Reversible Inactivation of Monkey Superior Colliculus. I. Curvature of Saccadic Trajectory

Abstract

Aizawa, Hiroshi and Robert H. Wurtz. Reversible inactivation of monkey superior colliculus. I. Curvature of saccadic trajectory. J. Neurophysiol. 79: 2082–2096, 1998. The neurons in the intermediate layers of the monkey superior colliculus (SC) that discharge before saccadic eye movements can be divided into at least two types, burst and buildup neurons, and the differences in their characteristics are compatible with different functional contributions of the two cell types. It has been suggested that a spread of activity across the population of the buildup neurons during saccade generation may contribute to the control of saccadic eye movements. The influence of any such spread should be on both the horizontal and vertical components of the saccade because the map of the movement fields on the SC is a two-dimensional one; it should affect the trajectory of saccade. The present experiments used muscimol injections to inactivate areas within the SC to determine the functional contribution of such a spread of activity on the trajectory of the saccades. The analysis concentrated on saccades made to areas of the visual field that should be affected primarily by alteration of buildup neuron activity. Muscimol injections produced saccades with altered trajectories; they became consistently curved after the injection, and successive saccades to the same targets had similar curvatures. The curved saccades showed changes in their direction and speed at the very beginning of the saccade, and for those saccades that reached the target, the direction of the saccade was altered near the end to compensate for the initially incorrect direction. Postinjection saccades had lower peak speeds, longer durations, and longer latencies for initiation. The changes in saccadic trajectories resulting from muscimol injections, along with the previous observations on changes in speed of saccades with such injections, indicate that the SC is involved in influencing the eye position during the saccade as well as at the end of the saccade. The changes in trajectory when injections were made more rostral in the SC than the most active burst neurons also are consistent with a contribution of the buildup neurons to the control of the eye trajectory. The results do not, however, support the hypothesis that the buildup neurons in the SC act as a spatial integrator.