The Journal of Neurophysiology Vol. 78 No. 2 August 1997, pp. 848-859
Copyright ©1997 by the American Physiological Society

Participation of Caudal Fastigial Nucleus in Smooth Pursuit Eye Movements. II. Effects of Muscimol Inactivation

Farrel R. Robinson¹, Andreas Straube², and Albert F. Fuchs¹

¹ Department of Physiology and Biophysics and Regional Primate Research Center, University of Washington, Seattle, Washington 98195-7330; and ² Department of Neurology, Grosshadern Clinic, University of Munich, 81377 Munich, Germany

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Robinson, Farrel, R., Andreas Straube, and Albert F. Fuchs. Participation of the caudal fastigial nucleus in smooth pursuit eye movements. II. Effects of muscimol inactivation. J. Neurophysiol. 78: 848-859, 1997. We studied the effect of temporarily inactivating the caudal fastigial nucleus (CFN) in three rhesus macaques trained to make smooth pursuit eye movements. We injected the -aminobutyric acid A agonist muscimol into one or both CFNs where we had recorded pursuit-related neurons a few minutes earlier. Inactivating the CFN on one side impaired pursuit in one monkey so severely that it could not follow step-ramp targets moving at 20°/s, the target velocity that we used to test the other two monkeys. We tested this monkey with targets moving at 10°/s. In all three monkeys, unilateral CFN inactivation either increased the acceleration of ipsilateral step-ramp pursuit (in 2 monkeys, to 144 and 220% of normal) or decreased the acceleration of contralateral pursuit (in 1 monkey, to 71% of normal). Muscimol injected into both CFNs in two of the monkeys left both ipsilateral and contralateral acceleration nearly normal in both monkeys (101% of normal). Unilateral CFN inactivation also impaired the velocity of maintained pursuit as the monkeys tracked a target moving at a constant velocity or oscillating sinusoidally. Averaged across both types of movements in all three monkeys, gains for ipsilateral, contralateral, upward, and downward pursuit were 94, 67, 84, and 73% of normal, respectively. Unilateral CFN inactivation also impaired the monkeys' ability to suppress their vestibuloocular reflex (VOR). Averaged across the two monkeys VOR gain during suppression increased from 0.06 to 0.25 during yaw rotation and from 0.21 to 0.59 during pitch rotation. Bilateral CFN inactivation reduced pursuit gains in all directions more than unilateral injection did. In the two monkeys tested, ipsilateral, contralateral, upward, and downward gains went from 94, 86, 85, and 74% of normal, respectively, after we inactivated one CFN to 88, 73, 80, and 64% of normal after we also inactivated the second CFN. We can explain many, but not all, of the effects of CFN activation on smooth pursuit with the behavior of CFN neurons, and the assumption that the activity of each CFN neuron helps drive pursuit movements in the direction that best activates that neuron. We conclude that the CFN, like the flocculus-ventral paraflocculus, is a cerebellar region involved in control of smooth pursuit.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Removal of the cerebellum abolishes a monkey's ability to make smooth pursuit eye movements (Burde et al. 1975; Westheimer and Blair 1973). Currently we know about two separate parts of the cerebellum that participate in smooth pursuit. One is the ventrolateral cortex, including the flocculus and ventral paraflocculus. The other is the caudal medial region of the cerebellum, including lobules VI and VII of the vermal cortex and the region in the caudal fastigial nucleus (CFN) to which this cortex projects (Yamada and Noda 1987).

Earlier findings strongly implicated the flocculus and ventral paraflocculus, grouped together earlier as the "flocculus" (Gerrits and Voogd 1989), in pursuit. The flocculus and ventral paraflocculus contain many Purkinje cells that discharge during pursuit at rates proportional to eye velocity (Lisberger and Fuchs 1978; Miles et al. 1980; Stone and Lisberger 1990). Evidence that these cells indeed are involved in smooth pursuit comes from the findings that electrical stimulation in these structures elicits slow eye movements (Belknap and Noda 1987; Lisberger and Pavelko 1988; Shidara and Kawano 1993) and that bilateral ablation of the flocculus, ventral paraflocculus, and part of the dorsal paraflocculus reduces smooth pursuit gain to ~64% of normal (Zee et al. 1981).

Because bilateral removal of the flocculus and ventral paraflocculus does not abolish pursuit but complete cerebellectomy does, we infer that some other part of the cerebellum must participate in pursuit. One candidate area is the posterior medial part of the cerebellum. Lobules VI and VII of the cerebellar vermis receive a projection from the dorsolateral pontine nucleus (Brodal 1979; Glickstein et al. 1994; Thielert and Thier 1993; Yamada and Noda 1987). This projection implicates lobules VI and VII in pursuit because several lines of evidence indicate that the dorsolateral pons participates in pursuit. First, the dorsolateral pons contains neurons that discharge during pursuit (Mustari et al. 1988; Suzuki and Keller 1984; Thier et al. 1988). Second, electrical stimulation in the dorsolateral pontine nuclei evokes smooth eye movements (May et al. 1985). Third, lesions in the dorsolateral pons cause deficits in both initiation and maintenance of pursuit (May et al. 1988). Furthermore, in lobules VI and VII of the cerebellar vermis, where dorsolateral pontine projections terminate, the activity of some Purkinje cells modulates well during smooth pursuit (Suzuki and Keller 1988a,b; Suzuki et al. 1981). This activity probably influences pursuit because electrical stimulation of lobules VI and VII reliably alters ongoing pursuit (Krauzlis and Miles 1994).

The Purkinje cells in lobules VI and VII project to the caudal part of the most medial, or fastigial, nucleus of the cerebellum (Yamada and Noda 1987), raising the possibility that this region, the CFN, participates in pursuit. Two other facts also implicate the CFN in smooth pursuit. First, the CFN, like vermal lobules VI and VII, receives a large direct input from the dorsolateral pontine nucleus (Noda et al. 1990). Second, the activity of CFN neurons modulates reliably and vigorously during pursuit (Fuchs et al. 1994b).

If the CFN is important to visual pursuit, inhibition of CFN activity should affect pursuit movements. To investigate this possibility, we injected muscimol into the CFN and studied the effects on smooth pursuit characteristics. We have published preliminary reports on the CFN role in smooth pursuit previously (Fuchs et al. 1994a; Robinson et al. 1994).

