Katherine, BinderJoseph, CohenFrancine, DeutschSanders, LisaJiang, Yue2011-05-232011-05-232011-05-23http://hdl.handle.net/10166/873Previous research shows that listeners direct temporally selective attention to the initial portions of words in continuous speech. This processing strategy is useful since speech signals change too rapidly for listeners to form representations of every detail, and the initial portions of words are particularly helpful in auditory lexical access. However, very little is known about the cues that direct temporal attention during speech processing. In the current study, event-related brain potentials (ERPs) elicited by word and syllable onsets in artificial languages were compared and lexical stress and the regularity of lexical duration were manipulated as potential attention cues. The results suggest that listeners can use stress or rhythmic regularity to direct attention to upcoming word onsets. The study provides information about how listeners preferentially process the most relevant segments in rapidly changing speech streams, an important step in extracting meaning from spoken language.en-USAttribution-NonCommercial-ShareAlike 3.0 Unportedspeech perceptionERPtemporally selective attentionrhythmstressThesispublic