Abstract
The current study examined the roles of binocular disparity and projected size in the detection of curved trajectories. Subjects were shown computer-generated displays of a sphere that traveled in the horizontal plane from a simulated starting position 8.6cm to the left and 122.1cm away from the observer to an ending position 8.5cm to the right of and 105.0cm away from the observer. On each trial observers were shown two displays. In one display the motion of the sphere was along a straight trajectory at a constant 3D speed; in the other the motion was along a constant curved trajectory at a constant 3D speed. To produce binocular disparity, all displays included reference objects presented 209.6cm away from the observer. A two-alternative forced choice procedure was used without feedback and observers were asked to indicate which display simulated a curved trajectory. Three independent variables were manipulated: viewing condition (binocular vs. monocular), type of curve (concave vs. convex relative to the observer), and projected size (constant 3D or constant 2D). The eight conditions were run in separate blocks counterbalanced across sessions. We found that, relative to monocular viewing, binocular viewing resulted in greater detection performance for convex curved trajectories. There was no significant difference between monocular and binocular viewing for concave curvatures. In addition, no systematic effects of projected size were observed. These results indicate that binocular information is more important than size information in detecting convex-curved trajectories. In addition, the importance of disparity information for convex trajectories suggests that change in disparity at the start of the motion path (which was greater for convex as compared to concave trajectories) is important for detecting curved trajectories.
Meeting abstract presented at VSS 2012