Draft Do Not Quote Submitted to Language and Speech Lexical Eeects in Phoneme Monitoring: Time-course versus Attentional Accounts Attention and Phoneme Monitoring 1

Pennsylvania. We would like to thank Juan Segui and the members of the Labo-ratoire de Sciences Cognitives et Psycholinguistique for their help, comments, and fruitful discussion, Jean-Luc Aucouturier for his technical help with speech editing and compression routines, and Pierre Clergeot and Denis Acker for access to a large number of subjects and testing facilities at the Ecole Sainte Genevieve, Ver-sailles. This research was carried out with the help of CNET (Convention 837BD28 00790 9245 LAA/TSS/CMC), CNRS (ATP \Aspects Cognitifs et Neurobiologiques du Langage"), the European Science Foundation (TW 86/17), the Minist ere de la Recherche et de l'Enseignement Sup erieur d ecision n o Abstract Under what conditions do lexical factors innuence phoneme detection times? Experiment 1 measured subjects' latencies to detect initial phonemes in monosyllabic and disyllabic words that were preceded by a semantically related or unrelated word. One group of subjects was instructed to pay attention to the semantic relations between words, and a second group was asked to focus on acoustic-phonetic information. A signiicant priming eeect was found, only for monosyllabic words, and only in the rst group. In Experiment 2, previously observed frequency eeects (Dupoux and Mehler, 1990) disappeared when the detection task was biased towards acoustic-phonetic information. In Experiment 3, two student populations were tested with exactly the same instruction set and showed markedly diierent results: One group showed a consistent lexical superiority eeect on monosyllabic items while the other group showed no such eeect. Taken together, these results suggest that the presence or absence of lexical eeects is extremely sensitive to attentional parameters that can be aaected by explicit biasing instructions and/or individual diierences. Importantly , these eeects cannot be accounted for in terms of mean reaction time diierences (where slow reaction times would be expected to lead to stronger lexical innuences than fast ones). The results reported here are consistent with the view that phoneme detection can be carried out using either of two quite diierent routes. Implications for current models of lexical and prelexical processing are discussed.