Influence of peripheral and stereoscopic vision on driving performance in a power wheelchair simulator

Abstract

Exercising in a Virtual Environment as an adjunct training method, can enhance safety, cost-effectiveness, training quality, and can be used as an assessment tool. In the realm of power wheelchair (PWC) simulation a number of systems have been developed over the past few years. Unfortunately, these simulators are: 1) rather simple, in particular lack correct physics simulation, 2) do not support peripheral vision, 3) are not suitable as a standard assessment tool, and 4) are not available commercially, except for one system. This study investigates factors influencing user's driving performance in a PWC simulator. It addresses the central research question: Do peripheral vision and/or stereoscopic viewing have an influence on PWC driving performance? This research directly investigates these issues by treating standard display viewing, peripheral vision, and stereoscopic 3D viewing as independent variables to see which one influences driving performance in a PWC simulator, which was developed for this purpose. This study also compares the users' sense of presence as a side effect and considers their preference for each of the three independent variables. In a randomized within-subject design, 24 participants performed the same navigation tasks across three conditions, namely monoscopic narrow field of view (narrow-FOV), monoscopic wide field of view (wide-FOV), and stereoscopic narrow field of view (stereo-FOV). The number of path collisions, number of wall collisions, and time spent was recorded to measure individual performance, whereas the sense of presence was measured using a standard questionnaire. In terms of driving performance, the results indicate that although the number of path and wall collisions was fewest in the wide FOV condition, there were no significant differences found between the means. However, time spent and overall driv¬ing performance in the wide FOV condition was significantly better compared with other conditions. The result of the users' sense of presence assessment shows that the wide-FOV and stereo-FOV were rated significantly higher by the participants. Further, the wide-FOV condition was preferred by 83% of the subjects when they had to choose between the conditions. The wide-FOV seems to be a promising interface for complementing training or assessment of PWC users.