How cerebellar motor learning keeps saccades accurate.

The neuronal substrate underlying the learning of a sophisticated task has been difficult to study. However, the advent of a behavioral paradigm that deceives the saccadic system into thinking it is making an error has allowed the mechanisms of the adaptation that corrects this error to be revealed in a primate. The neural elements that fashion the command signal for the generation of accurate saccades involve subcortical structures in the brain stem and cerebellum. In this review we show that sites in both those structures also are involved with the gradual adaptation of saccade size, a form of motor learning. Pharmacological manipulation of the oculomotor vermis (lobules VIc and VII) impairs mechanisms that either increase or decrease saccade size during adaptation. The net saccade-related simple spike (SS) activity of its Purkinje cells is correlated with the changes in saccade characteristics that occur during adaptation. These changes in SS activity are driven by an error signal delivered over climbing fibers, which create complex spikes whose probability of occurrence reflects the motor error between the actual and desired saccade size. These climbing fibers originate in the part of the inferior olive that receives projections from the superior colliculus (SC). Disabling the SC prevents adaptation and stimulation of the SC just after a normal saccade produces a surrogate error signal that drives adaptation without an actual visual error. Therefore, the SC provides not only the initial command that generates a saccade, as shown by others, but also the error signal that ensures that saccades remain accurate.

Adaptation and adaptation transfer characteristics of five different saccade types in the monkey.

Shifts in the direction of gaze are accomplished by different kinds of saccades, which are elicited under different circumstances. Saccade types include targeting saccades to simple jumping targets, delayed saccades to visible targets after a waiting period, memory-guided (MG) saccades to remembered target locations, scanning saccades to stationary target arrays, and express saccades after very short latencies. Studies of human cases and neurophysiological experiments in monkeys suggest that separate pathways, which converge on a common locus that provides the motor command, generate these different types of saccade. When behavioral manipulations in humans cause targeting saccades to have persistent dysmetrias as might occur naturally from growth, aging, and injury, they gradually adapt to reduce the dysmetria. Although results differ slightly between laboratories, this adaptation generalizes or transfers to all the other saccade types mentioned above. Also, when one of the other types of saccade undergoes adaptation, it often transfers to another saccade type. Similar adaptation and transfer experiments, which allow inferences to be drawn about the site(s) of adaptation for different saccade types, have yet to be done in monkeys. Here we show that simian targeting and MG saccades adapt more than express, scanning, and delayed saccades. Adaptation of targeting saccades transfers to all the other saccade types. However, the adaptation of MG saccades transfers only to delayed saccades. These data suggest that adaptation of simian targeting saccades occurs on the pathway common to all saccade types. In contrast, only the delayed saccade command passes through the adaptation site of the MG saccade.

An experimental vestibular neural prosthesis: design and preliminary results with rhesus monkeys stimulated with modulated pulses.

A vestibular neural prosthesis was designed on the basis of a cochlear implant for treatment of Meniere's disease and other vestibular disorders. Computer control software was developed to generate patterned pulse stimuli for exploring optimal parameters to activate the vestibular nerve. Two rhesus monkeys were implanted with the prototype vestibular prosthesis and they were behaviorally evaluated post implantation surgery. Horizontal and vertical eye movement responses to patterned electrical pulse stimulations were collected on both monkeys. Pulse amplitude modulated (PAM) and pulse rate modulated (PRM) trains were applied to the lateral canal of each implanted animal. Robust slow-phase nystagmus responses following the PAM or PRM modulation pattern were observed in both implanted monkeys in the direction consistent with the activation of the implanted canal. Both PAM and PRM pulse trains can elicit a significant amount of in-phase modulated eye velocity changes and they could potentially be used for efficiently coding head rotational signals in future vestibular neural prostheses.

Auditory outcomes following implantation and electrical stimulation of the semicircular canals.

We measured auditory brainstem responses (ABRs) in eight Rhesus monkeys after implantation of electrodes in the semicircular canals of one ear, using a multi-channel vestibular prosthesis based on cochlear implant technology. In five animals, click-evoked ABR thresholds in the implanted ear were within 10 dB of thresholds in the non-implanted control ear. Threshold differences in the remaining three animals varied from 18 to 69 dB, indicating mild to severe hearing losses. Click- and tone-evoked ABRs measured in a subset of animals before and after implantation revealed a comparable pattern of threshold changes. Thresholds obtained five months or more after implantation--a period in which the prosthesis regularly delivered electrical stimulation to achieve functional activation of the vestibular system--improved in three animals with no or mild initial hearing loss and increased in a fourth with a moderate hearing loss. These results suggest that, although there is a risk of hearing loss with unilateral vestibular implantation to treat balance disorders, the surgery can be performed in a manner that preserves hearing over an extended period of functional stimulation.