	METHODS

Abstract Introduction Methods Results Discussion References

Animal preparation and recording

The subjects for these experiments were three adolescent male rhesus macaques (Macaca mulatta) that were trained to follow a moving target spot with their eyes for an applesauce reward. The surgical preparation, training, and recording procedures are described in a previous paper (Fuchs et al. 1994b). Briefly, in a sterile surgical procedure with the monkey under deep anesthesia, we placed a wire coil around one eye so that we could measure eye position with the electromagnetic search coil technique (Robinson 1963). We also cut a small hole in the top of the skull, and over it implanted a stainless steel cylinder, 21 mm diam, aimed with a stereotaxic apparatus at the midline between the two CFNs. Finally, we secured acrylic lugs to the monkey's head so that the head could be stabilized during the recording sessions.

Muscimol injections

We collected smooth pursuit data from monkeys 1 and 2 after the same muscimol injections that we used to study the role of the CFN in saccade generation (Robinson et al. 1993). Monkey 3 was not part of that earlier study, but we used the same procedures with it as with the other two. Before injecting muscimol into the cerebellar nuclei, we made electrode penetrations into the intended injection site to locate pursuit-related neurons. Once we found the region of pursuit-related activity in the CFN, we withdrew the electrode and replaced it with an injection pipette. This pipette consisted of a length of 32-gauge hypodermic tubing with a pulled glass micropipette tip glued over one end. The glass tip had an inner diameter of ~25 µm. The pipette was connected by several feet of polyethylene tubing to a solenoid valve. We put the pipette in the same guide tube that we had just used for recording. We advanced the tip of the pipette to the dorsal-ventral center of the region from which we had just recorded pursuit-related activity.

To make the injections, we filled the pipette and several centimeters of the connected tube with a solution consisting of 1 mg/ml (8.75 mM) muscimol in normal saline. We injected the muscimol solution using a solenoid valve system (WPI PV830) with which we delivered brief (~15 ms) air pressure pulses to move the meniscus a calibrated distance down the tube. We used a ×50 microscope to view the meniscus and measure its advance down the tube. We were able to resolve injected volumes to within ~10 nl. Each injection took 5-20 min and consisted of 0.5-1.0 µl of muscimol solution. After each injection we waited 5 min and then withdrew the injection pipette.

We performed seven unilateral injections in three monkeys and three bilateral injections in two monkeys. We also did three control injections outside the CFN in two monkeys. In one control experiment, we injected muscimol into the fastigial nucleus 2.5 mm rostral to the most rostral eye-movement-related cells. In the other two, we injected muscimol 4 mm lateral to the pursuit-related cells of the fastigial nucleus in an area where we also had recorded pursuit-related neurons. Histological reconstructions confirmed thatthis lateral eye-movement area is the "distinct region" of eye movement neurons on the border between the posterior interpositus and the lateral cerebellar nuclei described by Van Kan et al. (1993).

Stimulus conditions

Immediately before each injection, we recorded normal eye movements while the monkeys tracked a laser spot projected onto a screen. The horizontal and vertical positions of the laser spot were controlled by mirror galvanometers in the light path. We controlled these galvanometers with a computer or a function generator. We recorded smooth-pursuit eye movements elicited by a target moving with a step-ramp time course (Rashbass 1961). Step-ramp movement consists of a rapid step of the target in one direction followed immediately by constant-velocity movement in the opposite direction. The size of the initial step was adjusted to the subsequent constant velocity so that the target crossed its starting position in 100-200 ms and the monkey initiated pursuit smoothly without catch-up saccades. Except where noted, the target always started at the center of gaze and moved toward the periphery. Horizontal and vertical target velocities were usually 20°/s, but we tested pursuit in monkey 2 with targets moving at 10°/s. This monkey could not track targets at 20°/s after CFN inactivation, though it could when its CFN was normal.

We also had the monkeys track targets oscillating ±10° with a sinusoidal velocity variation. We measured eye movements in response to horizontal and vertical target oscillation frequencies of 0.25, 0.5, and 0.8 Hz. By oscillating the chair in which the monkey sat with its head stabilized, we also were able to measure the animals' ability to suppress the vestibuloocular reflex (VOR). For this test, we used oscillations of ±10°, 0.5 Hz, in either the yaw or pitch plane as the monkey fixated a target spot that moved with the chair rotations.

Data collection and analysis

We recorded analog voltages corresponding to horizontal and vertical eye and target positions on video tape using a PCM recording adapter (Vetter 4000A) and digitized these signals from the tape. We analyzed step-ramp tracking with the analysis program described previously (Fuchs et al. 1994b). Briefly, we used the program to scroll through the record of eye movements, remove saccades, and save measurements of several parameters for each pursuit eye movement.

To quantify the changes caused by CFN inactivation, we made two measurements of step-ramp tracking, average acceleration, the change in velocity between the onset of pursuit and the peak velocity at the end of acceleration, and maintained pursuit gain, i.e., (eye velocity)/(target velocity) during the period after acceleration when eye velocity was nearly constant. We sorted step-ramp tracking movements into those toward the side of the inactivated CFN, called ipsilateral movements, and those away from the injected side, called contralateral movements. Note that, because we injected into the right CFN in some experiments and the left CFN in the others, the averaged measurements of ipsilateral and contralateral postinjection movements each include both rightward and leftward movements. Because our postinjection measurements included both rightward and leftward movements, we pooled rightward and leftward preinjection movements to make up our normal measurements. We compared ipsilateral and contralateral movements after CFN inactivation with normal movements.

To characterize how CFN inactivation affected sinusoidal pursuit, we printed selected epochs of eye position and velocity on a thermal strip chart recorder (Astro Med 9500). A hardware circuit differentiated the eye position voltage to provide us with eye velocity. We measured the peak velocity of each half cycle of eye movement and calculated pursuit gain as (peak eye velocity)/(peak target velocity).

We compared the means of our measurements with a factorial analysis of variance (ANOVA) using the Bonferroni correction for repeated tests. We considered P < 0.05 as significant. With the Bonferroni correction this means that, for example, with 10 possible pair-wise comparisons among five different means, an individual P value had to be 0.005 to be significant.

	RESULTS

Abstract Introduction Methods Results Discussion References

The oculomotor effects of the muscimol injections were evident within ~10 min after the end of the injection and were still clear with no apparent changes at the end of the recording 90-120 min later. The monkeys completely recovered from the effects of the muscimol within 24 h. After each muscimol injection into one CFN, all monkeys exhibited the saccade dysmetria, abnormal saccade trajectories, and eye position offset that we have described earlier (Robinson et al. 1993). In addition to these abnormalities, horizontal smooth pursuit was interrupted by many more contralateral saccades than ipsilateral saccades. Before the injection, sinusoidal pursuit contained roughly equal numbers of catch-up saccades in each direction, i.e., monkeys 1, 2, and 3 made 56, 59, and 50% (n = 34, 63, 111) of their saccades to the right. With one CFN inactivated, the monkeys' eyes fell behind a target moving contralaterally, so that 93, 81, and 73% (n = 232, 215, 177) of their corrective saccades were away from the side of the injection. This increase in the proportion of contralateral catch-up saccades was a clear qualitative indication that pursuit was abnormal after CFN inactivation. We evaluated postinjection pursuit deficits in step-ramp and sinusoidal tracking.

