Adaptive Changes in the Angular VOR: Duration of Gain Changes and Lack of Effect of Nodulo-Uvulectomy

Abstract

Alterations in the gain of the vertical angular vestibulo-ocular reflex (VOR) are dependent on the head position in which the gain changes were produced. We determined how long gravity-dependent gain changes last in monkeys after four hours of adaptation, and whether the adaptation is mediated through the nodulus and uvula of the vestibulocerebellum. Vertical VOR gains were adaptively modified by rotation about an interaural axis, in phase or out of phase with the visual surround. Vertical VOR gains were modified with the animals in one of three orientations: upright, left-side down, or right-side down. Monkeys were tested in darkness for up to four days after adaptation using sinusoidal rotation about an interaural axis that was incrementally tilted in 10° steps from vertical to side down positions. Animals were unrestrained in their cages in normal light conditions between tests. Gravity-dependent gain changes lasted for a day or less after adaptation while upright, but persisted for two days or more after on-side adaptation. These data show that gravity-dependent gain changes can last for prolonged periods after only four hours of adaptation in monkeys, as in humans. They also demonstrate that natural head movements made while upright do not provide an adequate stimulus for rapid recovery of vertical VOR gains that were induced on side. In two animals, the nodulus and uvula were surgically ablated. Vertical gravity-dependent gain changes were not significantly different before and after surgery, indicating that the nodulus and uvula do not have a critical role in producing them.

INTRODUCTION

The angular vestibulo-ocular reflex (VOR) causes the eyes to counter-rotate during head movement to stabilize vision. If ocular counterrotation is not adequate, slip of retinal images provides a stimulus that, over time, will change or adapt the gain of the VOR so that eye velocity more closely equals head velocity (see Ref. 3 for review). In both monkeys and humans, the first significant gain changes occur as early as 20–40 min after onset of the conditioning procedure,^4–7 and two hours of adaptation will produce gain changes in the monkey of more then 20–25%.^6–8 If adaptation is continued for an additional two hours, there is only a slight additional gain change (about 5%). At that point, the gain stabilizes and is unchanged even if stimulation is prolonged for up to 8 h.^8–13

In several studies, it has been demonstrated that the VOR gain can be modified in relation to particular contexts such as eye position^14,15 and orientation of the head with regard to gravity.^16–19 If vertical VOR gain changes are adapted in a particular head orientation, the gain changes are maximal when animals or humans are tested in the head orientation in which the VOR gain was adapted, and the changes gradually decrease as the head is oriented away from this position.^1,2 In humans, the gain changes can persist for two to three days after on-side adaptation.² We questioned whether the persistence of the gain changes was due to limited head movements in side-down position in normal daily behavior. If so, then the gain changes induced while upright should not last as long as gain changes induced on side. A determination of the comparative persistence of the vertical VOR gain changes after upright or on-side adaptation was the first objective of this study.

The finding that the gravity-dependent changes occur almost immediately after assuming the head position in which the gains were adapted, and that the extent of gain change decreases as the head is positioned away from the adapted orientation, suggests that the otoliths, which code information about head position provide an important input to central canal units to determine angular VOR gains. Whether other brain structures also convey head orientation information to central VOR-related units is unknown. The nodulus and uvula of the vestibulocerebellum are involved in orienting the axis of eye velocity generated through the VOR toward the spatial vertical.^20–25 A second goal of this study was to determine whether the nodulus and uvula are also involved in processing head orientation signals to implement the gravity-dependent gain changes.

METHODS

Two cynomolgus (Macaca fascicularis, M9358 and M0102) and one rhesus (Macaca mulatta, M98065) monkeys were used in this study. The experiments conformed to the Guide for the Care and Use of Laboratory Animals²⁶ and were approved by the Institutional Animal Care and Use Committee. Experimental techniques have been described in detail in previous publications, and only essential details will be provided here. Under anesthesia, a head mount was implanted on the skull to provide painless head fixation in stereotaxic coordinates.^8,27 Two scleral search coils were implanted on the left eye. One measured the horizontal and vertical components of eye position.²⁸ Another coil, placed approximately orthogonal to the frontal coil, was used to measure the torsional component of eye position. Nodulouvulectomy was performed in two animals after pretesting.²⁰ The extent of lesions was determined in histological sections.

