Abstract
Newer display technologies permit more flexible stimulus design and presentation possibilities, yet their characteristics must support comfortable binocular vision. The objective of this study was to determine the stimulus size needed to allow reflex vergence responses to maintain accurate eye alignment. Twelve typical adults (19-54 years) viewed grayscale natural images (SYNS dataset) presented dichoptically with radii of 1,2,4,8, or 16 deg and a gradual reduction in edge contrast to a mean luminance surround. The stimulus aperture was always stationary. In condition 1, step changes in disparity (0,0.5,1,2,3,4, or 6 deg) were introduced for 1.5s. Final vergence amplitudes (last 0.25s) were computed. In condition 2, disparity varied between ±4deg following a random walk in velocity for 40s. Cross-correlations between stimulus and vergence velocities were performed. Eye movements were recorded with an Eyelink 1000 (SR Research). In condition 1, for the 16deg radius, the disparity tuning function peaked at 3deg disparity for convergent and 2deg for divergent stimuli (mean±SD amplitudes of 1.79±1.12deg and 1.53±0.66deg). Stimulus radius had no significant effect on vergence amplitude for disparities ≤1deg, as all radii elicited accurate responses, and for disparities ≥4deg, as vergence changes were not disparity-driven. For disparities of 2-3deg, response amplitudes decreased gradually with decreasing stimulus size. In condition 2, for the 16deg radius, the vergence peak correlation of 0.59±0.10 was not significantly different from that for 8, 4, and 2deg radii, whereas tracking performance was significantly poorer with 1deg stimulus radius (peak correlation of 0.43±0.12, p<0.001). Stimulus size had no significant effect on vergence latency (0.177±0.030s for 16deg, p=0.175). These results demonstrated the interaction effect between stimulus size and disparity on step vergence responses, indicating the need for careful consideration of matching stimulus content in the two eyes, whereas dynamic vergence tracking seemed more robust to the effect of stimulus size.