Effect of inactivation and disinhibition of the oculomotor vermis on saccade adaptation.

The ability to adapt a variety of motor acts to compensate for persistent natural or artificially induced errors in movement accuracy requires the cerebellum. For adaptation of the rapid shifts in the direction of gaze called saccades, the oculomotor vermis (OMV) of the cerebellum must be intact. We disrupted the neural circuitry of the OMV by manipulating gamma aminobutyric acid (GABA), the transmitter used by many neurons in the vermis. We injected either muscimol, an agonist of GABA, to inactivate the OMV or bicuculline, an antagonist, to block GABA inhibition. Our previous study showed that muscimol injections cause ipsiversive saccades to fall short of their targets, whereas bicuculline injections cause most ipsiversive saccades to overshoot. Once these dysmetrias had stabilized, we tested the monkey's ability to adapt saccade size to intra-saccadic target steps that produced a consistent saccade under-shoot (amplitude increase adaptation required) or overshoot (amplitude decrease adaptation required). Injections of muscimol abolished the amplitude increase adaptation of ipsiversive saccades, but had either no effect, or occasionally facilitated, amplitude decrease adaptation. In contrast, injections of bicuculline impaired amplitude decrease adaptation and usually facilitated amplitude increase adaptation. Neither drug produced consistent effects on the adaptation of contraversive saccades. Taken together, these data suggest that OMV activity is necessary for amplitude increase adaptation, whereas amplitude decrease adaptation may involve the inhibitory circuits within the OMV.

Behavior of the oculomotor vermis for five different types of saccade.

Single unit and lesion studies have implicated the oculomotor vermis of the cerebellum in the control of targeting saccades to jumping visual targets. However, saccades can be made in a variety of other target situations where they can occur with different reaction times (express or delayed saccades) in response to a remembered target location (memory-guided saccades) or between several targets that are always visible (scanning saccades). Here we ask whether the oculomotor vermis contributes to generating all these types of saccades by examining the simple spike discharge of its Purkinje cells. Twenty-six of 32 P-cells (81%) exhibited qualitatively similar phasic firing patterns for targeting, express, scanning, delayed, and memory-guided saccades. The remaining six exhibited a different pattern for just scanning saccades. Although a sensitive test of discharge patterns revealed significant differences for some pairs of saccade types in ∼29% of P-cells, there was no cell-to-cell consistency as to which pairs were associated with different patterns. Also, a less sensitive comparison identified substantially fewer cells (∼15%) with different patterns. Thus the lack of any consistent difference in firing for different saccade types leads us to conclude that the oculomotor vermis is not likely to contribute differently to targeting, express, scanning, delayed, or memory-guided saccades.

Effects of GABA agonist and antagonist injections into the oculomotor vermis on horizontal saccades.

The oculomotor vermis (OMV) of the cerebellum is necessary for the generation of the accurate rapid eye movements called saccades. Large lesions of the midline cerebellar cortex involving the OMV cause saccades to become hypometric and more variable. However, saccades were not examined immediately after these lesions so the interpretation of the resulting deficits might have been contaminated by some adaptation to the saccade dysmetria. Therefore, to better understand the contribution of the OMV to normal saccades, we impaired its operation locally by injecting small amounts of either an agonist or antagonist of γ-aminobutyric acid (GABA), which is a ubiquitous neurotransmitter throughout the cerebellar cortex. Muscimol, a GABA agonist, inactivated part of the OMV, whereas bicuculline, an antagonist, disinhibited it. Muscimol caused all ipsiversive horizontal saccades from 5 to 30° to become hypometric. In contrast, bicuculline produced an amplitude-dependent dysmetria: ipsiversive horizontal saccades elicited by target steps <10° became hypometric, whereas those in response to larger steps became hypermetric. At the transition target amplitude, saccade amplitudes were quite variable with some being hypo- and others hypermetric. After most injections of either agent, saccades had lower peak velocities and longer durations than pre-injection saccades of the same amplitude. The longer durations were associated with a prolongation of the deceleration phase. Both agents produced inconsistent effects on contraversive saccades. These results establish that the oculomotor vermis helps control the characteristics of normal ipsiversive saccades and that GABAergic inhibitory processes are a crucial part of this process.

