Observations on Listener Responses from Multiple Perspectives

In this paper we present three studies that investigate the individual differences in nonverbal listening behavior. Besides collecting a corpus of listener responses, we asked people to watch a video of a speaker and indicate where they would produce a listener response. Also we asked people to judge the appropriateness of listener responses that we generated using a virtual human. The combination of the multiple perspectives collected in these studies provides us with a rich data set in which different types of response opportunities are distinguishable. There are moments where there is high agreement between these multiple perspectives that a listener response is appropriate or inappropriate, moments where a listener response is controversial and moments neither a response was given nor a response was judged inappropriate (neutral). We will show that different contextual characteristics can be used to discriminate these response opportunities. Observations show relations to sentence structure, conversational structure and proximity of earlier responses.

[1]  Louis-Philippe Morency,et al.  A probabilistic multimodal approach for predicting listener backchannels , 2009, Autonomous Agents and Multi-Agent Systems.

[2]  Dirk Heylen,et al.  Appropriate and Inappropriate Timing of Listener Responses from Multiple Perspectives , 2011, IVA.

[3]  J. Bavelas,et al.  Listeners as co-narrators. , 2000, Journal of personality and social psychology.

[4]  Louis-Philippe Morency,et al.  Parasocial consensus sampling: combining multiple perspectives to learn virtual human behavior , 2010, AAMAS.

[5]  Louis-Philippe Morency,et al.  Learning Backchannel Prediction Model from Parasocial Consensus Sampling: A Subjective Evaluation , 2010, IVA.

[6]  Dirk Heylen,et al.  Learning and evaluating response prediction models using parallel listener consensus , 2010, ICMI-MLMI '10.

[7]  Joakim Nivre,et al.  On the Semantics and Pragmatics of Linguistic Feedback , 1992, J. Semant..

[8]  Dirk Heylen,et al.  The MultiLis Corpus - Dealing with Individual Differences in Nonverbal Listening Behavior , 2010, COST 2102 Training School.

[9]  Dennis Reidsma,et al.  Elckerlyc - A BML Realizer for continuous, multimodal interaction with a Virtual Human , 2009 .

[10]  A. Dittmann,et al.  Relationship between vocalizations and head nods as listener responses. , 1968, Journal of personality and social psychology.

[11]  V. Yngve On getting a word in edgewise , 1970 .

[12]  Steven H. Lewis,et al.  Listener Responsiveness and the Coordination of Conversation , 1982 .