Displaying Depth in Computer Systems: Lessons from Two-Dimensional Works of Art

Virtual depth displays depend on static, monocular cues. Models of integrating monocular cues may be continuous (additive) or discontinuous. Previous research using simple displays and a small number of cues supported continuous cue integration. The present research is designed to expand the understanding of how the visual system integrates information from multiple pictorial cues by investigating combinations of one to ten pictorial cues in visually-rich, two-dimensional displays (paintings and photographs). Participants estimated depth in target paintings and photographs relative to a standard two dimensional display. Certain results suggest that the visual system integrates cues in a largely additive way, but after a number of cues are present there may be an additional boost in perceived depth resulting in a best-fittingdiscontinuous model of cue combination. However, this discontinuous effect may be due to designdecisions made by the painters rather than exclusively to the perceptual processes of the viewers. Analyses of these design decisions provide lessons for the design of two-dimensional displays.