Prediction of illusory motion direction from eye tracking data
At a Glance
Section titled âAt a Glanceâ| Metadata | Details |
|---|---|
| Publication Date | 2015-09-01 |
| Journal | Journal of Vision |
| Authors | Oliver Flynn, Arthur Shapiro |
| Institutions | American University |
Abstract
Section titled âAbstractâThe perception of motion can be generated by modulating the contrast relationships among objects, backgrounds, and thin lines separating them (Gregory and Heard, 1983; Shapiro et al. 2005; and others). This motion can be called âillusoryâ since no elements in the image physically change location. Here we examine a variant of this type of motion to see the extent to which it produces eye movements and whether we can predict the direction of illusory motion by examining only the eye movement data. The stimulus consisted of a large mid-luminance colored diamond (visual angle 16 degrees) bordered by thin (0.07 deg) edges and surrounded by a background (19 deg). Both the edges and background modulated in luminance between light and dark at 2 Hz; due to differences in timing between these elements, the diamond appeared to move continously in one direction (up, down, left, or right, depending on condition) though the diamond itself did not change position (Flynn and Shapiro, 2012). An eye tracking camera (50 Hz, Cambridge Research Systems) mounted to a chin rest tracked the gaze of naive observers (n = 30) who were asked simply to âwatchâ the stimulus on a computer monitor. Each observer saw 50 3-second trials presented in ranomized blocks of 5 conditions (up, down, left, right, no motion). An analysis of the slope of the eye movementsâchange in position shows that observersâ eye movements drift in the direction of perceived motion. Additionally, an anylsis of the eye postion following saccades shows that observersâ gazes center around the leading edge of the stimulus in each motion condition. We conclude that illusions of motion appear to create reliable patterns of eye movements which resemble responses to real motion. Eye tracking data can be used to predict the direction of illusory motion perceived by the observer. Meeting abstract presented at VSS 2015