Listener Responses as a Collaborative Process: The Role of Gaze

Abstract

The authors examined precisely when and how listeners insert their responses into a speaker's narrative. A collaborative theory would predict a relationship between the speaker's acts and the listener's responses, and the authors proposed that speaker gaze coordinated this collaboration. The listener typically looks more at the speaker than the reverse, but at key points while speaking the speaker seeks a response by looking at the listener, creating a brief period of mutual gaze called here a gaze window. The listener was very likely to respond with "mhm," a nod, or other reaction during this period, after which the speaker quickly looked away and continued speaking. This model was tested with 9 dyads in which 1 person was telling a close-call story to the other. The results confirmed the model for each dyad, demonstrating both collaboration in dialogue at the microlevel and a high degree of integration and coordination of audible and visible acts, in this case, speech and gaze.While listening to someone tell a story, listeners regularly make brief appropriate responses (sometimes called "back channels," after Yngve, 1970). The goal of the research reported here was to understand the timing of these responses, that is, why listeners respond when they do. Is this a haphazard event, is it determined by characteristics of the listener, or is it related to actions of the speaker (and if so, how)? We have pursued these questions first through an inductive qualitative analysis, then by statistical tests of the hypotheses we had developed. The data were videotapes of unacquainted dyads in which one person told a close-call story to the other.

From social behaviour to brain synchronization: Review and perspectives in hyperscanning

“…Gaze direction subtends referential communication and synchronisation during interindividual exchanges. For example, during conversation, gaze contributes to speech understanding as well as to the organization of turn-taking with eye contact enabling the converses to be coordinated and synchronized [13].…”

Continuous interaction with a virtual human

“…Such work mostly concerns gaze behavior. In a detailed study, Bavelas et al [4] conclude that 83% of Listener Responses in their corpus occur during mutual gaze, confirming earlier intuitions of Kendon [28] and Duncan Jr. [15]. Head movements also have been associated with eliciting Listener Responses [26].…”

“…The two dialogs labeled as q3ec1 and q3ec5 were discarded due to a buzz in the speech signal 4. Anderson et al[2] structure a dialogue into three levels: transactions, that accomplish a major subtask in the dialogue such as getting from one waypoint to the next; conversational games that fulfill a purpose within the transactions such as getting a question answered, getting something clarified, consisting of initiations followed by responses; and dialogue moves, which are the various types of initiations and responses that make up a conversational game.…”

The importance of gaze and gesture in interactive multimodal explanation

“…In particular, gaze (e.g. Bavelas et al 2002), and gesture (e.g. Cosnier 2000;McNeil 2000) have been studied as playing an integral role in human interaction.…”