Eye Movement Conjugacy While Watching a Video Reveals Greater Vertical than Horizontal Disconjugacy in Human Subjects

Abstract

Purpose: We have developed a novel method for quantitation of both horizontal and vertical eye movement conjugacy that is performed by comparing the position of the pupils over time while a subject is eye tracked watching a video playing in a moving aperture or on a full screen monitor. The purpose of this work is to compare the horizontal, vertical and total conjugacy of eye movements with these different stimuli.

Methods: 57 normal control subjects who denied neurologic or ophthalmic dysfunction were binocularly eye tracked with a camera placed at a fixed distance while watching a music video playing continuously in an aperture moving in a rectangular configuration. Ten of these subjects also watched a video moving in a square pattern within an aperture. For both the square and rectangle pattern of viewing, the aperture was approximately 1/9 of the screen size of a 17” diameter monitor. 17 of the original 57 subjects viewed a 40 second video playing on the same full screen 17” diameter computer monitor. Right and left eye positions were compared to assess conjugacy of eye movement in Cartesian space relative to time. All data were analyzed using XLSTAT version 2012.6.02 (Addinsoft SARL, Paris, France) and MedCalc version 12.6.1 (MedCalc Software, Ostend, Belgium). A p-value of ≤ 0.05 was deemed as statistically significant. We compared eye-tracking parameters among the groups using the Kruskal-Wallis test.

Results: Normal subjects demonstrated conjugate eye movement that was not impacted by age or sex. Subjects demonstrated significantly increased total conjugacy of eye movement when viewing the smaller moving aperture relative to the full screen. There was no difference in total, horizontal, or vertical conjugacy of eye movement between the square and rectangular patterns of movement. Horizontal eye movement was significantly more conjugate than vertical eye movement regardless of whether the eyes were following an aperture moving in a square or a rectangle.

Conclusions: Normal control subjects have eye movements that are more conjugate when viewing a video playing in a small aperture versus on a larger screen, and in the horizontal versus vertical plane, as assessed by our algorithm. Since a disconjugate gaze may result from intracranial pathology the algorithm we have developed may be useful for automatable assessment and monitoring of patients with neurologic dysfunction.

Keywords: 523 eye movements: conjugate • 524 eye movements: recording techniques • 525 eye movements: saccades and pursuits