ASA 127th Meeting M.I.T. 1994 June 6-10

4pSP9. Attentional allocation during speech perception: Evidence from phoneme monitoring.

Lee H. Wurm

Arthur G. Samuel

Dept. of Psychol., State Univ. of New York at Stony Brook, Stony Brook, NY 11794

Previous work [Frauenfelder et al., JEP:HPP 16, 77--91 (1990)] tested a critical prediction of the TRACE model of speech perception [J. L. McClelland and J. L. Elman, Cog. Psychol. 18, 1--86 (1986)]. Frauenfelder et al. found no evidence of inhibitory lexical effects on performance of a phoneme monitoring task, and concluded that the TRACE model was inaccurate. However, recent experiments [L. H. Wurm and A. G. Samuel, J. Acoust. Soc. Am. 94, 1883(A) (1993)] showed that the earlier conceptualization of the problem had several limitations, and that the question of inhibitory lexical effects was still unanswered. This experiment extends that work, examining specifically the role of attentional allocation during the phoneme monitoring task. In dual-task paradigm [e.g., M. I. Posner and S. J. Boies, Psychol. Rev. 78, 391--408 (1971)], stimuli were presented dichotically to subjects, who simultaneously monitored for specified phoneme targets in speech and for frequency modulations in a pure tone. Reaction times to frequency modulations were measured as a function of their locations relative to phoneme changes in nonword stimuli. The results of this line of research clarify the role played by attentional processes during speech perception, and whether indirect lexical inhibition should be included in models of lexical access. [Work supported by AFOSR.]