Express saccades: the conditions under which they are realized and the brain structures involved

Abstract

Repeated eye movements made to visual targets can yield a bimodal distribution of saccadic latencies the first mode of which was termed express saccades by Fischer and Boch. We have carried out a series of experiments in monkeys to identify the conditions necessary for express saccade generation, the rules involved, and the brain structures responsible. Express saccades are produced to singly appearing targets but not when several stimuli appear at the same time. Learning plays a central role in express saccade generation. What is learned is the vector of the saccade to be produced, not the location of the target is space or the orbital position of the eye. The frequency with which express saccades are made depends on expectation; express saccades seldom arise when targets appear at unexpected locations. Increasing the gap time between fixation spot termination and target onset and decreasing the number of expected target locations increases the frequency of express saccade generation. Express saccades are made with high frequency to the second of two successive targets as long as the vector is one that has been learned. Pre-cuing can increase the frequency of express saccades only within a limited range. Free viewing of natural scenes commonly yields saccades in the express range but the distribution is not bimodal. Lesions of the frontal eye fields, of the medial eye fields, and of areas V4 and MT do not interfere with the generation of express saccades. Neither does the disruption of either the magnocellular or parvocellular systems. But superior colliculus lesions eliminate express saccades; no recovery is evident It appears therefore that the cortical areas involved in express saccade generation send their signals to the brainstem through the superior colliculus.

Footnotes

 Supported by the National Eye Institute under grant EY008502.