Abstract
Abstract: :
Purpose: Some visual functions have been shown to undergo "normalization", such that a substantial threshold elevation produces little or no change in the perceived magnitude of suprathreshold stimuli. For example, we found previously that suprathreshold position offsets are perceived veridically despite up to a 12-fold elevation of the position-offset threshold. Here, we examine the effect of threshold elevation on the perception of suprathreshold speed, using a stimulus that includes minimal position information. Methods: A gray-scale random-dot pattern (50% contrast, 2 min/dot; 3.3 sq. deg field; 13 ms/frame; 50% limited-lifetime spectral coherence) moved horizontally on a bright background for 500 ms. Each trial consisted of two 500-ms intervals separated by about 3 s. In each interval, the direction of motion was randomly right or left. Two types of experiments were conducted. In threshold experiments, the observer indicated which interval contained the motion signal. Motion thresholds were measured during steady fixation with and without voluntary 1.5 Hz head movement. In subsequent suprathreshold matching experiments, the same observers compared the perceived speed of suprathreshold motion during voluntary head movement (4.3 and 8.6 deg/s, corresponding to approximately 3x and 6x threshold) and when the head was stationary. On each trial, the observer indicated whether the motion interval with or without head movement contained a higher speed. Points of subjective equality between the head-movement and head-stationary conditions were determined from psychometric functions, obtained using the method of constant stimuli. To remove potential biases, the data were averaged across experiments in which head movement occurred during the first or second interval of each trial. Results: The mean motion threshold increases from approximately 0.7 deg/s when the head is stationary to approximately 1.5 deg/s during 1.5 Hz horizontal head movement. During voluntary head movement, the perceived speed of both suprathreshold target velocities is approximately 20% less than when the head is stationary. Conclusion: The failure of perceived speed to exhibit "normalization" during voluntary head movement implies that suprathreshold motion signals are not compensated for the resulting elevation of motion thresholds. If the perception of suprathreshold speed also fails to normalize when position information is available, then substantial and potentially dangerous errors in the perception of object speed could occur under conditions of observer motion.
Keywords: 477 motion-2D • 585 space perception • 586 spatial vision