During testing, the monkey’s head was fixed to a plastic frame, which held two sets of field coils that generated orthogonal oscillating magnetic fields at the same frequency. The primate chair was centered in a four-axis vestibular stimulator surrounded by an optokinetic drum with black and white vertical stripes. To calibrate eye movements, the animals were rotated in light at 30°/s about a spatial vertical axis. Animals were upright for calibration of yaw eye movements, left-side down (LSD) for pitch movements, and prone for roll movements. It was assumed that horizontal and vertical gains were close to unity when upright or side down,^28,29 and torsional gains were assumed to be 0.6 when the rotation was performed around a naso-occipital axis aligned with the spatial vertical.^30,31 Positive directions of eye movement were leftward for yaw, downward for pitch, and clockwise (from the animal’s point of view) for roll components.

Vertical VOR gains were adapted by oscillating the monkeys in light for four hours about an interaural axis in one of three positions: upright, LSD, or right-side down (RSD). Gains were decreased by rotating the animal and visual surround in the same direction and increased by rotating the animal and visual surround in opposite directions. Sinusoids of 0.25 Hz and 0.5 Hz as well as steps of velocity¹ were used for gain adaptation.

To measure gravity-dependent effects on vertical VOR gain adaptation [(eye velocity)/(head velocity)], animals were oscillated sinusoidally at 0.5 Hz (60°/s peak velocity) in darkness about a pitch (interaural) axis that was either upright or tilted toward side-down positions in roll in 10° increments up to 90°. Ten cycles of data were obtained in each head orientation. Because the animals were always rotating in pitch, canal activation was the same in every head orientation. Eye position voltages and voltages related to the velocity of the chair oscillation and the tilt of the axis of rotation were recorded with amplifiers having a bandpass of DC to 40 Hz. Data were acquired by computer and analyzed off-line. Voltages were digitized at 1000 Hz/channel with 16-bit resolution. Voltages related to eye position were digitally differentiated by finding the slope of the least squares linear fit over 25-ms time period. This corresponds to a filter, which has a 3-dB cutoff above 40 Hz, the cutoff frequency of the filters used for data acquisition. Saccades were eliminated using a maximum likelihood ratio criterion.³²

The gain of the VOR was tested sequentially for 2–4 days after adaptation. Desaccaded eye velocity was fitted with sinusoids to estimate the gain and phase of the response in each head orientation (temporal gain and phase). Changes in gain were expressed as a percentage relative to the preadapted level for each head orientation and plotted as a function of head tilt. To analyze the magnitude of the gravity-dependent VOR gain adaptation, we applied a sine approximation with a bias (C) to the data

$$y=A*\text{cos}\left(x+B\right)+C$$ (1)

where A is a half of the magnitude of the gravity specific effect and B is the angle of the head tilt where the maximal gain changes occurred, i.e., the spatial phase. C is the bias of the sinusoid, which gives a measure of the gravity-independent gain changes.

Standard t tests and ANOVA were used to analyze pairs or sets of data, respectively. A generally accepted statistical approach for data-model comparison, the c² test, provided a robust statistical analysis if there were several hundred data points.³³ This assumption failed if the sample size was small. An analysis of variance (ANOVA) is less sensitive to any non-normality in the data distribution.³⁴ To avoid possible complications in the statistical analysis, we utilized a reduced case of the analysis of variance (F-statistic).³⁵

RESULTS

Gravity-Dependent and Gravity-Independent Gain Changes of Normal Animals

When animals were sinusoidally oscillated about an interaural axis in darkness before adaptation, the induced vertical eye velocities were about the same in all tested head orientations (Fig. 1A). After the vertical VOR gain was decreased with the animal in the LSD position, modulation of the eye velocities induced by interaural rotation was reduced maximally when the animal was LSD in darkness (Fig. 1B, top trace), and peak eye velocity increased toward the original value as the animal was tilted toward the right-side down (RSD) position (Fig. 1B, middle and low traces). As previously shown,¹ there was an up-down asymmetry in the velocity decreases, and the gain changes were larger for downward (+) than upward (−) slow-phase velocities (Fig. 1B). The induced eye velocity and retinal slip were minimal, and visual suppression was complete when head was pitched up (Fig. 1C, positive eye velocity), but not when the head was pitched down (Fig. 1C, negative velocities). This asymmetry is likely related to a difference in visual suppression, which could result in a difference in upward and downward gain adaptation.⁸ Since we were concerned with the overall vertical VOR performance in this study, these differences will not be considered further.

FIGURE 1.

