A Dynamic Type of Vernier Acuity in the Visual Prediction of Motion Trajectory

Abstract

The purpose of the experiment was to analyse a dynamic type of Vernier acuity in the visual prediction of the trajectory of linear apparent motion. The subjects were shown a single moving dot, which moved from the left end to the right end of a CRT. The change in dot position between two successive frames was 0.15 deg. Midway through the trajectory the dot moved behind an amodal occluding square, thus being invisible in this area. The vertical position where the dot re-emerged from behind the occluding square was varied, with 14 conditions for deviations both above and below the ‘correct’ locus of straight movement. A unit step of the deviation was 0.029 deg. There were three SOA conditions (13 ms, 21 ms, and 26 ms). In a 2AFC paradigm, the subjects had to report on which of two presentations the dot reappeared in the ‘correct’ position. We applied Probit analysis to fit a cumulative normal distribution to the result of each SOA condition and each deviation direction. We defined a value giving 75% correct response as a threshold value. Within each SOA condition, there was no difference between the threshold for deviations above or below the correct position, nor did threshold values between three SOA conditions show significant differences (ANOVA). We compare and discuss Vernier-acuity thresholds with and without amodal occlusion of the locus.