Electrical microstimulation and unit recording were performed in dorsomedial frontal cortex of four alert monkeys to identify an oculomotor area whose existence had been postulated rostral to the supplementary motor area. Contraversive saccades were evoked from 129 sites by stimulation. Threshold currents were lower than 20 microA in half the tests. Response latencies were usually longer than 50 ms (minimum: 30 ms). Eye movements were occasionally accompanied by blinks, ear, or neck movements. The cortical area yielding these movements was at the superior edge of the frontal lobe just rostral to the region from which limb movements could be elicited. Depending on the site of stimulation, saccades varied between two extremes: from having rather uniform direction and size, to converging toward a goal defined in space. The transition between these extremes was gradual with no evidence that these two types were fundamentally different. From surface to depth of cortex, direction and amplitude of evoked saccades were similar or changed progressively. No clear systematization was found depending on location along rostrocaudal or mediolateral axes of the cortex. The dorsomedial oculomotor area mapped was approximately 7 mm long and 6 mm wide. Combined eye and head movements were elicited from one of ten sites stimulated when the head was unrestrained. In the other nine cases, saccades were not accompanied by head rotation, even when higher currents or longer stimulus trains were applied. Presaccadic unit activity was recorded from 62 cells. Each of these cells had a preferred direction that corresponded to the direction of the movement evoked by local microstimulation. Presaccadic activity occurred with self-initiated as well as visually triggered saccades. It often led self-initiated saccades by more than 300 ms. Recordings made with the head free showed that the firing could not be interpreted as due to attempted head movements. Many dorsomedial cortical neurons responded to photic stimuli, either phasically or tonically. Sustained responses (activation or inhibition) were observed during target fixation. Twenty-one presaccadic units showed tonic changes of activity with fixation. Justification is given for considering the cortical area studied as a supplementary eye field. It shares many common properties with the arcuate frontal eye field. Differences noted in this study include: longer latency of response to electrical stimulation, possibility to evoke saccades converging apparently toward a goal, and long-lead unit activity with spontaneous saccades.

The voluntary control of gaze implies the ability to make saccadic eye movements specified by abstract instructions, as well as the ability to repress unwanted orientating to sudden stimuli. Both of these abilities are challenged in the antisaccade task, because it requires subjects to look at an unmarked location opposite to a flashed stimulus, without glancing at it. Performance on this task depends on the frontal/prefrontal cortex and related structures, but the neuronal operations underlying antisaccades are not understood. It is not known, for example, how excited visual neurons that normally trigger a saccade to a target (a prosaccade) can activate oculomotor neurons directing gaze in the opposite direction. Visual neurons might, perhaps, alter their receptive fields depending on whether they receive a pro- or antisaccade instruction. If the receptive field is not altered, the antisaccade goal must be computed and imposed from the top down to the appropriate oculomotor neurons. Here we show, using recordings from the supplementary eye field (a frontal cortex oculomotor centre) in monkeys, that visual and movement neurons retain the same spatial selectivity across randomly mixed pro- and antisaccade trials. However, these neurons consistently fire more before antisaccades than prosaccades with the same trajectories, suggesting a mechanism through which voluntary antisaccade commands can override reflexive glances.

The oculomotor system has long been thought to rely on an accurate representation of eye displacement or position in a successful attempt to reconcile a stationary target's retinal instability (caused by motion of the eyes) with its corresponding spatial invariance. This is in stark contrast to perceptual localization, which has been shown to rely on a sluggish representation of eye displacement, achieving only partial compensation for the retinal displacement caused by saccadic eye movements. Recent studies, however, have begun to case doubt on the belief that the oculomotor system possess a signal of eye displacement superior to that of the perceptual system. To verify this, five humans and one monkey (Macaca nemestrina) served as subjects in this study of oculomotor localization abilities. Subjects were instructed to make eye movements, as accurately as possible, to the locations of three successive visual stimuli. Presentation of the third stimulus (2-ms duration) was timed so that it fell before, during, or after the subject's saccade from the first stimulus to the second. Localization errors in each subject (human and nonhuman) were consistent with the hypothesis that the oculomotor system has access to only a damped representation of eye displacement--a representation similar to that found in perceptual localization studies.

Reviews on the visual system generally praise its amazing performance. Here we deal with its biggest weakness: sluggishness. Inherent delays lead to mislocalization when things move or, more generally, when things change. Errors in time translate into spatial errors when we pursue a moving object, when we try to localize a target that appears just before a gaze shift, or when we compare the position of a flashed target with the instantaneous position of a continuously moving one (or one that appears to be moving even though no change occurs in the retinal image). Studying such diverse errors might rekindle our thinking about how the brain copes with real-time changes in the world.

