Relative contributions of the two eyes to perceived egocentric visual direction in normal binocular vision

Abstract

Perceived egocentric direction (EVD) is based on the sensed position of the eyes in the orbit and the oculocentric visual direction (eye-centered, OVD). Previous reports indicate that in some subjects eye-position information from the two eyes contributes unequally to the perceived EVD. Findings from other studies indicate that the retinal information from the two eyes may not always contribute equally to perceived OVD. The goal of this study was to assess whether these two sources of information covary similarly within the same individuals. Open-loop pointing responses to an isolated target presented randomly at several horizontal locations were collected from 13 subjects during different magnitudes of asymmetric vergence to estimate the contribution of the position information from each eye to perceived EVD. For the same subjects, the direction at which a horizontally or vertically disparate target with different interocular contrast or luminance ratios appeared aligned with a non-disparate target estimated the relative contribution of each eye's retinal information. The results show that the eye-position and retinal information vary similarly in most subjects, which is consistent with a modified version of Hering's law of visual direction.

Figure 1

Corrected pointing errors are plotted against the asymmetric vergence demand. Each plot shows the data of a single subject. Positive and negative values on the x-axis indicate convergence and divergence demands, respectively. Positive and negative numbers on the y-axis specify pointing errors toward the right and left, respectively. Each point represents one pointing response. The best fitting straight lines are shown for pointing responses during changes in the position of the left (triangles) and right eyes (circles). The equation and coefficient of determination is included for each fitted line.

Figure 1

Corrected pointing errors are plotted against the asymmetric vergence demand. Each plot shows the data of a single subject. Positive and negative values on the x-axis indicate convergence and divergence demands, respectively. Positive and negative numbers on the y-axis specify pointing errors toward the right and left, respectively. Each point represents one pointing response. The best fitting straight lines are shown for pointing responses during changes in the position of the left (triangles) and right eyes (circles). The equation and coefficient of determination is included for each fitted line.

Figure 2

Points of subjective equality (PSEs) are shown for one subject for (a) different interocular log contrast ratios for horizontally disparate targets (b) different interocular log luminance ratios for horizontally disparate targets and (c) different interocular log contrast ratios for vertically disparate targets. In panels (a) and (b), the triangles and circles indicate PSEs on one set of trials for test targets with crossed and uncrossed disparity, respectively. In panel (c), the triangles and circles are PSEs for left-hyper and right-hyper disparity, respectively. The equation and coefficient of determination is included for each fitted line.

Figure 2

Points of subjective equality (PSEs) are shown for one subject for (a) different interocular log contrast ratios for horizontally disparate targets (b) different interocular log luminance ratios for horizontally disparate targets and (c) different interocular log contrast ratios for vertically disparate targets. In panels (a) and (b), the triangles and circles indicate PSEs on one set of trials for test targets with crossed and uncrossed disparity, respectively. In panel (c), the triangles and circles are PSEs for left-hyper and right-hyper disparity, respectively. The equation and coefficient of determination is included for each fitted line.

Figure 2

Points of subjective equality (PSEs) are shown for one subject for (a) different interocular log contrast ratios for horizontally disparate targets (b) different interocular log luminance ratios for horizontally disparate targets and (c) different interocular log contrast ratios for vertically disparate targets. In panels (a) and (b), the triangles and circles indicate PSEs on one set of trials for test targets with crossed and uncrossed disparity, respectively. In panel (c), the triangles and circles are PSEs for left-hyper and right-hyper disparity, respectively. The equation and coefficient of determination is included for each fitted line.

Figure 3

Comparison between the relative weighting of retinal information, as estimated using targets with different log contrast ratios (x axis) and log luminance ratios (y axis) in the two eyes. Each data point represents the results of one subject. The dashed line specifies perfect agreement. The solid line is the best fit to the data.

Figure 4

Comparison between the relative weighting of retinal information, as estimated using targets with different inter-ocular contrast ratios with horizontal (x axis) and vertical disparity (y axis). The dashed and solid lines are as in Figure 3.

Figure 5

Comparison between the relative weighting of eye-position (x axis), as determined from pointing responses, and the relative weighting of retinal information, estimated using (a) horizontally disparate targets with different contrast in the two eyes, (b) horizontally disparate targets with different luminance in the two eyes, and (c) vertically disparate targets with different contrast in the two eyes. The relative weighting of eye position is specified as the log ratio of the slopes fit to the results when the right and left eye positions varied. The relative weighting of retinal information is given as the log inter-ocular contrast or luminance ratio that resulted in perceived alignment between targets with crossed and uncrossed (or right and left hyper) disparity. In each panel, the dashed and solid lines have the same significance as in Figure 3.

Figure 5

Comparison between the relative weighting of eye-position (x axis), as determined from pointing responses, and the relative weighting of retinal information, estimated using (a) horizontally disparate targets with different contrast in the two eyes, (b) horizontally disparate targets with different luminance in the two eyes, and (c) vertically disparate targets with different contrast in the two eyes. The relative weighting of eye position is specified as the log ratio of the slopes fit to the results when the right and left eye positions varied. The relative weighting of retinal information is given as the log inter-ocular contrast or luminance ratio that resulted in perceived alignment between targets with crossed and uncrossed (or right and left hyper) disparity. In each panel, the dashed and solid lines have the same significance as in Figure 3.

Figure 5

Comparison between the relative weighting of eye-position (x axis), as determined from pointing responses, and the relative weighting of retinal information, estimated using (a) horizontally disparate targets with different contrast in the two eyes, (b) horizontally disparate targets with different luminance in the two eyes, and (c) vertically disparate targets with different contrast in the two eyes. The relative weighting of eye position is specified as the log ratio of the slopes fit to the results when the right and left eye positions varied. The relative weighting of retinal information is given as the log inter-ocular contrast or luminance ratio that resulted in perceived alignment between targets with crossed and uncrossed (or right and left hyper) disparity. In each panel, the dashed and solid lines have the same significance as in Figure 3.