Inferring interaction to support collaborative learning in 3D virtual environments through the user's avatar Non-Verbal Communication

When a group of students, placed around a shared workspace and working on a predefined task, is observed, the teacher can intuitively, to a certain extent, understand the collaboration that is taking place within the group without listening to the students' discourse. Following this analogy, we propose that some Non-Verbal Communication (NVC) cues, adapted to 3D virtual environment features, could be useful to infer the quality of collaborative interaction to such an extent that collaborative learning can be automatically fostered. The NVC cues that have been selected for that purpose and the related collaborative learning indicators that can be inferred will be discussed in this paper.