Pursuit of step-ramp targets after unilateral injections

HORIZONTAL STEP-RAMP TRACKING. Unilateral injection of muscimol into the CFN impaired pursuit so badly in monkey 2 that, in all directions, this animal followed the usual step-ramp target velocity of 20°/s only with saccades. Monkey 2 did have occasional episodes of pursuit to targets moving horizontally at 10°/s. Monkeys 1 and 3 tracked targets moving at 20°/s but with the abnormalities described below. Therefore, our analysis of step-ramp pursuit consists of measurements from monkeys 1 and 3 while they tracked targets moving at 20°/s and from monkey 2 while it tracked targets moving at 10°/s. When tracking targets moving at 10°/s, monkey 2 exhibited the same pattern of deficits that monkey 3 showed when tracking targets at 20°/s, although the deficit was more severe for monkey 2.

Figure 1 shows how CFN inactivation affected step-ramp pursuit in monkeys 1 and 3. After unilateral CFN inactivation, both monkeys often accelerated their eyes beyond the 20°/s target velocity (see arrows) soon after the onset of ipsilateral pursuit, i.e., rightward in Fig. 1B and leftward in Fig. 1C. We saw no evidence that this overshoot became smaller during a series of trials. For example, in a series of 20 ipsilateral tracking movements recorded from monkey 3 after CFN inactivation, the average velocity at the end of acceleration was larger for the last three movements (29°/s) than for the first three (22°/s).

View larger version (28K): [in this window] [in a new window]   FIG. 1. Examples of horizontal step-ramp pursuit in monkeys 1 and 3 after unilateral caudal fastigial nucleus (CFN) inactivation. Each panel shows several superimposed examples of horizontal eye position (HE) and eye velocity (H.E) during rightward (upward deflections) and leftward (downward deflections) pursuit. Dashed lines, middle shows zero eye velocity; top and bottom show velocity of target, 20°/s to right and left, respectively. A: normal pursuit. B: pursuit after inactivation of right CFN. C: pursuit after inactivation of left CFN. Small arrows point to velocity overshoots at end of eye acceleration.

Figure 1 also shows that maintained eye movements in monkey 3 also were slightly slower and/or of shorter duration than normal during contralateral pursuit, i.e., leftward in Fig. 1B and rightward in Fig. 1C. Monkey 1 could maintain nearly normal contralateral eye velocity during step-ramp pursuit but, as we show below, had more difficulty during sinusoidal pursuit.

Figure 2 summarizes the step-ramp tracking data from our three monkeys. Inactivation of one CFN either increased the acceleration of ipsilateral pursuit or decreased the acceleration of contralateral pursuit (Fig. 2A, UNI values). The relative sizes of the increase and decrease were different in the three monkeys. Only the contralateral decrease in monkey 1 and the ipsilateral increase in monkeys 2 and 3 were significant.

View larger version (50K): [in this window] [in a new window]   FIG. 2. Characteristics of horizontal (A and B) and vertical (C and D) step-ramp tracking showing effects of unilateral and bilateral CFN inactivation. Data shown for monkeys 1 and 3 were collected as they pursued a target moving at 20°/s. Data for monkey 2 were collected as that animal pursued a target moving at 10°/s because monkey 2 could not pursue targets moving at 20°/s after CFN inactivation. A and B,  (NORM), combined average normal rightward and leftward pursuit. Striped bars (UNI, IPSI, and CTRA), show average acceleration and maintained velocity gain characteristics of pursuit toward and away from side of inactivated CFN. Stippled bars for monkeys 1 and 3 (BI, IPSI and CTRA), characteristics of pursuit toward and away from side of first injection, but after a second injection. C and D,  (NORM), measurement of normal upward (UP) or downward (DOWN) pursuit. Striped bars (UNI, UP, and DOWN), characteristics of upward and downward pursuit after unilateral CFN inactivation. Stippled bars for monkey 3, characteristics of upward (BI, UP) or downward pursuit (BI, DOWN) after bilateral CFN inactivation. NS, values not significantly different from normal (P > 0.05). Large error bars = SD. Small error bars = SE. Numbers in parentheses = number of horizontal (A and B) and vertical (C and D) movements measured for each monkey. Dashed lines, characteristics of normal pursuit.

In addition to affecting acceleration, CFN inactivation also caused a significant decrease in maintained velocity gain in monkeys 2 and 3, but not in monkey 1 (Fig. 2B, UNI values). After unilateral injection, contralateral maintained velocity gain fell to 92, 27, and 79% of normal in monkeys 1, 2, and 3, respectively. Gain reductions were smaller for ipsilateral pursuit. The maintained velocity gain of ipsilateral pursuit was significantly lower than normal in monkey 3 (89% of normal) but not in monkeys 1 and 2 (99 and 94% of normal, respectively).

Our samples of several step-ramp conditions contain very few measurements, i.e., upward pursuit in monkey 1 (n = 3), ipsilateral and contralateral pursuit in monkey 2 (both ns = 4), and ipsilateral pursuit after bilateral inactivation in monkey 3 (n = 4; Fig. 2). Although we presented about the same number of target movements in these conditions as we did in the others, the monkeys tracked fewer of them. We can only speculate about why a monkey tracked fewer trials in some conditions than in others. However, it is unlikely that the monkeys failed to track many movements in some conditions because they were sick or disoriented (see DISCUSSION). Although some conditions have only three trials, these data are generally consistent with data in the same conditions from the other monkeys or similar conditions in the same monkey. For example, the finding that the maintained velocity gain of contralateral pursuit in monkey 2 fell to ~27% of normal is based on only four measurements (Fig. 2B). However, this result is very similar to the finding, based on 17 measurements, that during sinusoidal pursuit, contralateral pursuit gain in monkey 2 fell to ~23% of normal (Fig. 4B).

View larger version (51K): [in this window] [in a new window]   FIG. 4. Gain during horizontal (A) and vertical (B) sinusoidal smooth pursuit (±10°, 0.5 Hz) after unilateral and bilateral inactivation of the CFN. A: solid bars (NORM), combined averages of normal rightward and leftward pursuit gains. Striped bars (UNI, IPSI, and CONTRA), pursuit gain toward and away from inactivated CFN. Stippled bars, pursuit gain toward (BI, IPSI) or away from (BI, CTRA) side of first injection after second injection. B: solid bars (NORM), gains of normal upward (UP) or downward (DOWN) pursuit. Striped bars, gains of upward (UNI, UP) or downward (UNI, DOWN) pursuit after unilateral injection. Stippled bars, upward (BI, UP) or downward (BI, DOWN) pursuit gain after bilateral CFN inactivation. Large error bars = SD. Small error bars = SE. Numbers in parentheses are number of movements measured for each monkey. Dashed lines, characteristics of normal pursuit.

