How Foot Tracking Matters: The Impact of an Animated Self-Avatar on Interaction, Embodiment and Presence in Shared Virtual Environments

The use of a self-avatar representation in head-mounted displays has been shown to have important effects on user behavior. However, relatively few studies focus on feet and legs. We implemented a shared virtual reality for consumer virtual reality systems where each user could be represented by a gender-matched self-avatar controlled by multiple trackers. The self-avatar allowed users to see their feet, legs and part of their torso when they looked down. We implemented an experiment where participants worked together to solve jigsaw puzzles. Participants experienced either no-avatar, a self-avatar with floating feet, or a self-avatar with tracked feet, in a between-subjects manipulation. First, we found that participants could solve the puzzle more quickly with self-avatars than without self-avatars; but there was no significant difference between the latter two conditions, solely on task completion time. Second, we found participants with tracked feet placed their feet statistically significantly closer to obstacles than participants with floating feet, whereas participants who did not have a self-avatar usually ignored obstacles. Our post-experience questionnaire results confirmed that the use of a self-avatar has important effects on presence and interaction. Together the results show that although the impact of animated legs might be subtle, it does change how users behave around obstacles. This could have important implications for the design of virtual spaces for applications such as training or behavioral analysis.