WGA-HRP was used to examine projections to the brainstem from the supplementary eye field (SEF). The SEF was defined electrophysiologically in awake, behaving monkeys and connections were compared to those of the arcuate frontal eye field (FEF), area 6DC, and primary motor cortex. The SEF was found to have either direct or indirect connections with almost every known pre- and paraoculomotor structure of the brainstem. The SEF was found to project bilaterally to layers I and IV of a tangentially widespread region of the superior colliculus. Terminal label was evident in the pretectal olivary nucleus, nucleus of the optic tract, nucleus raphe interpositus (omnipause region), nucleus prepositus hypoglossi, the perioculomotor cap of the central gray, dorsal central gray, nucleus reticularis tegmenti pontis, nucleus reticularis pontis oralis, and to multiple nuclei of the basis pontis (most densely to the dorsomedial nucleus). Bilateral projections were found in the parvicellular red nucleus. Reciprocal connections were present in the nucleus limitans, the mesencephalic reticular formation, locus coeruleus, and the serotonergic nuclei of the raphe complex (dorsalis and central superior). Overall patterns of connectivity were similar to those of the FEF and markedly different from those of the contiguous dorsocaudal area 6 or primary motor cortex. It was concluded that observed patterns of SEF-brainstem connectivity further justifies viewing this region as a distinct eye field that is likely to serve preparatory and trigger functions in the generation of saccadic eye movements.

The region in and around the thalamic internal medullary lamina (IML) in the cat recently has been shown to contain neurons active with ocular saccades and responding to visual stimuli. In the present study, single-unit microelectrode recordings were made in the corresponding thalamic region of the alert monkey in order to determine whether neurons with similar properties existed. Our objective was to specify the functional characteristics of these thalamic cells in the monkey, since 1) cell populations in the central thalamus form an important link between brain stem structures, such as superior colliculus and paramedian pontine reticular formation, and cortical areas, such as frontal eye field and inferior parietal lobule; and 2) most neurophysiological information on these structures with regard to gaze mechanisms has been obtained in primates. In this first part of the study we report observations on 164 thalamic units whose activity was related to the performance of spontaneous eye movements, head fixed. The animals had been trained on a visual discrimination task but photic stimuli were used only for calibrating the eye-position recording and for inducing small saccades and smooth pursuit. The experiments were performed in dim red light and in total darkness. Three types of units were found: 67 saccadic burst units, 58 saccade pause-rebound units, and 39 eye-position units. Sixty-two of the burst units had a directional preference. Most of the on-directions were contraversive, and it was in such units that the lead time of firing before saccades was the longest (up to at least 400 ms). Some of the burst units had a movement field, others fired more intensively and with a longer lead time, depending on the eccentricity of the eye position reached in orbit. The five units with no directional preference were the ones showing the best relation of burst duration with saccade duration. Three types of pause-rebound units were distinguished, depending on whether the saccadic pause or the postsaccadic burst was the most conspicuous event or the pause occurred after saccade offset. The three types were called, respectively, omnipausers, omnirebound cells, and late pausers. Omnipausers and omnirebound cells had no directional preference but their typical firing patterns occurred very consistently with all saccades, even less than 2 degrees. In a few units, the rebound progressively faded away in total darkness. The relation of firing rate of eye-position units with eccentricity of the eyes in orbit was analyzed. Fluctuations in time and a hysteresis effect were found to affect this relation.(ABSTRACT TRUNCATED AT 400 WORDS)

Saccades were electrically evoked from the frontal eye field (FEF) of two trained monkeys while saccade-cells were recorded from the intermediate layers of the superior colliculus (SC). We found that FEF microstimulation, eliciting saccades of a given vector, excited SC saccade-cells encoding the same vector and inhibited all others. Such a mechanism can prevent competing commands from arising simultaneously in different structures.

. Neuronal activities were recorded in the supplementary eye field (SEF) of 3 macaque monkeys trained to perform antisaccades pseudorandomly interleaved with prosaccades, as instructed by the shape of a central fixation point. The prosaccade goal was indicated by a peripheral stimulus flashed anywhere on the screen, whereas the antisaccade goal was an unmarked site diametrically opposite the flashed stimulus. The visual cue was given immediately after the instruction cue disappeared in the immediate-saccade task, or during the instruction period in the delayed-saccade task. The instruction cue offset was the saccade gosignal. Here we focus on 92 task-related neurons: visual, eye-movement, and instruction/fixation neurons. We found that 73% of SEF eye-movement-related neurons fired significantly more before antisaccades than prosaccades. This finding was analyzed at 3 levels: population, single neuron, and individual trial. On individual antisaccade trials, 40 ms before saccade, the firing rate of eye-movementrelated neurons was highly predictive of successful performance. A similar analysis of visual responses (40 ms astride the peak) gave less-coherent results. Fixation neurons, activated during the initial instruction period (i.e., after the instruction cue but before the stimulus) always fired more on antisaccade than on prosaccade trials. This trend, however, was statistically significant for only half of these neurons. We conclude that the SEF is critically involved in the production of antisaccades.