When words fail: Collaborative gestures during clarification dialogues

Patrick G T Healey, Nicol Plant, Christine Howes, Mary Lavelle

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

13 Citations (Scopus)

Abstract

Research on co-speech gestures has primarily focussed on speakers. However, in conversation non-speaking addressees can also gesture. Often this is to provide concurrent feedback such as backchannels but sometimes it involves gestures that relate to the specific content of the speaker's turn. We hypothesise that non-speakers should contribute most actively during clarification sequences i.e. at the moments when mutual-understanding is threatened. We test this hypothesis using a corpus of story-telling dialogues, captured using video and motion capture. The results show that during clarification sequences speaker and non-speaker behaviours tend to merge. Non-speakers in particular move their hands faster and produce more than twice as many content-specific specific gestures in overlap with speaker turns. These results underline the collaborative nature of conversation, the strategic importance of nonverbal resources for sustaining mutual-understanding and the critical role of clarification and repair in successful communication.

Original languageEnglish
Title of host publicationAAAI Spring Symposium - Technical Report
PublisherAI Access Foundation
Pages23-29
Number of pages7
VolumeSS-15-07
ISBN (Print)9781577357117
Publication statusPublished - 2015
Event2015 AAAI Spring Symposium - Palo Alto, United States
Duration: 23 Mar 201525 Mar 2015

Conference

Conference2015 AAAI Spring Symposium
Country/TerritoryUnited States
CityPalo Alto
Period23/03/201525/03/2015

Fingerprint

Dive into the research topics of 'When words fail: Collaborative gestures during clarification dialogues'. Together they form a unique fingerprint.

Cite this