Multimodal support for social dynamics in co-located meetings

Abstract

In this paper, we present a system that employs perceptual technologies (i.e. technologies that perceive the context through sensors such as cameras and microphone) to provide feedback about people's behaviour in small group meetings. The system measures aspects of behaviour that are relevant to the social dynamics of the meeting, speaking time and gaze behaviour, and provides visual feedback about these aspects to the meeting participants through a peripheral display. We describe the system properties and the perceptual components. Also, we present a study aimed at evaluating the effect of such a system on meeting behaviour. Groups of participants, amounting to 82 participants in all, discussed topics of general interest. Analysis of the data of 58 participants showed that feedback influenced the behaviour of the participants in such a way that it made over-participators speak less and underparticipators speak more. Analysis of the micro-patterns of six participants indicated that feedback on gaze behaviour had little effect on the interaction dynamics. We conclude that perceptual technologies can be used to build services that may help people to improve their meeting skills and we consider some ways in which such systems may be deployed in meetings.

Activation of a Co-located Meeting by Visualizing Conversation and Facilitating Participant's Behavior

“…It motivates the participant to talk, changes the direction of the conversation, and designs the field of conversation. Terken et al [4] proposed a tabletop system that represents amount of speaking time and listening time as circles projected in front of the participants. The size of the circles increases when a person is speaking and listening.…”

“…[2,3,5]), however, our system provides not only visualization but also scores for enhancing conversation with gamification and facilitation messages for suggesting participants' speaking/listening behavior. Terken et al [4] proposed a visual feedback based on participants' speaking time and listening time. Their system has only positive feedback that the size of projected circles increases along with the increase of the amount of speaking/ listening time.…”

“…Bachour [15] pointed out and verified that member's unbalanced paticipation leads decreasing of motivation in collaborative learning. DiMicco [5], Terken [4] and Schiavo [11] all together mentioned that balancing paticipant's total utterlance time is important for collaborative decision making, which is an opinion-convergent type discussion, not only for brainstorming which is an opinion-divergent one. Recently facilitator, who coordinates an active discussion, becomes popular and has an important role in discussion management.…”

“…(3), applying Gini coefficient as a measure of equality. Similar metrics inspired by the Gini coefficient have already been adopted for measuring participation equality in group conversations [4,11].…”

“…On the other hand, in a real communication, it sometimes ends up as one-way conversation because of someone's too much speaking, and people cannot understand a less-speaking person's thought. Considering the situations where a balanced participation is desirable, such as brainstorming and decision-making meeting, the amount of speaking time should be well controlled [4,5,11,15].…”