Conversation Clock

Contacts:

Tony Bergstrom

Abstract:

Tables are places to cooperate and discuss amongst others. Conversation Clock seeks to augment the developing dialog by displaying a representation of conversation to all present. The table demonstrates turn-taking, domination, interruption, and activity throughout a conversation.

Conversation Clock

Motivation:

Conversation is an ephemeral aspect of social interaction. Past comments, utterances, and interactions help to lead the conversation, but the present moment is the most influential. Conversation Clock allows inspection of past conversation to be examined in the present. The visual nature allows conversation to continue without interruptions while providing reminders of past interaction and meta-cues of the entire history.

Group interaction has a different set of social norms than one on one interaction. Work done with Visiphone looked at remotely mediate two person conversations. It showed that a visualization allows you to "see things you know, but do not realize you know." Visiphone directly influenced the interaction between two individuals without forcing a change. How does this interaction change when the individuals are no longer remote? Does a visualization provide the same influence Conversation Clock examines the larger and more complex interaction environments of co-located individuals.

Approach:

Local interaction often shares the common physical object of the table. It is a convenient object to sit around and collaborate. Tables have features that support this interaction - participants can face each other and set their belongs down while not feeling fully exposed to a group.

The conversation clock utilized a round table with seats for four participates. Each seat has a personal microphone to monitor and associate audio input of individuals. The visualization is projected into the center of the table and is easily seen by all. The visualization itself is designed to be without a necessary orientation. This assures that all individuals are seeing the same image.

The conversation clock utilized a round table with seats for four participates. Each seat has a personal microphone to monitor and associate audio input of individuals. The visualization is projected into the center of the table and is easily seen by all. The visualization itself is designed to be without a necessary orientation. This assures that all individuals are seeing the same image.

Conversation is not just speech. People fidget. While our table does not sense visual input, the microphones, by virtue of their table mounting, can detect non verbal activity at the table. This provides a fuller picture of activity.

What does it show?

The Conversation Clock is a visualization of tick marks along concentric rings. Just like a clock, each ring contains a single minute of tick marks. Upon completion of a ring, all rings shrink toward the center point creating space for a new outer ring. The resulting image resembles the rings of a tree, in which the inner rings represent earlier times in the history of the conversation. Tick marks are parallel to the gradient of the circle. Each mark is colored according to the microphone from which input was received and sized to indicate the amplitude of the associated waveform.

conversation clock

Showing the history, aspects like activity and turn taking become easy to observe. If an individual has not been speaking, their lack of aural contribution is made clear in the rings. Of course, if individual is speaking at length and dominating the conversation, one can easily observe this as well. Aspects such as interruption, silences, and argument also make visual impressions on the table.

Group interaction can take a wide variety of forms - business meetings, social games, casual conversation, team meetings, group interviews, interrogations, etc. Each interaction's social norms differ. Interpretation of individuals' contribution are left to the observers - no judgment is provided from the table. Similarly, cultural context of a conversation drastically affects what one would expect to see in a visualization. Thus, the visualization is only properly viewed with an understanding of the context in which it was made.

Video:

A conversation about ferocious pandas. MOV

Related:

Donath, J., Karahalios, K., and Viegas, F. Visiphone. Proceedings of ICAD2000.

Publications:

Tony Bergstrom and Karrie Karahalios. Seeing More: Visualizing Audio Cues. Proceedings of INTERACT 2007. PDF

Tony Bergstrom and Karrie Karahalios. Conversation Clock: Visualizing audio patterns in co-located groups. HICSS 2007. PDF

Anthony Bergstrom. Visualization of Audio Augmenting Social Interactions. Master's Thesis. University of Illinois Urbana-Champaign, December 2006. PDF

Karrie G. Karahalios and Tony Bergstrom. Visualizing audio in group table conversation. IEEE TableTop2006. PDF