Changes in simple spike activity of some Purkinje cells in the oculomotor vermis during saccade adaptation are appropriate to participate in motor learning.

Adaptation of saccadic eye movements provides an excellent motor learning model to study theories of neuronal plasticity. When primates make saccades to a jumping target, a backward step of the target during the saccade can make it appear to overshoot. If this deception continues for many trials, saccades gradually decrease in amplitude to go directly to the back-stepped target location. We used this adaptation paradigm to evaluate the Marr-Albus hypothesis that such motor learning occurs at the Purkinje (P)-cell of the cerebellum. We recorded the activity of identified P-cells in the oculomotor vermis, lobules VIc and VII. After documenting the on and off error directions of the complex spike activity of a P-cell, we determined whether its saccade-related simple spike (SS) activity changed during saccade adaptation in those two directions. Before adaptation, 57 of 61 P-cells exhibited a clear burst, pause, or a combination of both for saccades in one or both directions. Sixty-two percent of all cells, including two of the four initially unresponsive ones, behaved differently for saccades whose size changed because of adaptation than for saccades of similar sizes gathered before adaptation. In at least 42% of these, the changes were appropriate to decrease saccade amplitude based on our current knowledge of cerebellum and brainstem saccade circuitry. Changes in activity during adaptation were not compensating for the potential fatigue associated with performing many saccades. Therefore, many P-cells in the oculomotor vermis exhibit changes in SS activity specific to adapted saccades and therefore appropriate to induce adaptation.

Head-free gaze shifts provide further insights into the role of the medial cerebellum in the control of primate saccadic eye movements.

This study examines how signals generated in the oculomotor cerebellum could be involved in the control of gaze shifts, which rapidly redirect the eyes from one object to another. Neurons in the caudal fastigial nucleus (cFN), the output of the oculomotor cerebellum, discharged when monkeys made horizontal head-unrestrained gaze shifts, composed of an eye saccade and a head movement. Eighty-seven percent of our neurons discharged a burst of spikes for both ipsiversive and contraversive gaze shifts. In both directions, burst end was much better timed with gaze end than was burst start with gaze start, was well correlated with eye end, and was poorly correlated with head end or the time of peak head velocity. Moreover, bursts accompanied all head-unrestrained gaze shifts whether the head moved or not. Therefore we conclude that the cFN is not part of the pathway that controls head movement. For contraversive gaze shifts, the early part of the burst was correlated with gaze acceleration. Thereafter, the burst of the neuronal population continued throughout the prolonged deceleration of large gaze shifts. For a majority of neurons, gaze duration was correlated with burst duration; for some, gaze amplitude was less well correlated with the number of spikes. Therefore we suggest that the population burst provides an acceleration boost for high acceleration (smaller) contraversive gaze shifts and helps maintain the drive required to extend the deceleration of large contraversive gaze shifts. In contrast, the ipsiversive population burst, which is less well correlated with gaze metrics but whose peak rate occurs before gaze end, seems responsible primarily for terminating the gaze shift.

Identifying sites of saccade amplitude plasticity in humans: transfer of adaptation between different types of saccade.

To view different objects of interest, primates use fast, accurate eye movements called saccades. If saccades become inaccurate, the brain adjusts their amplitudes so they again land on target, a process known as saccade adaptation. The different types of saccades elicited in different behavioral circumstances appear to utilize different parts of the oculomotor circuitry. To gain insight into where adaptation occurs in different saccade pathways, we adapted saccades of one type and examined how that adaptation affected or transferred to saccades of a different type. If adaptation of one type of saccade causes a substantial change in the amplitude of another, that adaptation may occur at a site used in the generation of both types of saccade. Alternatively, if adaptation of one type of saccade transfers only partially, or not at all, to another, adaptation occurs at least in part at a location that is not common to the generation of both types of saccade. We produced significant amplitude reductions in memory-guided, delayed, targeting and express saccades by moving the target backward during the saccade. After memory-guided saccades were adapted, the amplitude of express, targeting and delayed saccades exhibited only a partial reduction. In contrast, when express, targeting, or delayed saccades were adapted, amplitude transfer to memory-guided saccades was more substantial. These results, combined with previously published data, suggest that there are at least two sites of adaptation within the saccadic system. One is used communally in the generation of express, targeting, delayed and memory-guided saccades, whereas the other is specific for the generation of memory-guided saccades.

