined either small or large sets of related concepts, and they were either strongly or weakly related to their targets. Targets either were or were not studied prior to the priming task. The results indicated that, for taxonomic primes, shorter decision latencies were obtained when set size was smaller and when the target was studied before the priming task. In contrast, for rhymes, neither set size nor prior study had reliable effects. For all three types of primes, decision latencies were faster for stronger than for weaker prime-to-target relationships. These findings are contrasted with the results of manipulating these variables in episodic tasks such as cued recall, and they are interpreted in the context of a components-of-processing approach.