The perception and acoustic features of Korean ditropic sentences.

Abstract

Ditropic sentences are utterances that convey either a literal or an idiomatic meaning (e.g., It broke the ice). This study investigated listener’s ability to discriminate between literal or idiomatic meanings and examined the acoustic features contributing to this distinction. Ten ditropically ambiguous Korean sentences were audio-recorded by four native speakers of Korean. Each utterance was produced twice with either a literal or idiomatic meaning. Fifteen native Korean subjects listened to a randomized presentation of these utterances singly and in pairs without other context and identified each as literal or idiomatic. Listeners successfully discriminated the intended idiomatic or literal meanings (singletons = 70.65%, pairs = 75.67%). These results were consistent with those of Van Lancker and Canter [(1981)] for English ditropic sentences. Each utterance was acoustically analyzed in terms of means and variations in fundamental frequency, duration, and intensity. Analyses of variance revealed significantly longer durations and greater variation in syllable duration for literal than idiomatic sentences, whereas idiomatic sentences were characterized by significantly greater variation in intensity than literal sentences. Some prosodic cues for Korean differed from those found previously for English [Van Lancker et al. (1981)] and French [Abdelli-Baruh et al. (2007)]. These results further understanding of use of prosody in sentential linguistic contrasts.