Subthreshold activation of the superior colliculus drives saccade motor learning.

How the brain learns and maintains accurate precision movements is currently unknown. At times throughout life, rapid gaze shifts (saccades) become inaccurate, but the brain makes gradual adjustments so they again stop on target. Previously, we showed that complex spikes (CSs) in Purkinje cells of the oculomotor cerebellum report the direction and amplitude by which saccades are in error. Anatomical studies indicate that this error signal could originate in the superior colliculus (SC). Here, we deliver subthreshold electrical stimulation of the SC after the saccade lands to signal an apparent error. The size of saccades in the same direction as the simulated error gradually increase; those in the opposite direction decrease. The electrically adapted saccades endure after stimulation is discontinued, exhibit an adaptation field, can undergo changes in direction, and depend on error timing. These electrically induced adaptations were virtually identical with those produced by the visually induced adaptations that we report here for comparable visual errors in the same monkeys. Therefore, our experiments reveal that an additional role for the SC in the generation of saccades is to provide a vector error signal that drives dysmetric saccades to adapt. Moreover, the characteristics of the electrically induced adaptation reflect those of error-related CS activity in the oculomotor cerebellum, suggesting that CS activity serves as the learning signal. We speculate that CS activity may serve as the error signal that drives other kinds of motor learning as well.

Complex spike activity in the oculomotor vermis of the cerebellum: a vectorial error signal for saccade motor learning?

Brain stem signals that generate saccadic eye movements originate in the superior colliculus. They reach the pontine burst generator for horizontal saccades via short-latency pathways and a longer pathway through the oculomotor vermis (OMV) of the cerebellum. Lesion studies implicate the OMV in the adaptation of saccade amplitude that occurs when saccades become inaccurate because of extraocular muscle weakness or behavioral manipulations. We studied the nature of the possible error signal that might drive adaptation by examining the complex spike (CS) activity of vermis Purkinje (P-) cells in monkeys. We produced a saccade error by displacing the target as a saccade was made toward it; a corrective saccade approximately 200 ms later eliminated the resulting error. In most P-cells, the probability of CS firing changed, but only in the error interval between the primary and corrective saccade. For most P-cells, CSs occurred in a tight cluster approximately 100 ms after error onset. The probability of CS occurrence depended on both error direction and size. Across our sample, all error directions were represented; most had a horizontal component. In more than one half of our P-cells, the probability of CS occurrence was greatest for error sizes<5 degrees and less for larger errors. In the remaining cells, there was a uniform increased probability of CS occurrence for all errors

Activity changes in monkey superior colliculus during saccade adaptation.

Saccades are eye movements that are used to foveate targets rapidly and accurately. Their amplitude must be adjusted continually, throughout life, to compensate for movement inaccuracies due to maturation, pathology, or aging. One possible locus for such saccade adaptation is the superior colliculus (SC), the relay for cortical commands to the premotor brain stem generator for saccades. However, previous stimulation and recording studies have disagreed as to whether saccade adaptation occurs up- or downstream of the SC. Therefore we have reexamined the behavior of SC burst neurons during saccade adaptation under conditions that were optimized to produce the biggest possible change in neuronal activity. We show that behavioral adaptation of saccade amplitude was associated with significant increases or decreases, in the number of spikes in the burst and/or changes in the shape of the movement field in 35 of 43 SC neurons tested. Of the 35, 29 had closed movement fields and 14 were classified indeterminate because the movement field could not be definitively diagnosed. Changes in the number of spikes occurred gradually during adaptation and resulted from correlated changes in burst lead and duration without consistent changes in peak burst rate. These data indicate that the great majority of SC neurons show a change in discharge in association with saccade amplitude adaptation. Based on these and previous results, we speculate that the site for saccade adaptation resides in the SC or that the SC is the final common pathway for adaptive changes that occur elsewhere in the saccade system.

Contribution of the frontal eye field to gaze shifts in the head-unrestrained monkey: effects of microstimulation.

