Effects of priming on the naming accuracy of preschoolers with word-finding deficits.

Eight preschoolers with word-finding (WF) deficits and 16 controls with normal word-finding abilities (8 preschoolers and 8 adults) named 40 pictured objects under primed and unprimed conditions. Each picture could be correctly labeled with a simple noun or a compound (e.g., cane or walking stick). The primed condition involved semantic primes for both the simple and compound targets as well as a partial lexical prime for the compound targets. All participant groups decreased naming errors when given the primes. Two results indicated that the participants made use of the lexical primes. The first was a shift in form of correct responses from simple nouns in the unprimed condition to compound nouns in the primed condition. The second was an increase in errors that incorporated the lexical prime in the primed condition. There were limits to the benefit that the WF group derived from the primes. First, the primes did not enable the WF group to compensate fully for their naming problems. The gap between the error rates of the WF group and the control groups was not reduced in the primed condition. Second, the quality of errors made by the WF group did not improve in response to primes. Compared to the controls, the WF group made proportionately more errors that indicated no access to the target neighborhood (particularly "I don't know" responses) in the unprimed condition. With primes, the controls further reduced their use of these errors, but the WF group did not. When members of the control groups did make errors, they were more likely than the WF children to produce a word substitution that bore a close semantic, visual, or phonological relation to the target in both unprimed and primed conditions. These limitations on the benefit of priming for participants in the WF group suggest deficiencies in size, elaboration, or organization of their lexicons.