Abstract
The relationship between syntactic and prosodic cues to sentence boundaries was examined as a function of language dominance. Groups of English-dominant and Hebrew-dominant subjects were timed on their responses to tones at the ends of complete and incomplete sentences in both languages, and they identified each utterance as either “finished” or “unfinished”. The truncated sentences were lexically and syntactilly well-formed, through prosodically incomplete. Reaction time was also measured in a second experiment, in which explicit attention was not called to intonation. The results indicated that listeners perceive acoustic cues to sentence boundaries, though the tendency to reprocess these cues in terms of syntax increases at less attention is directed to prosodic features. In experiment II, both the subjects who were aware of the unfinished intonation and those who reported noticing unusual about the sentences responsed significantly more slowly to incomplete utterances. These findings indicate that post-sentence identification tasks may fail to detect that objective acoustic information is in fact initially perceived. The two dominance groups responded similarly to the same stimuli in Experiment I, producing no errors in identifying unfinished sentences in English and the largest number of incorrect responses on unfinished in Hebrew. These results provide support for perceptual effects of differences in segmental lenthening between English and Hebrew uncovered in an earlier study.

This publication has 18 references indexed in Scilit: