Short-latency disparity vergence eye movements: dependence on the preëxisting vergence angle.

We recorded the vergence eye movements that are elicited at ultra-short latencies when binocular disparities are applied to large-field patterns (Busettini, C., Miles, F.A. and Krauzlis, R.J. (1996). J. Neurophysiol., 75: 1392-1410) and determined their dependence on the preëxisting vergence angle (PVA). The search coil technique was used to record the movements of both eyes in four healthy subjects (two with presbyopia). Using dichoptic viewing, the two eyes saw identical images each consisting of a fixation cross at the centre of a random-dot pattern in a circular aperture. The subject fixated the crosses and then the images (crosses, random dots, windows) moved horizontally (1.5 degrees/s) in opposite directions so as to bring the eyes to the desired horizontal vergence position without changing the accommodation demand. After a further 800-1200 ms to permit fusion at this new vergence angle (now, the PVA), a disparity step was applied and, 200 ms later, the screen changed to uniform grey, marking the end of the trial. The disparity steps could have one of six magnitudes and four directions (crossed, uncrossed, right-hyper, left-hyper) while the PVA was varied systematically. The horizontal and vertical disparity vergence responses (DVRs) of one of the presbyopes consistently showed robust linear dependence on the PVA (r(2)>0.96). The horizontal DVRs of the other three subjects showed no sensitivity to the PVA and their vertical DVRs showed only very weak dependence. The experiment was repeated on one of the non-presbyopes after cycloplegia, but the outcome was the same, indicating that the negative findings were not due to the influence of the vergence-accommodation response. Our data indicate that the DVRs can be scaled by the PVA, but most subjects do not show this effect, perhaps because they relied on other distance cues that are uninformative in our experimental situation.