COMPETITION IN SPOKEN WORD RECOGNITION - SPOTTING WORDS IN OTHER WORDS

Although word boundaries are rarely clearly marked, listeners can rapi dly recognize the individual words of spoken sentences. Some theories explain this in terms of competition between multiply activated lexica l hypotheses; others invoke sensitivity to prosodic structure. We desc ribe a connectionist model, SHORTLIST, in which recognition by activat ion and competition is successful with a realistically sized lexicon. Three experiments are then reported in which listeners detected real w ords embedded in nonsense strings, some of which were themselves the o nsets of longer words. Effects both of competition between words and o f prosodic structure were observed, suggesting that activation and com petition alone are not sufficient to explain word recognition in conti nuous speech. However, the results can be accounted for by a version o f SHORTLIST that is sensitive to prosodic structure.