Group decision making in multi-user virtual reality and video conferencing

Head-mounted displays enable social interactions in immersive virtual environments. However, it is yet unclear whether the technology is suitable for collaborative work between distant group members. Other communication technologies such as online chat tools or video conferencing systems have repeatedly led to poorer group performance compared to face-to-face interaction. It has been argued that lack of social presence and increased cognitive load might be responsible for these results. The use of a multi-user Virtual Reality (VR) system can help to overcome some of these limitations. Multi-user VR enables users to convey spatially directed, non-verbal behavior (i.e., selective gaze and gestures), which was shown to be beneficial for group performance. Moreover, the entire visual field is taken by the VR display, thus allowing to reduce extraneous cognitive load by means of controlling task-irrelevant stimuli. We set out to compare multi-user VR, video conferencing, and face-to-face interaction in terms of collaborative performance in a group decision task. Furthermore, we examine whether performance is influenced by the feeling of presence, social presence, and cognitive load. Using the hidden profile paradigm, we have started testing small groups of three participants, who are working on a personnel selection case. In the hidden profile paradigm, each participant receives an individual information set containing shared and unshared information about four fictional candidates. The correct (i.e., most favorable) candidate can only be identified if sufficient unshared information is exchanged during the subsequent group discussion. The groups are randomly assigned to discuss the alternatives in a multi-user VR environment via virtual representatives (i.e., avatars), in a video conference, or in a face-to-face setting. Data collection (approx. 150 students) is currently ongoing. Consistent with previous research, we expect poorer performance of the video conferencing system compared to the face-to-face condition. In contrast, we expect better performance in the VR condition, presumably comparable to the face-to-face condition. We will present preliminary analyses and discuss our findings with respect to the potential of multi-user VR as a tool for online collaboration