The role of the primate frontal eye field (FEF) has been inferred primarily from experiments investigating saccadic eye movements with the head restrained. Three recent reports investigating head-unrestrained gaze shifts disagree on whether head movements are evoked with FEF stimulation and thus whether the FEF participates in gaze movement commands. We therefore examined the eye, head, and overall gaze movement evoked by low-intensity microstimulation of the low-threshold region of the FEF in two head-unrestrained monkeys. Microstimulation applied at 200 or 350 Hz for 200 ms evoked large gaze shifts with substantial head movement components from most sites in the dorsomedial FEF, but evoked small, predominantly eye-only gaze shifts from ventrolateral sites. The size and direction of gaze and eye movements were strongly affected by the eye position before stimulation. Head movements exhibited little position dependency, but at some sites and initial eye positions, head-only movements were evoked. Stimulus-evoked gaze shifts and their eye and head components resembled those elicited naturally by visual targets. With stimulus train durations >200 ms, the evoked gaze shifts were more likely to be accomplished with a substantial head movement, which often continued for the entire stimulus duration. The amplitude, duration and peak velocity of the evoked head movement were more strongly correlated with stimulus duration than were those of the gaze or eye movements. We conclude that the dorsomedial FEF generates a gaze command signal that can produce eye, head, or combined eye-head movement depending on the initial orbital position of the eye.

Complex spike activity of purkinje cells in the oculomotor vermis during behavioral adaptation of monkey saccades.

Throughout life, the oculomotor system can correct itself when saccadic eye movements become inaccurate. This adaptation mechanism can be engaged in the laboratory by displacing the target when the saccade toward it is in flight. Forward and backward target displacements cause gradual increases and decreases in saccade amplitude, respectively. Equipped with this paradigm, we asked whether Purkinje cells (P-cells) in the vermis of the oculomotor cerebellum, lobules VIc and VII, changed their complex spike (CS) discharge during the behavioral adaptation of horizontal saccades. We tested the hypothesis that CS activity would change only when a targeting saccade caused an error in eye position relative to the target, i.e., during the error interval between the primary and corrective saccades. We examined only those P-cells whose simple spike activity exhibited either a burst or pause with saccades in several directions. Approximately 80% of such P-cells exhibited an increase in CS activity during the error interval when the adaptation paradigm imposed horizontal eye-position errors in one direction and a decrease in activity for errors in the other. As adaptation progressed and errors were reduced, there was no consistent change in the CS activity. These data suggest that the CS activity of P-cells in the oculomotor vermis signals the direction but not the magnitude of eye-position error during saccade adaptation. Our results are consistent with cerebellar learning models that have been proposed to explain adaptation of the vestibulo-ocular reflex so similar mechanisms may also underlie plasticity of this precision voluntary oculomotor behavior.

Amplitude adaptation occurs where a saccade is represented as a vector and not as its components.

When saccades become inaccurate, their amplitude is adapted. We examined, in humans, whether this adaptation occurs where the saccade is represented as a vector or as its horizontal and vertical components. In one experiment, we behaviorally reduced the amplitude of clockwise oblique saccades and examined the transfer to saccades made to other target amplitudes and directions. In a second, we adapted rightward saccades of the same size as the rightward component of the clockwise oblique saccades and examined the effect on oblique saccades. The results of both experiments imply that adaptation occurs where the saccade command is represented as a vector.

Effect of pharmacological inactivation of nucleus reticularis tegmenti pontis on saccadic eye movements in the monkey.

The superior colliculus (SC) provides signals for the generation of saccades via a direct pathway to the brain stem burst generator (BG). In addition, it sends saccade-related activity to the BG indirectly through the cerebellum via a relay in the nucleus reticularis tegmenti pontis (NRTP). Lesions of the oculomotor vermis, lobules VIc and VII, and inactivation of the caudal fastigial nucleus, the cerebellar output nucleus to which it projects, produce saccade dysmetria but have little effect on saccade peak velocity and duration. We expected similar deficits from inactivation of the NRTP. Instead, injections as small as 80 nl into the NRTP first slowed ipsiversive saccades and then gradually reduced their amplitudes. Postinjection saccades had slower peak velocities and longer durations than preinjection saccades with similar amplitudes. Contraversive saccades retained their normal kinematics. When the gains of ipsiversive saccades to 10 degrees target steps had fallen to their lowest values (0.28 +/- 0.19; mean +/- SD; n = 10 experiments), the gains of contraversive saccades to 10 degrees target steps had decreased very little (0.82 +/- 0.11). Eventually, ipsiversive saccades did not exceed 5 degrees , even to 20 degrees target steps. Moreover, these small remaining saccades apparently were made with considerable difficulty because their latencies increased substantially. When ipsiversive saccade gain was at its lowest, the gain and kinematics of vertical saccades to 10 degrees target steps exhibited inconsistent changes. We argue that our injections did not compromise the direct SC pathway. Therefore these data suggest that the cerebellar saccade pathway does not simply modulate BG activity but is required for horizontal saccades to occur at all.

