Abstract
A grid on the ground, in an otherwise dark environment, can be delineated by both its global layout (linear perspective) and local texture density. However, it is unclear how these cues affect absolute distance judgment. To explore this, we extended an earlier study with a 1.2x3m texture surface (3×4 matrix, 1m from observer; He et al, 2000) in which we showed that false linear perspective (matrix converging distally), more than linear perspective (parallel matrix), caused observers to overestimate absolute distance (more so verbally than when instructed to perform visually directed action task/blindfolded walking). First, we reduced the size of the global texture matrix to 2×2. Fluorescent star shaped objects were used to form texture matrix with (a) linear perspective, (b) false perspective. Similar to our results with the larger 3×4 texture matrix, observers were better able to judge absolute distance in the blindfolded walking task, than in the verbal report task. Overall, judgments were better with linear perspective than false linear perspective cues. Second, we tested distance judgments with texture density cue using a 3×4 matrix with (a) regularly spaced star objects, (b) false foreshortening, i.e. density of star objects higher at the distal end of the matrix. Our results showed that distance judgments were not significantly different in the two conditions, which were about similar to the results in the linear perspective case. Together these findings suggest that the visual system is more vulnerable to the false linear perspective cue, than to the false foreshortening cue.
Supported by grants from A&S, U of L; Knights Templar Eye Foundation, Inc. and SCO Research Funds.