Abstract
Knowledge of human visual space perception in the intermediate distance (>2m) is predominantly based on testing in a static setting, where neither the observer nor target move. To extend our understanding, we investigated observer’s ability to localize a target in the dark while walking. Previous studies have revealed a stationary observer characteristically locates a dimly-lit target in the dark at the intersection between the observer’s intrinsic bias (an implicit slant surface/curve) and its projection line from the eyes. Our present study explored how the intrinsic bias changed when the observer either walked forward to, or backward away from, a static target. We used the blind-walking-gesturing task to measure perceived location of a 0.2 degree dimly-lit (0.16 cd/m2) target. The target was placed at one of 12 locations (distance: 1.5, 3.25, 5.75 and 7.0 m; height: 0.14, 0.74 m and eye level). First, in a baseline stationary condition, observers (n=4) stood still and judged the location of the static target presented for 3.0 sec. Confirming previous findings, the judged locations can be transcribed by the intrinsic bias curve. Second, in a backward-walking condition, the observer walked backward for 1.25 m while viewing the static target. They then responded to the perceived target location with the blind-walking-gesturing task. Third, in a forward-walking condition, the observer followed the same protocol except for walking toward the target by 1.25 m. We found in comparison with the stationary condition, average judged distances were significantly farther in the backward-walking condition (p<0.001), and the intrinsic bias largely remained at the baseline position before stepping backward. In the forward-walking condition, however, judged distances were similar to that in the stationary condition (p>0.05). Altogether, these findings indicate that during self-motion in the dark, the dynamics of the intrinsic bias depend on the observer’s movement directions, thus affecting perceived space.