DIFFERENTIAL SENSITIVITY OF PERCEPTUAL IDENTIFICATION FOR WORDS AND PSEUDOWORDS TO TEST EXPECTATIONS - IMPLICATIONS FOR THE LOCUS OF WORD-FREQUENCY EFFECTS

Four experiments using a study-test paradigm provide evidence of differences in the nature of perceptual enhancement effects for words compared with results for meaningless pseudowords. Subjects studied words (Experiments 1, 2, and 4) or pseudowords (Experiments 3 and 4) presented from 1 to 6 times in a list, then performed perceptual identification tests of studied and nonstudied items. Better identification of studied than of nonstudied items (i.e., a priming effect) was found for words and nonwords, but for words the function relating enhancement to the amount of prior exposure (i.e., a repetition effect) depended on whether subjects expected a later recall test. The results support dual-process accounts of word identification that assume a flexible use of either lexical code activation or episodic trace retrieval. A framework for understanding priming, repetition, and word frequency effects is discussed.