Évaluation de l'impact de l'incarnation d'un égo-avatar temps réel sur la marche

Abstract

In recent years, virtual reality (VR) has become increasingly accessible and less expensive. Given this ease of access, many studies have continued to use VR, which was already available before, but at much higher costs. With VR, it is possible to place a user in a virtual environment (VE) as well as to give him a virtual body known as virtual self-avatar. This self-avatar is best presented in real time and offers a visual feedback showing the own movements of the user in the virtual world. It has been shown that the self-avatars have an impact on psychological behavior, such as influencing rhythm patterns as well as the fluidity of movement when playing drums, but it has also been shown that avatars can induce a drift of movement when it is controlled. For example, a person performing a hand movement with a virtual hand shifted from the actual position of his hand, this person will use the position of the virtual hand to perform the task. In gait rehabilitation, those self-avatars are also used. They are generally used to give a biofeedback to the participant who is walking in the VE, or to influence his gait. However, no study aims to isolate the impact of these self-avatars on gait. It’s then interesting to ask what the impact of these self-avatars on gait behavior is. The objective of this study is to evaluate and quantify the impact that self-avatars can have on gait in VR, more precisely what specific parameters of gait pattern can be changed by the presence of these real time virtual self-avatars. For this project, the participants were presented to two studies. The first (study A) was designed to isolate the impact that a real-time virtual self-avatar can have on gait. In this study, the participants had to walk in four different environments, real and virtual, as well as with and without presence of a real-time virtual self-avatar. The absence of self-avatar seems to induce a more confident motor behavior, characterized by greater steps and a higher pace in the virtual world. The second study, Study B, aims to assess the impact of two different self-avatars on gait patterns in VR. Participants had to perform a motor task, either to walk by performing a foot placement task, under two different conditions, with a human-looking self-avatar and with a self-avatar resembling the fictional creature of Godzilla. The difference in self-avatar does not induce a change in the motor pattern of the participants, but the participants significantly exaggerate their movements when performing the foot placement task when they embody the Godzilla character. The participants also underestimate the time when he embodies the Godzilla character. In order to further isolate the impact of the self-avatar in VEs, the system used in this study should be modified to use another medium than the fixed-speed treadmill. Finally, the results that emerge from this study show that a real-time self-avatar does not induce a change in the motor behavior when walking, but the appearance of it can influence the behavior of a participant when performing targeted motor tasks.

Date	16 Oct 2019
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	David Labbé (Supervisor)