Abstract
Purpose :
Small eye movements (drifts, microsaccades and tremors) keep the eye in continuous motion, even as observers attempt to fixate a target. In this study we examine eye velocity distributions from typical observers as they fixate targets presented at different gaze angle. We hypothesise that the mechanics of the oculomotor plant will bias eye velocity towards primary position.
Methods :
Twelve typical observers with normal vision were asked to fixate a target (0.4° green dot) presented for 12s at either 0° (primary position) or at eccentricity of ±16° horizontal or ±16° vertical. The targets were presented in a random order against a black background in an otherwise dark room. Observers viewed the targets binocularly and eye movements were recorded at 1000Hz from the eye with better visual acuity. Saccades and blinks were excluded from eye movement data and a bivariate probability density function of target-relative eye velocity was calculated after filtering eye position data. The centroid of the isocontour that encompassed the highest 68% of data was examined to determine any directional bias.
Results :
Preliminary results show the mean x coordinate of the centroid has a leftward bias (-0.030 ± 0.092°/s) during right-gaze (+16°), and a rightward bias (+0.039 ± 0.120°/s) during left-gaze (-16°). This difference (0.072°/s) was statistically significant [t(11) = 2.697, p = 0.010, d = 0.779]. In contrast, the mean y coordinate of the centroid had a downward bias during up-gaze (-0.217 ± 0.310°/s) and down-gaze (-0.230 ± 0.281°/s). This difference (0.012°/s) was not statistically significant [t(11) = 0.101, p = 0.461, d = 0.029]. There were no significant differences between any of the eccentric gaze positions and primary position (i.e. 0°).
Conclusions :
We have demonstrated that eye velocity during foveal fixation is dependent on gaze angle. Eye velocity is biased towards primary position during horizontal gaze, and downward during vertical gaze. These findings may relate to differences underlying horizontal and vertical oculomotor control. We speculate that the magnitude of the horizontal differences observed will increase with larger gaze angles (i.e. beyond 16°). Our results may have implications for aspects of foveal vision that are critically dependent on eye velocity, e.g., visual acuity, with thresholds potentially varying as a function of horizontal gaze angle.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.