VERTICAL STEP-RAMP TRACKING. After unilateral CFN inactivation, monkey 2 would not pursue vertical step-ramp targets even at 10°/s so our quantitative results consist of measurements from monkeys 1 and 3. Neither monkey showed a significant change in average eye acceleration at the start of vertical pursuit (Fig. 2C, UNI values). However, inactivating one CFN significantly reduced the gain of both upward and downward maintained pursuit in monkey 1 but reduced only downward pursuit gain in monkey 3 (Fig. 2D, UNI values).

Pursuit of step-ramp targets after bilateral CFN inactivation

As we report above, the first injection decreased contralateral acceleration in monkey 1 and increased ipsilateral acceleration in monkey 3. The second injection (Fig. 2A, BI values) eliminated the effects of the first so that acceleration was not significantly different from normal in either direction for either monkey (ipsilateral: 99 and 102% of normal in monkeys 1 and 3; contralateral: 99 and 107%). For bilateral injections, the terms ipsilateral and contralateral refer to the side of the first injection.

In contrast to its effects on acceleration, bilateral inactivation did not eliminate the reduction in maintained velocity gain caused by unilateral inactivation. On the contrary, bilateral injection made contralateral pursuit gains even lower than after unilateral inactivation (84 vs. 92% of normal in monkey 1; 70 vs. 79% of normal in monkey 3) but these further reductions were not significant (Fig. 2B, BI values). Unlike contralateral gain, ipsilateral gain was nearly the same as after unilateral inactivation (97 vs. 99% of normal in monkey 1, 88 vs. 89% of normal in monkey 3). It is not clear why bilateral inactivation should have different effects on ipsilateral and contralateral gains.

We tested only monkey 3 with vertical step-ramps after bilateral injections. After bilateral inactivation left upward and downward acceleration was nearly normal (Fig. 2C, BI values). The gain of downward, but not upward, pursuit was significantly lower after bilateral inactivation than after unilateral inactivation (downward gain: 65 vs. 82% of normal; upward gain: 86 and 94% of normal after unilateral and bilateral injections, respectively) (Fig. 2D, BI values).

PURSUIT ONSET LATENCY. If CFN activity helps initiate pursuit, inactivation of the CFN should increase pursuit latency. However, CFN inactivation had only small and inconsistent effects on pursuit latency. Only monkey 1 had a significant increase (from 112 to 117 ms) in horizontal pursuit latency after both unilateral and bilateral CFN inactivation; monkeys 2 and 3 did not. (We did not test monkey 2 with bilateral inactivation.)

Pursuit of sinusoidally oscillating targets after unilateral injections

Inactivation of the CFN on one side reduced the gain of both horizontal and vertical sinusoidal tracking movements. Examples of smooth eye movements during pursuit of a target spot oscillating ±10° horizontally at 0.5 Hz appear in Fig. 3. Normally the monkeys followed the target well (gray traces), needing only a few catch-up saccades to the right or the left in about equal numbers. After inactivation of the left CFN, pursuit velocities (black traces) to the right (Fig. 3, upward deflections) did not reach normal peaks. Catch-up saccades occurred much more frequently away from the side of the injection than toward it, regardless of the direction in which the target was moving.

View larger version (25K): [in this window] [in a new window]   FIG. 3. Examples from 3 monkeys of horizontal sinusoidal smooth pursuit eye movements (±10°, 0.5 Hz) before (PRE INJ, gray traces) and after (POST INJ, black traces) inactivation of left CFN. HT, horizontal target position; HE, horizontal eye position; (H.E), horizontal eye velocity.

Figure 4 summarizes the gain of sinusoidal smooth pursuit after CFN inactivation. Unilateral inactivation significantly reduced the gain of contralateral pursuit in all three monkeys to 87, 23, and 88% of normal in monkeys 1, 2, and 3, respectively. There were also small reductions in the gain of ipsilateral pursuit (to 96, 78, and 94% of normal, respectively) that were significant only in monkeys 2 and 3 (Fig. 4A, UNI values).

Unilateral CFN inactivation also affected the gains of vertical pursuit. Downward pursuit gain fell significantly in all three monkeys to 78, 51, and 91% of normal in monkeys 1, 2, and 3, respectively. Upward gains also fell, to 86, 80, and 92% of normal, but the reductions were significant in only monkeys 2 and 3 (Fig. 4B, UNI values).

Pursuit of sinusoidally oscillating targets after bilateral injections

After the gain reduction caused by inactivating one CFN, inactivating the other significantly reduced the gains of contralateral and downward pursuit even further in both monkeys. Contralateral gain fell to 66 and 75% of normal in monkeys 1 and 3, respectively. Downward gain fell to 55 and 72% of normal (Fig. 4, A and B, BI values).

Inactivating the second CFN also reduced ipsilateral gain significantly in monkey 3 (to 81% of normal) but not in monkey 1 (89% of normal). Upward gain fell significantly in monkey 1 (to 56% of normal) but not in monkey 3 (to 89% of normal; Fig. 4, A and B, BI values). Again, ipsilateral and contralateral refer to the side of the first injection.

Pursuit of targets oscillating at different frequencies

Inactivation of one or both CFNs caused only a modest, if any, decrease in pursuit gain as peak target velocity increased from 16 to 50°/s (Fig. 5). For both monkeys tested, normal pursuit gain showed little (monkey 3) or no (monkey 1) decrease with increasing target velocity. After unilateral inactivation, contralateral gain was lower than normal but did not change much with increasing target velocity. In contrast, ipsilateral gain decreased with target velocity for both monkeys. After bilateral inactivation, pursuit gain (average of left and right) was lower than after unilateral inactivation. Gain decreased only slightly with increasing target velocity. (Unfortunately, we did not collect data at 0.25 Hz from monkey 1 after a bilateral injection.)

View larger version (17K): [in this window] [in a new window]   FIG. 5. Smooth pursuit gain at 3 different peak target velocities during tracking of a target oscillating sinusoidally (±10°). NORMAL, averages of leftward and rightward gains before CFN inactivation. BI, averaged leftward and rightward gains after inactivation of both CFNs. IPSI, gains of pursuit toward side of inactivated CFN after unilateral injection. CONTRA, gains of pursuit away from side of inactivated CFN after unilateral injection. Error bars = SD. To avoid overlap of error bars, IPSI points are displaced slightly to right and BI points slightly to left.