(A, B) Eye velocities induced by sinusoidal oscillation around an interaural axis with the animal left-side down (top trace), upright (middle trace) and right-side down (bottom trace), before (A) and after (B) the VOR gain was decreased while LSD. The heavy lines are slow-phase velocity, and saccadic velocities, which were removed for the analysis, are shown by the thin vertical dotted lines. The thin solid lines show head velocity, inverted to facilitate comparison. (C) Suppression of eye velocities induced by sinusoidal rotation in a subject-stationary visual surround. Downward (+) eye velocities were suppressed more completely than upward eye velocities (−). (D) Vertical VOR gains were not dependent on head orientation before adaptation (dashed line; dotted lines around the dashed line represent ± 1 SD). After the gain was decreased in the LSD position (solid symbols), gain values varied as a function of head tilt, and were smallest when the head was close to the position of adaptation (−90°). (E) Gain changes as a function of head tilt re gravity from the data in D are shown by the solid symbols. The heavy line shows a sinusoidal fit to these data. The dashed line is the bias of the sinusoidal fit, which is the gravity-independent gain change. The peak-to-peak amplitude of the sinusoid (dotted lines) represents the magnitude of the gravity-dependent gain changes.

The gain of the vertical VOR was calculated for each head orientation in the unadapted state and plotted as a function of head tilt to obtain the spatial responses. Before adaptation, vertical VOR gains were independent of the angle of head tilt (Fig. 1D, dashed line, ± 1 SD, dotted lines), and the average gains were approximately the same over all head orientations. After the pitch VOR gain was decreased for four hours with animal LSD; the measured gain was minimal when the animal was tested LSD; and gain values progressively increased as the monkey was reoriented toward the contralateral side down (Fig. 1D, filled symbols).

Two types of VOR gain changes are induced when the VOR is adapted: one is dependent on the head orientation re position in which gain was changed, while the other is independent of head position.¹ When the gain changes were fit with a sine function (Eq. 1; FIG. 1E), the bias of the sinusoid represented the gravity-independent gain change, and the peak-to-peak amplitude, the gravity-dependent changes. Gravity-independent gain changes in FIGURE 1E were −21%, and the amplitude of the fitted sine was »14%. Therefore, the peak-to-peak gravity-dependent gain changes were 28%, that is, twice the amplitude.

Prolonged Effects of Gravity-Dependent VOR Gain Changes

In two animals, the pitch component of the VOR gain was adaptively increased and decreased in each of the three positions, and the animals were tested sequentially for up to four days after adaptation. Between tests, animals were caged, but were otherwise unrestricted in their daily activity. After the VOR gain was increased in animal M98065 while in LSD position, the gravity-dependent gain changes were 42% when tested on the same day (Fig. 2A). One and two days after adaptation, the gravity-dependent gain changes were 20% (Fig. 2E) and 15% (Fig. 2I), respectively. Gravity-independent gain changes were 20% on the day of adaptation, but only 8% and 4% on the first and second days. When the VOR gain was decreased in the RSD head position in the same animal, the gravity-dependent gain changes were −57% on the day of adaptation (Fig. 2B), and decreased to 28% (Fig. 2F) and 22% (Fig. 2J), when tested one and two days afterward. The gravity-independent gain changes were −31% on the day of adaptation, and −17% and −13% on the two following days. Thus, four hours of on-side adaptation produced gravity-dependent gain changes in the monkeys that were present for at least two days after adaptation (F-statistic, P <0.05).

FIGURE 2.

Gain changes of the vertical VOR in two animals after adaptation in on-side (1^st and 2^nd columns) and upright (3^rd and 4^th columns) positions. Data are show for the day of adaptation (A–D), one day after adaptation (E–H) and two days after adaptation (I–L). The dashed line represents the bias of the sinusoidal fits to the gain changes in the different positions. The inserts below show approximate head positions and axes of rotation during testing and can be related to the data in the graphs above.

When the vertical VOR gain was adaptively increased with the animal M0102 in the upright position, the peak-to-peak gravity-dependent changes were 33% on the day of adaptation (Fig. 2C), but rapidly decreased to 23% (Fig. 2G) and 9% (Fig. 2K) in the following two days. These gain changes were only significant on the day of adaptation and one day afterward (F-statistic, P <0.05). Gravity-independent changes were » 12% on the day of adaptation and on both subsequent days. Gain decreases with the animal adapted in the upright position were smaller, although significant (P <0.05), than when the gains were modified on-side (Fig. 2D). In one experiment, the gravity-dependent gain changes were 12% and gravity-independent changes were −14% in the day of adaptation. Changes observed in the two days following adaptation were not significant (Fig. 2H, L).

