Experimenting with the Gaze of a Conversational Agent

We have carried out a pilot experiment to investigate the effects of different eye gaze behaviors of a cartoon-like talking face on the quality of human-agent dialogues. We compared a version of the talking face that roughly implements some patterns of humanlike behavior with two other versions. We called this the optimal version. In one of the other versions the shifts in gaze were kept minimal and in the other version the shifts would occur randomly. The talking face has a number of restrictions. There is no speech recognition, so questions and replies have to be typed in by the users of the systems. Despite this restriction we found that participants that conversed with the optimal agent appreciated the agent more than participants that conversed with the other agents. Conversations with the optimal version proceeded more efficiently. Participants needed less time to complete their task