According to current models of word recognition, the time to recognize a word would be affected by the frequency of its neighbors. The study examined the role of neighborhood frequency in spoken word recognition, using a definition of the neighborhood in terms of the last candidates of the cohort aligned from onset with the input. Two sets of disyllabic CVCV words were selected which differed by the relative frequency of occurrence of their more frequent neighbor (one set of words having no more frequent neighbor, and the other at least one). They were compared in three different tasks: lexical decision, repetition and gating. No effect of relative frequency was obtained either with the lexical decision task, or the repetition task. The gating task also failed to show reliable evidence for the influence of neighborhood frequency on both recognition points and isolation points (for which subjects' confidence is not taken into account). The results are discussed in terms of the nature of the competitor space of a spoken word.
展开▼