The gravity-dependent gain changes at different times after adaptation are summarized in FIGURE 3. The gravity-dependent gain increases and decreases were significant in all instances just after adaptation varying from 15–55% and were approximately the same for gain increases and decreases for all positions (Fig. 3A). When adaptation was done in the on-side position, gain changes were significant within the first two days after adaptation or longer (Fig. 3A, filled symbols). After upright adaptation, however, significant gain changes were observed only in one case and only on the first day after adaptation (Fig. 3B).

FIGURE 3.

Average changes of the vertical VOR gain (%) in two animals at different times after adaptation on side (A) and upright (B). The filled symbols represent significant values.

Gravity-Dependent VOR Gain Adaptation after Nodulo-Uvulectomy

Two animals were tested for their ability to generate gravity-dependent VOR gain adaptation before and after surgical ablation of the nodulus and uvula. One animal was M98065, whose data are described in the previous section. The second animal was M9358, which had been used in single-unit experiments from the nodulus and uvula that are described elsewhere.²¹ The previous experiments in M9358 were completed about one year before the control data for the present experiments were taken.

When M98065 was rotated in darkness with the steps of velocity (60°/s), the velocity storage time constant had symmetrical responses for leftward and rightward rotations before and after ablation, but the time constants were reduced from »40 s before to »20 s one month after surgery. Since isolated removal of the nodulus and uvula prolongs the horizontal time constant of velocity storage and causes a loss of habituation (see Ref. 36 for review), this suggested that the lesion had extended beyond the nodulus and uvula in this animal. On histological analysis, the vermis of lobules VIIa– X were ablated, but the lesion extended into the fastigial nucleus and the caudal interpositus nucleus on the right. There was also a residual tag of nodulus tissue just above the choroids plexus, but there were no Purkinje cells in the remaining tissue. Since the nodulus and uvula were almost completely removed in this animal, we considered that it could be used to determine whether gravity-dependent adaptation could be induced after nodulo-uvulectomy.

When the vertical VOR gain was increased in the LSD position before ablation (Fig. 4A), the peak-to-peak gravity-dependent and gravity-independent gain changes in animal M98065 were 38% and 16%, respectively. Two months after the surgery, there were both gravity-dependent and -independent gain changes of 39% and 21%, respectively, in response to the same stimulus (Fig. 4E). The head orientations in which peak gravity-dependent changes occurred were also the same before and after (−127° and −128°). Similar results were obtained when gain was increased in RSD (Fig. 4B, F), or decreased in LSD (Fig. 4C, G), or RSD positions (Fig. 4D, H). The averages for all conditions were 42 ± 10% for the gravity-dependent changes before and 36 ± 10% after surgery, and the gravity-independent changes were 22 ± 10% and 25 ± 4% before and after surgery (Table 1).

FIGURE 4.

Gain increases (1^st and 2^nd columns) and decreases (3^rd and 4^th columns) in M98065. The gain changes were similar before (A–D) and two months after (E–H) nodulo-uvulectomy

TABLE 1.

Parameters of gain changes obtained from M98065 before and two months after the nodulo-uvulectomy

Gain changes	Head tilt	Gain parameters	Before surgery	Two months after surgery
Increase	LSD	Gravity-dependent gain	38%	39%
		Phase	−127°	−128°
		Gravity-independent gain	16%	21%
	RSD	Gravity-dependent gain	32%	34%
		Phase	122°	123°
		Gravity-independent gain	10%	22%
Decrease	LSD	Gravity-dependent gain	42%	23%
		Phase	−128°	−108°
		Gravity-independent gain	−30%	−25%
	RSD	Gravity-dependent gain	56%	46%
		Phase	110°	124°
		Gravity-independent gain	−31%	−30%

Gain changes were all significant (P <0.05) and were not substantially different before and after operation.

