Does Brief Exposure to a Self-avatar Effect Common Human Behaviors in Immersive Virtual Environments?

A plausible assumption is that self-avatars increase the realism of immersive virtual environments (VEs), because self-avatars provide the user with a visual representation of his/her own body. Consequently having a self-avatar might lead to more realistic human behavior. To test this hypothesis we compared human behavior in VE with and without providing knowledge about a self-avatar with real human behavior in real-space. This comparison was made for three tasks: a locomotion task (moving through the content of the VE), object interaction task (interacting with the content of the VE), and social interaction task (interacting with other social entities within the VE). In the locomotion task, we measured participants’ velocity and walking ration when they walked from point A to point B. In the object interaction task we measured participants’ times needed to stick a triangle into several differently oriented openings of the same shape. Finally, in the social interaction task participants walked from point A to point B while an avatar was standing half-way on the direct route from A to B. Here we measured the minimal distance between the participant and the avatar (interpersonal distance). Surprisingly, we did not find effects of a self-avatar exposure on any of these tasks. However, participant’s VE and real world behavior differed significantly. These results challenge the claim that knowledge about the self-avatar substantially influences natural human behavior in immersive VEs.