L701: Accommodation to Speech Production in Task-Related Dialogue

Abstract

Using the experimental setting of a joint construction task where visual contact between the two members of dyad is blocked, the communicative channels of eye-track and mouse-track are studied in order to ascertain whether their presence reduces the need for distinct introductory mentions of a referent; a phenomenon which has been found in the visual communication channel, as reported by Anderson et al (1997). A series of multiple regressions on the absolute durations of introductory and second identical mentions, and on the duration differences between corresponding first and second tokens reject this hypothesis, showing that the presence of one extra communicative channel reduces the overall length of tokens, two extra channels increases the overall length of tokens, but showing no effect of communication channel on the difference in duration between first and second tokens. The results are discussed and explanations for these findings are put forward.