Effects on suppression of the VOR

In normal intact monkeys, the eyes move very little when a monkey suppresses its VOR during yaw rotation (Fig. 6, gray records). After inactivation of the right CFN, there was a small but consistent sinusoidal oscillation in horizontal eye position and velocity (Fig. 6, black records) in both monkeys tested (monkeys 1 and 3). These slow eye movements were unsuppressed remnants of the horizontal VOR because, like the VOR, they were in the direction opposite to head rotation.

View larger version (29K): [in this window] [in a new window]   FIG. 6. Examples of eye movements during suppression of vestibuloocular reflex (VOR; ±10°, 0.5 Hz) before (PRE INJ, gray records) and after (POST INJ, black records) inactivation of right CFN. HH, horizontal head position; HE, horizontal eye position; (H.E), horizontal eye velocity.

Inactivation of one CFN significantly increased suppressed VOR gain both ipsilaterally (from 0.08 to 0.30 in monkey 1 and from 0.05 to 0.27 in monkey 3) and contralaterally (from 0.08 to 0.26 monkey 1 and from 0.05 to 0.17 in monkey 3) during yaw oscillations at 0.5 Hz, ±10° (Fig. 7A). In addition, suppressed VOR gain increased for both upward (from 0.23 to 0.57 in monkey 1 and from 0.31 to 0.88 in monkey 3) and downward movements (from 0.13 to 0.48 in monkey 1 and from 0.19 to 0.42 in monkey 3) during oscillations in pitch of 0.5 Hz, ±10° (Fig. 7B).

View larger version (35K): [in this window] [in a new window]   FIG. 7. Horizontal (A) and vertical (B) VOR gain during suppression. A: solid bars, NORM, combined averages of the gain of normal left and right VOR movements. UNI, IPSI (), and CONTRA (), the gain of movements toward and away from injected CFN. B: upward (UP) and downward (DOWN) gains during suppression before (NORM) and after (UNI) muscimol injections into one CFN. Numbers in parentheses are number of movements measured for each monkey. Large error bars are SD. Small error bars are SE. Dashed lines, normal pursuit gain.

Does the CFN compensate for initial starting position in smooth pursuit?

We (Robinson et al. 1993) and others (e.g., Vilis and Hore 1981) have shown that the medial cerebellum helps reduce the effect of orbital position on saccade gain. If CFN activity compensates for starting position during pursuit as well as saccades, bilateral CFN inactivation should cause pursuit acceleration or gain to vary more than normal with starting position. Such an increase in position dependence did not occur in monkey 3, the one monkey tested. Normally step-ramp pursuit acceleration and maintained velocity gain were highest when pursuit started in the center of the orbit and were slightly smaller when pursuit started 10° to the left or right. This pattern changed very little after bilateral inactivation of the CFN. Pursuit gain changed little with starting positions within ±10° either before or after bilateral inactivation.

Control injections

After we injected muscimol into the fastigial nucleus 2.5 mm anterior to the eye-movement-related neurons in the CFN, monkey 1 tracked slowly moving targets accurately and made very few catch-up saccades. This injection did not cause any saccadic abnormalities either (Robinson et al. 1993).

In monkeys 1 and 3, we also injected muscimol into another eye-movement-related part of the cerebellar nuclei, the ventrolateral corner of the posterior interpositus nucleus (VPIN), 4 mm lateral to the CFN. The effect on smooth pursuit was different from that after CFN inactivation. VPIN inactivation produced only small and inconsistent effects on horizontal pursuit in both monkeys. Vertical pursuit, tested only in monkey 3, showed an increase in upward acceleration (to 131% of normal), a decrease in downward acceleration (to 72% of normal), and a small increase in maintained velocity gain in both directions (upward to 110% and downward to 114% of normal).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Pursuit deficits are attributable to CFN inactivation

The muscimol injections in this study affected smooth pursuit by inactivating the CFN. Injecting muscimol into other parts of the cerebellar nuclei, i.e., the rostral fastigial nucleus or the VPIN, did not cause the same abnormalities in pursuit or saccades as those produced by injections into the CFN. Because unilateral and bilateral injections caused significantly different deficits, we infer that little if any of the muscimol injected into one CFN spread to the contralateral CFN ~3 mm away. Furthermore, an injection 2.5 mm anterior to the CFN did not cause any abnormalities in pursuit or saccades, indicating that muscimol from that injection did not spread to the CFN.

Despite their eye movement abnormalities, the monkeys in this study showed no signs of distress or disorientation during our data collection. After CFN inactivation, they did not struggle or vocalize more than normal. Within the limits of their oculomotor deficits, all of the monkeys continued to perform the tracking task for 90-120 min after an injection. They obtained the same amount of reward after CFN inactivation as before. None of our monkeys exhibited nystagmus after a CFN inactivation. Aside from their oculomotor problems then, our monkeys' behavior in the test apparatus was normal. We conclude that the oculomotor deficits we observed are not due to a general disturbance in the condition of our monkeys, such as dizziness or disorientation.

Pursuit deficits after CFN inactivation

Inactivating the CFN on one side either increased the acceleration of ipsilateral pursuit or decreased the acceleration of contralateral pursuit. Subsequent inactivation of the remaining CFN makes horizontal pursuit acceleration nearly normal again. Inactivating one or both CFNs did not affect vertical pursuit acceleration.

Inactivating one CFN caused deficits in pursuit gain as well as acceleration. Contralateral and downward gain fell significantly in all three monkeys during both step-ramp and sinusoidal tracking (except for contralateral step-ramp tracking in monkey 1). The deficits in ipsilateral and upward pursuit gain were smaller and less consistent. Ipsilateral gain did not fall significantly in any monkey during step-ramp tracking and fell significantly during sinusoidal tracking in monkeys 2 and 3. Upward gain fell significantly during step-ramp tracking in monkey 1 but not monkey 3. During sinusoidal tracking, upward gain fell significantly in monkeys 2 and 3 but not monkey 1. We can account for many, but not all, of the pursuit deficits caused by CFN inactivation by assuming that the activity of each pursuit-related CFN neuron helps drive the eyes in the neuron's preferred direction. Below we elaborate on this explanation and suggest other possible mechanisms through which CFN inactivation causes deficits that we cannot account for with our premise. We also briefly consider several questions relevant to our current findings.

