Visually induced cross-axis postsaccadic eye drift.

1. It has been previously shown that, if a visual pattern is transiently moved just after every saccade, it is possible to induce horizontal, postsaccadic, ocular drift after horizontal saccades that persists in the dark. In this study we show that horizontal ocular drift can also be created after vertical saccades. Five human subjects viewed binocularly the interior of a full-field hemisphere filled with a random-dot pattern. They were encouraged to make frequent vertical saccades. During training, eye movements were recorded by the electrooculogram. A computer detected the end of every saccade and immediately moved the pattern to the left after up saccades and right after down saccades. The motion was exponential, its amplitude was 25% of the vertical component of the antecedent saccade, its time constant was 50 ms. Before and after 2-3 h of training, movements of both eyes were measured by the eye-coil/magnetic-field method while subjects were instructed to make vertical saccades in the dark, in the presence of the movable adapting pattern, and between stationary targets for calibration. 2. After training (approximately 20,000 saccades) all subjects developed a zero-latency, exponential ocular drift to the left after up saccades and to the right after down saccades. The amplitude of the horizontal drift, expressed as a percentage of the vertical component of the preceding saccade, was 2.7% in the dark. This rose to 10.2% in the presence of the movable adapting stimulus. The latter rise is not due to visual following systems but to a zero-latency increase in initial drift velocity. 3. The horizontal drifts were usually unequal between the two eyes, indicating the presence of disconjugate movements. We measured intrasaccadic disconjugate horizontal movements of all subjects. In agreement with studies by others of saccades in the light, we measured a divergence during up saccades (1.3 degrees) and a convergence for down (0.4 degrees), but in this case for spontaneous saccades in the dark. After training, these values increased for saccades in the dark but decreased in the light in the presence of the adapting stimulus. These changes were largely idiosyncratic and statistically significant in only a few subjects. 4. The cross-axis postsaccadic drifts were separated into their conjugate and disconjugate components. The disconjugate components were small and idiosyncratic, and the means were small for saccades in the dark. The only consistent trend was in the presence of the adapting stimulus where up saccades were often followed by convergence.(ABSTRACT TRUNCATED AT 400 WORDS)