PHYSIOLOGICAL CORRELATE OF FIXATION DISENGAGEMENT IN THE PRIMATES FRONTAL EYE FIELD

1. We recorded from the frontal eye field (FEF) of rhesus monkeys while they performed the gap task in which the fixation point disappears 200 ms before the appearance of the peripheral saccadic target. This gap allows the disengagement of fixation to begin before the acquisition of saccade coordinates, thereby greatly reducing saccade latency (''gap effect''). Very short-latency saccades obtained in this gap task have been called ''express saccades''. 2. We studied 145 FEF neurons that had presaccadic activity on conventional saccade tasks. When tested in the gap task with a 200-ms gap, nearly half of these neurons (69) increased their discharge rate in response to the disappearance of the fixation target. We call this increase a fixation-disengagement discharge (FDD). The mean latency of the start of the FDD relative to the fixation light extinction was 149 +/- 36 (SD) ms. 3. Gap-task trials with the saccade target in the cell's response held were randomly intermixed with trials having the target opposite to the cell's held. The FDD was present in both cases: on trials into the response field, the FDD was followed by the cell's presaccadic burst. On trials opposite the cell's field, the FDD activity was suppressed prior to the saccade. 4. The FDD was most likely to be found in cells that had the movement type of presaccadic activity, i.e., movement cells and visuomovement cells. FDD was observed in 57% of visuomovement cells A, B, and C, 50% with movement activity, and 18% purely visual. 5. The magnitude of FDD varied across cells but was usually less than the cell's saccade-related burst. With the 200-ms gap there was often a momentary reduction in spike rate between the FDD and the start of the saccade-related burst of movement cells. 6. We conclude that many FEF cells carry a physiological correlate of the disengagement of fixation. A FEF role in controlling fixation disengagement is consistent with the difficulties that patients with FEF lesions have in suppressing inappropriate saccades. In intact subjects, the FEF's FDD could prime oculomotor structures downstream (i.e., superior colliculus) such that the subsequent appearance of a peripheral target would trigger short-latency saccades.