Behavior of the position vestibular pause (PVP) interneurons of the vestibuloocular reflex during head-free gaze shifts in the monkey.

Most behavioral studies indicate that the efficacy (gain) of the vestibuloocular reflex (VOR) in primates is modulated during the voluntary head movements that accompany large shifts in the direction of gaze. However, the timing and degree of this modulation is the subject of some debate. The neurophysiological substrate for this apparent gain reduction has been sought in the behavior of the type I position vestibular pause (PVP) neuron, a well-known type of interneuron in the direct VOR pathway. With the head fixed, PVPs increase their firing rates with contraversive eye position and with ipsiversive passive head rotation and also cease firing (pause) for the duration of ipsiversive saccades. During head-free ipsiversive gaze shifts, the eyes and head move in the same direction. If the vestibular signal carried by PVPs provides the primary drive for the VOR, the vestibular signal should be present during ipsiversive gaze shifts to the degree that the VOR is present. Of 25 type I PVPs recorded, 21 ceased their discharge for the entire duration of the rapid, eye-saccade component of an ipsiversive gaze shift. The resumption of activity occurred, on average, 13 ms after the end of the saccade. These results suggest that the activity of the vast majority of PVP neurons do not reflect the state of the VOR, but rather PVPs are completely eliminated from participation in the reflex during head-free gaze movements. We conclude that if any modulation of the VOR does exist, it must occur through other, probably longer-latency, pathways.

The characteristics and neuronal substrate of saccadic eye movement plasticity.

Saccadic eye movements are shifts in the direction of gaze that rapidly and accurately aim the fovea at targets of interest. Saccades are so brief that visual feedback cannot guide them to their targets. Therefore, the saccadic motor command must be accurately specified in advance of the movement and continually modified to compensate for growth, injury, and aging, which otherwise would produce dysmetric saccades. When a persistent dysmetria occurs in subjects with muscle weakness or neural damage or is induced in normal primates by the surreptitious jumping of a target forward or backward as a saccade is made to acquire the target, saccadic amplitude changes to reduce the dysmetria. Adaptation of saccadic amplitude or direction occurs gradually and is retained in the dark, thus representing true motor plasticity. Saccadic adaptation is more rapid in humans than in monkeys, usually is incomplete in both species, and is slower and less robust for amplitude increases than decreases. Adaptation appears to be motor rather than sensory. In humans, adaptation of saccades that would seem to require more sensory-motor processing does not transfer to saccades that seem to require less, suggesting the existence of distributed adaptation loci. In monkeys, however, transfer from more simple to more complex saccades is robust, suggesting a common adaptation site. Neurophysiological data from both species indicate that the oculomotor cerebellum is crucial for saccadic adaptation. This review shows that the precise, voluntary behaviors known as saccadic eye movements provide an alternative to simple reflexes for the study of the neuronal basis of motor learning.

Discharge patterns of cerebellar output neurons in the caudal fastigial nucleus during head-free gaze shifts in primates.

Lesion studies in both human and non-human primates indicate that the cerebellum is important for accurate and stereotyped saccadic eye movements. Based on single-unit recordings and pharmacological inactivations in head-fixed monkeys, we suggested that the caudal fastigial nucleus (CFN) provides the brainstem saccade generator with a burst that helps accelerate contraversive saccades and decelerate ipsiversive ones. Here we examine this suggestion during head-free gaze shifts where there can be a 10-fold difference in saccade duration. First, the timing of the burst does not depend on whether the gaze shift has a head component. When a family of either ipsiversive or contraversive gaze shifts with a variety of saccadic durations is aligned on gaze onset, the high-frequency burst in the associated rasters occurs progressively later as saccade duration increases. Realignment of the same rasters with the end of the saccade reveals a tight timing of burst end with saccade end for all 10 CFN burst neurons studied. The delayed bursts for contraversive saccades were unexpected based on the early burst illustrated in the published head-fixed data. One hypothesis is that the late activity helps terminate contraversive as well as ipsiversive gaze shifts. An alternative explanation is that the late CFN burst could still be used as an excitatory drive to promote the late reacceleration or prolonged velocity plateau that is present during large gaze shifts.