CFN's role in horizontal smooth pursuit acceleration

We can explain the effects of CFN inactivation on horizontal pursuit acceleration if pursuit-related CFN activity drives the eyes to the contralateral side and some pursuit-related CFN neurons are active during ipsilateral pursuit acceleration. Some CFN neurons (11/46; see Fig. 3C of Fuchs et al. 1994b) prefer pursuit almost directly contralateral and therefore fire more than cells with other preferred directions during contralateral pursuit. During pursuit in their nonpreferred directions, 4 of 18 neurons tested exhibited a burst of spikes late in eye acceleration at the start of ramp pursuit. Further, most CFN neurons have substantial resting rates and discharge at tens of spikes/second even during pursuit in their nonpreferred direction (Fuchs et al. 1994b).

We propose, for the example of rightward step-ramp pursuit, that neurons in the left CFN start to fire a burst of action potentials before the onset of pursuit. Recording data show that about half of CFN neurons begin to fire before pursuit onset and their average lead is 27 ms (Fuchs et al. 1994b). The increased activity in the left CFN provides a rightward drive that helps accelerate the eyes to the right. This rightward drive from the left CFN is opposed by a leftward drive from the tonically active neurons in the right CFN. This opposition is small compared with the drive from the bursts in the left CFN so that the net effect of the two CFNs is to help accelerate the eyes to the right. Near the end of rightward eye acceleration, some neurons in the right CFN fire a burst of action potentials. This additional leftward drive helps slow the rightward acceleration so that the eyes accurately reach the target's velocity.

Eliminating the activity of one CFN causes an imbalance between the opposing drives of the two CFNs. Ipsilateral acceleration increases and eye velocity overshoots that of the target because both the tonic activity and the late burst that normally oppose ipsilateral acceleration are absent. Contralateral acceleration decreases because the burst of CFN activity that aids contralateral acceleration is gone. Inactivating both CFNs causes no imbalance between the two CFNs, and so horizontal pursuit acceleration is normal. This finding also indicates that the primary drive for pursuit acceleration is generated outside the CFN.

CFN's role in horizontal smooth pursuit gain

If our proposal that CFN activity drives the eyes to the contralateral side during horizontal pursuit is correct, it is easy to see why CFN inactivation reduces the gain of contralateral pursuit. That is, eliminating the contralateral drive provided by the CFN reduces contralateral eye velocity and gain. Recordings of CFN activity support this interpretation by showing that many CFN neurons fire when eye movements reflect large or increasing contralateral drive. So, during constant contralateral velocity in step-ramp pursuit, many CFN neurons fire at elevated rates. During sinusoidal pursuit many CFN neurons reach their peak rates near the peak of contralateral eye velocity (Fuchs et al. 1994b), when, presumably, the pursuit system is providing its peak contralateral drive to the eyes. A few CFN neurons also reach their highest rates soon after peak ipsilateral velocity when the eyes begin to accelerate toward the contralateral side.

Our suggestion does not explain the small reduction we observed in ipsilateral pursuit gain after CFN inactivation. We would expect that, if our proposal is correct, inactivating one CFN would increase, not decrease, ipsilateral pursuit gain. The reduction we observed in ipsilateral pursuit gain might result from one or both of the other consequences of CFN inactivation, i.e., saccade dysmetria or the eye position offset.

Inactivating one CFN makes ipsilateral saccades too large and contralateral saccades too small (Robinson et al. 1993). If a monkey made any catch-up saccades during ipsilateral pursuit, the eyes would overshoot the target. Therefore, a monkey might slow its ipsilateral pursuit after an ipsilateral catch-up saccade to let the target catch up with the fovea. For example, in Fig. 4 monkeys 2 and 3 both exhibit low ipsilateral (leftward) pursuit velocity immediately after their large ipsilateral saccades. Saccade dysmetria cannot account for the reduction in contralateral pursuit gain. Contralateral catch-up saccades after CFN inactivation are too small and so would require an increase in pursuit gain, not the decrease we observed, to make up for the hypometric catch-up saccades.

In addition to causing saccade dysmetria, unilateral CFN inactivation also makes a monkey direct its fovea ~2° ipsilaterally from the target. A monkey with an inactivated CFN does not try to correct this offset (Robinson et al. 1993). Previous work tells us that, in intact monkeys, a position offset between the target and eye elicits pursuit acceleration toward the target (Lisberger and Westbrook 1985; Morris and Lisberger 1987). However, the intact monkeys in those studies aim their foveae directly at a target when not thwarted by an imposed offset. After CFN inactivation, the situation is different. The offset does not elicit any pursuit acceleration and, in fact, a monkey will move its eyes away from the target if it is moved so that its image falls on the monkey's fovea. Therefore, even during pursuit, the target's image is always ~2° peripheral to the center of the fovea. There is currently no information on how an offset maintained by the monkey affects pursuit performance. It is at least plausible that the gain of pursuit maintained by an image that falls consistently outside the fovea is lower than the gain of normal pursuit when the target image is directly on the fovea.

It is possible that the eye position offset caused by CFN inactivation also reduces contralateral pursuit gain. Therefore, our proposal that pursuit-related CFN activity drives the eyes toward the contralateral side during contralateral pursuit is not the only possible explanation for reduced contralateral gain after CFN inactivation. However, we favor reduced contralateral drive as the explanation for the deficit in contralateral pursuit gain because it is consistent with the responses of CFN neurons. Still, our current data cannot eliminate the possibility that CFN inactivation reduces contralateral pursuit gain, at least in part, because the target is not on the fovea.

CFN's role in vertical smooth pursuit

Some pursuit-related CFN neurons prefer pursuit approximately downward (11/46; see Fig. 3C of Fuchs et al. 1994b) or upward (6/46). If we assume that each of these neurons drives the eyes in its preferred direction, we would expect that silencing these cells would reduce vertical pursuit gain, as we observed. The fact that CFN inactivation reduces downward gain more than upward gain may reflect our earlier finding that CFN neurons that prefer downward pursuit are more numerous than neurons that prefer upward pursuit.

If, as suggested by the reduction in vertical pursuit gains after CFN inactivation, some CFN activity helps maintain vertical pursuit, why does CFN inactivation not affect the acceleration of vertical pursuit? This is especially puzzling because some CFN neurons fire a burst of action potentials beginning before the start of step-ramp pursuit with significant vertical components (e.g., see Fuchs et al. 1994b, Fig. 7C). Available data do not tell us why CFN inactivation reduces the gain but not acceleration of vertical pursuit.

Why are the deficits of monkey 2 worse than those of monkeys 1 and 3?

CFN inactivation impaired monkey 2's pursuit much more than it impaired the pursuit of the other two monkeys, though the pattern of deficits was similar in all three. The severe deficits in monkey 2 did not occur because of insufficient training; monkey 2 was trained as long as the other animals and, when not injected, performed smooth pursuit nearly as well as monkeys 1 and 3. Therefore, monkey 2's larger deficits suggests that the CFN contributes more to smooth pursuit in some monkeys than in others.

