Previous research has shown that recognition of a word such as king can be facilitated by prior presentation of a homophone (thrown) of the word?s semantic associate (throne). The proposed experiments extend this research to implicit word retrieval by using a novel application of the traditional word-stem completion paradigm (e.g., thrown-k__). Word stems will be presented with no prior study phase, and participants will be asked to complete the stem as quickly as possible with the first word that comes to mind. This type of word-stem completion paradigm allows phonological priming to be measured without the threat of contamination from explicit retrieval processes. Further, this paradigm will measure associative phonological priming, i.e., the extent to which phonological priming of word-stem retrieval occurs across preexisting semantic associations versus newly-formed associations. In contrast to new associations, preexisting associations are likely to be stronger because they have been frequently activated over the course of time. In addition, the extent to which phonological priming occurs for young adults will be compared to that of elderly adults, who are thought to have deficits in the formation of new associations. To offset these declines, it is important to establish phonology?s role in word retrieval in order to isolate specific language and memory difficulties in old age. Predictions will be discussed within a theory of language perception and production (Node Structure Theory, MacKay, 1987).
