Disparity-induced and blur-induced convergence eye movement and accommodation in the monkey.

The dynamics of vergence eye movement and of ocular accommodation were studied in two monkeys trained to track a haploscopically presented target that appeared to move in depth. The target was presented under four conditions: monocular viewing, normal binocular viewing, accommodation-open-loop binocular viewing, and "conflict" viewing, in which the accommodation and vergence stimuli did not correspond to those produced by any real target in three-dimensional space. The first and third conditions were chosen because in each case only one of the two primary cues that guide accommodation and vergence was operative: blur in the first case and disparity in the third. We usually studied responses to apparent target movement directly toward or away from the right eye, in which accommodation was measured. The latencies of the accommodation responses to steps toward the monkey were approximately 180 and 240 ms in the two monkeys, while in both monkeys the latencies of convergence were approximately 160 ms. Neither the vergence latencies nor the accommodation latencies were greatly different in monocular and binocular viewing. Responses to a sinusoidally moving target (frequencies 0.1-1.2 Hz; peak-to-peak amplitude 0.5-4 diopters or meter-angles) were studied in the first three of the above viewing conditions. In binocular viewing, even with accommodation open-loop, vergence and accommodation showed much smaller phase lags than in monocular viewing. Furthermore, in response to step changes, both vergence and accommodation velocities were higher in binocular viewing than in monocular viewing. Thus the dynamic control of both vergence and accommodation relies predominantly on disparity signals. At low frequencies (0.2 or 0.3 Hz) the monkeys showed only a modest ability to separate their accommodation and vergence responses when presented with conflicting blur and disparity cues. A simple linear calculation based on the data above was used to predict the responses in such situations. The predicted and observed responses were in reasonable agreement.