Monkey 2's normal pursuit was not quite as good as that of monkeys 1 and 3. It accelerated more slowly and had slightly lower gain than did the pursuit of monkeys 1 and 3. These data may mean that monkeys with poor pursuit rely more on CFN activity to aid their pursuit. Data from more monkeys will be required to test this suggestion.

Horizontal pursuit gain at different target velocities

After either unilateral or bilateral CFN inactivation, horizontal pursuit gain decreased only slightly with increasing target velocity (Fig. 5). Evidently, within the range we tested CFN activity aids pursuit of different target frequencies about equally. This is surprising because CFN neurons modulate their activity more strongly for higher than lower target velocities (Fuchs et al. 1994).

It is also surprising that, when monkeys pursued an oscillating target with a peak velocity of ~50°/s, ipsilateral pursuit was impaired at least as much as contralateral pursuit. Accounting for this ipsilateral gain decrease at high target velocities will have to wait until we understand why ipsilateral pursuit gain is impaired after CFN inactivation.

VOR suppression

After unilateral injections of muscimol into the CFN, the monkeys exhibited deficits in suppressing their VORs. These data are generally similar to those of Kurzan et al. (1993), who produced suppression by coupling the movement of a surrounding scene to a monkey's oscillations. It has been suggested that smooth pursuit participates in VOR suppression because pursuit and VOR suppression develop together in humans (Herman et al. 1982) and decline together in both humans and monkeys as stimulus movement increases in frequency (Barnes et al. 1978; Lisberger et al. 1981). We interpret the deficits in VOR suppression after CFN inactivation as a consequence of the deficits in smooth pursuit.

Wallenberg Syndrome

Saccadic abnormalities after inactivation of the CFN are the same as those in patients with a lateral medullary stroke that causes a collection of symptoms called Wallenberg Syndrome (Robinson et al. 1993). We suggested that the stroke causes saccadic symptoms via a mechanism originally proposed by Waespe and Wichmann (1990), i.e., the stroke inhibits CFN activity by interrupting the axons of inferior olive neurons bound for the cerebellum. Our current data indicate that inhibition of the CFN also can account for the pursuit deficits of Wallenberg patients that, like monkeys with an inhibited CFN, exhibit reduced pursuit gain contralateral to their lesion (Leigh and Zee 1991).

Are the CFN and the flocculus-ventral paraflocculus the only cerebellar pursuit areas?

Our data help us understand how the cerebellum mediates smooth pursuit by identifying the CFN as a region in addition to the flocculus and ventral paraflocculus (F-VPF) that participates in both pursuit acceleration and maintenance. Do other cerebellar areas also contribute to smooth pursuit? Smooth pursuit cannot be abolished by compromising either the F-VPF or the CFN alone but can be with a complete cerebellectomy. From this we can infer either that a part of the cerebellum other than the F-VPF and CFN is involved in pursuit or that the F-VPF and CFN each can compensate partially for the loss of the other. One could examine these alternatives by testing pursuit in a monkey with bilateral lesions of both the F-VPF and CFN.

	ACKNOWLEDGEMENTS

  We thank S. Usher and D. Reiner for expert technical assistance and K. Elias for top-notch editing. We also thank C. Kaneko, L. Ling, J. Phillips, M. Pong, and R. Soetedjo for useful comments on an earlier version of this manuscript.

  This study was supported by National Institutes of Health Grants EY-10578, EY-00745, and RR-00166 and by the Deutsche Forschungsgemeinschaft (Heisenberg-Program).

	FOOTNOTES

  Address for reprint requests: F. R. Robinson, Dept. of Physiology and Biophysics, Box 357290, University of Washington, Seattle, WA 98195-7290. E-mail: robinsn{at}u.washington.edu