The time constant of the second animal (M9358), which was initially about 30–40 s before operation, was reduced to 8–9 s as a result of repeated testing. One week after nodulo-uvulectomy, central control of the VOR time constant was lost, and the evoked per-rotatory nystagmus did not diminish even after 5 min of constant velocity rotation in darkness. Two months after surgery, the time constants of the per- and post-rotatory nystagmus were »500 and »300 s for rotations to the left and right, respectively. Periodic alternating nystagmus in darkness was also present in this animal. Histological analysis demonstrated that the nodulus and uvula were completely ablated. Thus, the behavioral data obtained after ablation correlated with the histological analysis in this animal²⁰ and with previous results.^22,37–41

The gravity-dependent gain changes also persisted after nodulo-uvulectomy in this animal (Fig. 5A–D). Before operation, they were about 20 ± 6% (Table 2). In the first month after surgery, significant gravity-dependent gain changes were only present for gain increase while RSD (Table 2). Three months after surgery, however, there were significant gravity-dependent gain changes in three of the four conditions (Fig. 5F, G, H). In two of these conditions (Fig. 5F, G), the gravity-dependent gain increases in RSD and decreases in LSD changes were comparable to prelesion data (22% vs. 18% and 14% vs. 14%, respectively; Table 2). Thus, data obtained from both animals demonstrate that complete removal of the nodulus and uvula did not abolish the gravity-dependent gain changes.

FIGURE 5.

Gain increases (1^st and 2^nd columns) and decreases (3rd and 4th columns) in M9358. The gain changes were significant before surgery (A–D). Three months after surgery (E–H), the gains could not be increased in the LSD position (E), but the changes were close to the presurgical values in all other conditions (F–H).

TABLE 2.

Parameters of gain changes obtained from M9358 before and one and two months after the nodulo-uvulectomy

Gain changes	Head tilt	Gain parameters	Before surgery	One month after surgery	Three months after surgery
Increase	LSD	Gravity-dependent gain	28%	5%*	4%*
		Phase	−74°	−136°	−2°
		Gravity-independent gain	4%	5%*	7%*
	RSD	Gravity-dependent gain	22%	15%	18%
		Phase	97°	65°	81°
		Gravity-independent gain	17%	3%	9%
Decrease	LSD	Gravity-dependent gain	15%	8%*	14%
		Phase	−115°	−135°	−143°
		Gravity-independent gain	−11%	−9%*	−15%
	RSD	Gravity-dependent gain	18%	8%*	8%
		Phase	80°	69°*	51°
		Gravity-independent gain	−12%	−6%*	−9%

^*Indicates non-significant changes in gain and/or phase.

DISCUSSION

This study confirms previous findings that gain adaptation of the vertical VOR is dependent on head orientation re gravity, that modifications in gain are largest in the head orientation in which vertical gain was adapted, and that the changes in gain gradually decrease as animals are reoriented away from this position.^1,2 Additionally, we show here that the gravity-dependent gain changes that were induced by only four hours of adaptation persisted for several days in the monkey, similar to the persistence of the changes in vertical VOR gain in humans after one hour of adaptation.² Horizontal VOR gains will normalize within an hour in humans⁴² and animals⁴³ if the subjects are allowed to move their heads freely in light. However, adaptive changes in the horizontal VOR were maintained for over a week if the monkeys’ heads were immobilized.¹³ This implies that adaptive gain changes of the VOR will decay at a very slow rate spontaneously unless these changes are actively reversed. Based on the data presented in this study, we further suggest that the same conditions that had led to adaptation of the VOR must be present to readapt the VOR back to its original state, and that the orientation of the head relative to gravity is an important context for readaptation as for adaptation. The gravity-dependent gain changes of the vertical VOR induced in the upright position returned to normal faster than gains adapted on side. A likely explanation for this is that monkeys predominantly assume an upright position, so that changes related to movements that activate the vertical VOR about this position are detected and corrected more readily than for similar movements made in animals that were adapted on side.

The finding that a specific amount of gain change is induced in every head orientation after the VOR has been adapted in a particular position suggests that central neurons responsible for gain modification of the VOR receive both semicircular canal and otolith related inputs. The canal input sets the direction of the movement relative to the head and provides a measure that can be compared to the visual input, while the otolith input alters the gain of the VOR as a function of the orientation of the head re gravity. There are several places where this interaction could occur. One is in the nodulus and uvula. This portion of the vestibulocerebellum does not have a significant role in VOR gain adaptation,⁶ but does receive direct canal and otolith input,^44,45 as well as visual input,⁴⁶ and is important for gravity-dependent spatial orientation of the VOR. Data in this study rule out significant nodulus and uvular participation in the gravity-dependent gain changes, however. The flocculus, which is known to control VOR adaptation,⁴⁷ or the vestibular and/or cerebellar nuclei that receive convergent canal and otolith inputs,^48–52 are other likely sites that could contribute to the implementation of the gravity-dependent gain changes.