The Effect of Avatar Model in Stepping Off a Ledge in an Immersive Virtual Environment
Animated digital self-representations of the user in immersive virtual environments, called self-avatars, have been shown to be an aid in perceptual judgments in virtual environment and provide critical information for people deciding whether an action can be take or not. In this body of work, the size of the self-avatar is carefully calibrated to match the size of user. However, little attention has been paid to the graphical model used to represent the self-avatar. In this thesis, we further investigate the question of whether the form of the model can affect perceptual judgments in an IVE. We study this question in the context of affordance judgments, that is, properties of the virtual environment that represent possibilities for action. Our specific task concerns the judgment of stepping off a virtual ledge, a task we have studied before. In that work, we showed that the presence of a self-avatar provided important information in making the judgment of whether to step off the virtual ledge or not. In this work, we will again employ that task, but vary the underlying representation of the self-avatar across subjects to see if it affects this judgment. The forms of self-avatars vary between no self-avatar, a simple line-based skeleton avatar, or a full-body, rich polygonal, gender-matched self-avatar. Our results replicate our prior work, and show that presenting a self-avatar significantly affects people’s perceptual judgment in virtual environments. However, the form of the self-avatar seem to make no difference in such tasks.