Real faces and robot faces: The effects of representation on computer-mediated communication

While there is much research regarding audio, video and text based communication, there has been little work concerning how users communicate via avatars—that is graphical embodiments of remote users. The aim of this study was to explore the effects of different forms of representation, by examining how users communicate via high quality video images and basic graphical representations in different communicative contexts. Communication analysis revealed that video images facilitate turn-taking, although they are not necessarily perceived very differently from basic avatars in terms of questionnaire responses. Using eye-tracking techniques, we found that while participants generally gaze more often at video images, this is dependent on the communicative context and is not necessarily an advantage in a problem-solving situation. This study has demonstrated the value of employing various measures and tasks in order to evaluate computer-mediated interactions. The results have implications for the use of video and graphical representations in computer mediated communication and suggest that the benefits of video must be considered in relation to the user's requirements (e.g. communication process versus outcome) and context in which the system is to be applied (e.g. problem-solving or social).