  Received 28 October 1996; accepted in final form 9 April 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• BARNES, G. R., BENSON, A. J., PRIOR, A.R.J. Visual-vestibular interaction in the control of eye movement. Aviat. Space Environ. Med. 49: 557-564, 1978.[Medline]
• BELKNAP, D. B., NODA, H. Eye movements evoked by microstimulation in the flocculus of the alert macaque. Exp. Brain Res. 67: 352-362, 1987.[Medline]
• BRODAL, P. The pontocerebellar projection in the rhesus monkey: an experimental study with retrograde axonal transport of horseradish peroxidase. Neuroscience 4: 193-208, 1979.[Medline]
• BURDE, R. M., STROUD, H. M., ROPER-HALL, G., WIRTH, F. P., O'LEARY, J. L. Ocular motor dysfunction in total and hemicerebellectomized monkeys. Br. J. Ophthalmol. 59: 560-565, 1975.[Abstract/Free Full Text]
• FUCHS, A. F., ROBINSON, F. R., AND STRAUBE, A. Preliminary observations on the role of the caudal fastigial nucleus in the generation of smooth pursuit eye movements. In: Contemporary Ocular Motor and Vestibular Research: A Tribute to David A. Robinson, edited by A. F. Fuchs, Th. Brandt, U. Büttner, and D. S. Zee. Stuttgart: Thieme, 1994a, p. 165-170.
• FUCHS, A. F., ROBINSON, F. R., STRAUBE, A. Participation of the caudal fastigial nucleus in smooth pursuit eye movements. I. Neuronal activity. J. Neurophysiol. 72: 2714-2728, 1994b.[Abstract/Free Full Text]
• GERRITS, N. M., VOOGD, J. The topological organization of climbing and mossy fiber afferents in the flocculus and ventral paraflocculus in rabbit, cat, and monkey. In: The Olivocerebellar System in Motor Control (Exp. Brain Res. Series 17), edited by and P. Strata . Berlin: Springer-Verlag, 1989, p. 26-29
• GLICKSTEIN, M., GERRITS, N., KRALJ-HANS, I., MERCIER, B., STEIN, J., VOOGD, J. Visual pontocerebellar projections in the macaque. J. Comp. Neurol. 349: 51-72, 1994.[Medline]
• HERMAN, R., MAULUCCI, R., STUYCK, J. Development and plasticity of visual and vestibular generated eye movements. Exp. Brain Res. 47: 69-78, 1982.[Medline]
• KRAUZLIS, R. J., MILES, F. A. Effects of electrical stimulation of the cerebellar posterior vermis on smooth pursuit eye movements in monkeys. Soc. Neurosci. Abstr. 20: 858, 1994.
• KURZAN, R., STRAUBE, A., BÜTTNER, U. The effect of muscimol micro-injections into the fastigial nucleus on the optokinetic response and the vestibulo-ocular reflex in the alert monkey. Exp. Brain Res. 94: 252-260, 1993.[Medline]
• LEIGH, R. J., ZEE, D. In: The Neurology of Eye Movements, , 2nd ed.. Philadelphia, PA: Davis, 1991
• LISBERGER, S. G., EVINGER, C., JOHANSON, G. W., FUCHS, A. F. Relationship between eye acceleration and retinal image velocity during foveal smooth pursuit in man and monkey. J. Neurophysiol. 46: 229-249, 1981.[Free Full Text]
• LISBERGER, S. G., FUCHS, A. F. Role of primate flocculus during rapid behavioral modification of vestibulo-ocular reflex. I. Purkinje cell activation during visually guided horizontal smooth-pursuit eye movements and passive head rotation. J. Neurophysiol. 41: 733-763, 1978.[Abstract/Free Full Text]
• LISBERGER, S. G., PAVELKO, T. A. Brain stem neurons in modified pathways for motor learning in the primate vestibulo-ocular reflex. Science Wash. DC 242: 771-773, 1988.[Abstract/Free Full Text]
• LISBERGER, S. G., WESTBROOK, L. E. Properties of visual inputs that initiate horizontal smooth pursuit eye movements in monkeys. J. Neurosci. 5: 1662-1673, 1985.[Abstract]
• MAY, J. G., KELLER, E. L., CRANDALL, W. F. Changes in eye velocity during smooth pursuit tracking induced by microstimulation in the dorsolateral pontine nucleus of the macaque. Soc. Neurosci. Abstr. 11: 79, 1985.
• MAY, J. G., KELLER, E. L., SUZUKI, D. A. Smooth-pursuit eye movement deficits with chemical lesions in the dorsolateral pontine nucleus of the monkey. J. Neurophysiol. 59: 952-977, 1988.[Abstract/Free Full Text]
• MILES, F. A., FULLER, J. H., BRAITMAN, D. J., DOW, B. M. Long-term adaptive changes in primate vestibulo-ocular reflex. III. Electrophysiological observations in flocculus of alert monkey. J. Neurophysiol. 43: 1437-1476, 1980.[Free Full Text]
• MORRIS, E. J., LISBERGER, S. G. Different responses to small visual errors during the initiation and maintenance of smooth-pursuit eye movements in monkeys. J. Neurophysiol. 58: 1351-1369, 1987.[Abstract/Free Full Text]
• MUSTARI, M. J., FUCHS, A. F., WALLMAN, J. Response properties of dorsolateral pontine units during smooth pursuit in rhesus macaque. J. Neurophysiol. 60: 664-686, 1988.[Abstract/Free Full Text]
• NODA, H., SUGITA, S., IKEDA, Y. Afferent and efferent connections of the oculomotor region of the fastigial nucleus in the macaque monkey. J. Comp. Neurol. 302: 330-348, 1990.[Medline]
• RASHBASS, C. The relationship between saccadic and smooth tracking eye movements. J. Physiol. Lond. 159: 326-338, 1961.
• ROBINSON, D. A. A method of measuring eye movement using a scleral search coil in a magnetic field. IEEE Trans. Biomed. Electron. 10: 137-145, 1963.
• ROBINSON, F. R., FUCHS, A. F., STRAUBE, A. Role of the caudal fastigial nucleus in saccade generation. II. Effects of muscimol inactivation. J. Neurophysiol. 70: 1741-1758, 1993.[Abstract/Free Full Text]
• ROBINSON, F. R., STRAUBE, A., FUCHS, A. F. Smooth pursuit deficits after muscimol inactivation of the caudal fastigial nucleus in macaque. Soc. Neurosci. Abstr. 20: 856, 1994.
• SHIDARA, M., KAWANO, K. Role of Purkinje cells in ventral paraflocculus in short-latency ocular following responses. Exp. Brain Res. 93: 185-195, 1993.[Medline]
• STONE, L., LISBERGER, S. G. Visual responses of Purkinje cells in the cerebellar flocculus during smooth pursuit eye movements in monkeys. I. Simple spikes. J. Neurophysiol. 63: 1241-1261, 1990.[Abstract/Free Full Text]
• SUZUKI, D. A., KELLER, E. L. Visual signals in the dorsolateral pontine nucleus of the alert monkey: their relationship to smooth pursuit eye movements. Exp. Brain Res. 53: 473-478, 1984.[Medline]
• SUZUKI, D. A., KELLER, E. L. The role of the posterior vermis of the cerebellum in smooth pursuit eye movements. I. Eye and head movement-related activity. J. Neurophysiol. 59: 1-18, 1988a.[Abstract/Free Full Text]
• SUZUKI, D. A., KELLER, E. L. The role of the posterior vermis of the cerebellum in smooth pursuit eye movements. II. Target velocity-related Purkinje cell activity. J. Neurophysiol. 59: 19-40, 1988b.[Abstract/Free Full Text]
• SUZUKI, D. A., NODA, H., KASE, M. Visual and pursuit eye movements-related activity in the posterior vermis of the monkey cerebellum. J. Neurophysiol. 46: 1120-1139, 1981.[Free Full Text]
• THIELERT, C.-D., THIER, P. Patterns of projections from pontine nuclei and the nucleus reticularis tegmenti pontis to the posterior vermis in the rhesus monkey: a study using retrograde tracers. J. Neurophysiol. 337: 113-126, 1993.
• THIER, P., KOEHLER, W., BUETTNER, U. W. Neuronal activity in the dorsolateral pontine nucleus of the alert monkey modulated by visual stimuli and eye movements. Exp. Brain Res. 70: 496-512, 1988.[Medline]
• VAN KAN, P.L.E., HOUK, J. C., GIBSON, A. R. Output organization of intermediate cerebellum of the monkey. J. Neurophysiol. 69: 57-73, 1993.[Abstract/Free Full Text]
• VILIS, T., HORE, J. Characteristics of saccade dysmetria in monkeys during reversible lesions of medial cerebellar nuclei. J. Neurophysiol. 46: 828-838, 1981.[Abstract/Free Full Text]
• WAESPE, W., WICHMANN, W. Oculomotor disturbances during visual-vestibular interaction in Wallenberg's lateral medullary syndrome. Brain 113: 821-846, 1990.[Abstract/Free Full Text]
• WESTHEIMER, G., BLAIR, S. M. Oculomotor deficits in cerebellectomized monkeys. Invest. Ophthalmol. 12: 618-621, 1973.[Free Full Text]
• YAMADA, J., NODA, H. Afferent and efferent connections of the oculomotor vermis of the macaque monkey. J. Comp. Neurol. 265: 224-241, 1987.[Medline]
• ZEE, D. S., YAMAZAKI, A., BUTLER, P. H., GUCER, G. Effects of ablation of flocculus and paraflocculus on eye movements in primate. J. Neurophysiol. 46: 878-899, 1981.[Free Full Text]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society