Distance Perception and Cues to Distance in Virtual Reality

Distance perception in virtual reality is reviewed, and investigated in a series of experiments. In experiment 1, the systematic underestimation of distance found in virtual reality displays was investigated. This investigation used a video camera mounted on a head-mounted display to compare a photorealistic ”virtual” world to monocular viewing of the real world with restricted field of view and unrestricted viewing. In experiment 2, the use of angular declination from the horizon as a cue to distance was investigated in virtual reality using a virtual world with a variable horizon line height. Subjects performed both a verbal estimation task and a visually directed action task. Subject data were fit to power function models to avoid issues of linear underestimation. In experiment 3, distance perception in indoor and outdoor virtual environments was investigated, using the same tasks as experiment 2. Experiment 1 found significant distance underestimation in the head-mounted display condition, but not in either of the real-world viewing conditions. This suggests that neither restricted field of view, nor level of graphical detail are responsible for the distance underestimation found in virtual reality. Experiment 2 found similar performance between the two tasks, and significantly higher power function exponents when the horizon line was lowered. This supports declination from the horizon as a cue to distance, and is consistent with a theory of distance perception suggesting the two tasks use the same underlying representation of distance. Experiment 3 found similar performance between both tasks in the outdoor environment, but a significantly larger mean verbal exponent than motor exponent in the indoor condition. This dissociation between verbal and motor performance suggests that there might be some fundamental difference between the two tasks, or between the two environments used.