Abstract
Virtual reality (VR) technology is being used in an increasing number of applications. However, research has shown that we often perceive surface properties differently in real and virtual environments. To evaluate how well virtual platforms support realistic lightness perception, we measured lightness constancy in a physical scene, in VR, and on a 2D flat-panel display. Twelve observers participated in three conditions. In the physical condition, observers performed a lightness matching task on a custom-built apparatus where adjustable reflectance patches were visible through two 2-degree apertures. On each trial, the reference aperture was set to one of three reflectances (0.18, 0.39, 0.55). The match aperture had one of five illumination levels, between 1.25 and 3.05 times the illuminance at the reference aperture. Observers adjusted the reflectance at the match aperture until it appeared to match the reflectance at the reference aperture. In the VR condition, observers viewed an apparatus and room that replicated the physical condition, rendered in Unity, on an Oculus Rift S headset. In the flat-panel condition, observers viewed an apparatus like the one in the physical condition, rendered on an LCD screen using Unity. Thouless ratios for lightness constancy were significantly higher (indicating greater constancy) in the physical condition (mean and 95% confidence interval 0.87 ± 0.04) than in the flat-panel condition (0.79 ± 0.08). Thouless ratios were not significantly different in the VR condition (0.83 ± 0.08) than in the physical condition or the flat-panel condition. Thus in the simple scenes considered here, lightness constancy is moderately lower in virtual environments than in physical environments. This discrepancy should be considered when developing applications where realistic performance is critical, but our results suggest that VR can be a flexible alternative to flat panel displays and a reasonable proxy for real environments.