Dissociative masked repetition priming and word frequency effects in lexical decision and episodic recognition tasks

After encoding a study list under either intentional or shallow incidental learning instructions, subjects received either a lexical decision or an episodic recognition task in which each target was preceded by a masked prime that was either the target itself or an unrelated item. The four main findings were: (1) For lexical decisions, a facilitatory masked repetition priming effect occurred for both intentionally and incidentally studied words and nonwords and for nonstudied words, but not for nonstudied nonwords. (2) For episodic recognition, a facilitatory masked repetition priming effect occurred for both intentionally and incidentally studied words and nonwords but for neither nonstudied words nor nonwords. (3) Lexical decision performance was better for high- than low-frequency words. (4) This word-frequency effect for lexical decisions was not attenuated by masked repetition priming but was attenuated by prior study of the target, even under conditions in which episodic recognition was equivalent for the low- and high-frequency words. These results were discussed in terms of Forster's (Language and Cognitive Processes 1 (1985), 87-108) relevant-search-set hypothesis and familiarity-based processing theories of repetition priming effects in the lexical decision task.