Abstract
The ability of a user to move to different locations within a virtual environment (VE) is a fundamental action that subserves the activities of exploration and manipulation. By empirical analogy, the perceptual information used to locomote to a target within a virtual environment is compared to the perceptual information used to walk to a location in the real world. An experiment is reported that had participants move to a location as accurately as possible within a VE where a target object was presented. The amount of visual feedback available to participants was manipulated. Three conditions were compared: static viewing of the target and virtual environment before locomotion, the disappearance of the target object as movement to the object was initiated, and locomotion to the target while both object and environment were present. In addition, the composition of virtual environments was either textured or polygonal. Error measures indicated that users locomote within VE's with less accuracy than those that walk blindfolded in the real world. Texture had its largest effect on the accuracy of movement when optic flow was not available, that is, static estimates of distance. Discussions center on the relative contribution of visual, cognitive, and proprioceptive information to VE user movement accuracy.
Get full access to this article
View all access options for this article.