| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Neurophysiol (November 1, 2002). 10.1152/jn.00796.2001
Submitted on 27 September 2001
Accepted on 15 July 2002
Department of Neurology, University of Zurich, 8091 Zurich, Switzerland
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Misslisch, H. and B.J.M. Hess. Combined Influence of Vergence and Eye Position on Three-Dimensional Vestibulo-Ocular Reflex in the Monkey. J. Neurophysiol. 88: 2368-2376, 2002. This study examined two kinematical features of the rotational vestibulo-ocular reflex (VOR) of the monkey in near vision. First, is there an effect of eye position on the axes of eye rotation during yaw, pitch and roll head rotations when the eyes are converged to fixate near targets? Second, do the three-dimensional positions of the left and right eye during yaw and roll head rotations obey the binocular extension of Listing's law (L2), showing eye position planes that rotate temporally by a quarter as far as the angle of horizontal vergence? Animals fixated near visual targets requiring 17 or 8.5° vergence and placed at straight ahead, 20° up, down, left, or right during yaw, pitch, and roll head rotations at 1 Hz. The 17° vergence experiments were performed both with and without a structured visual background, the 8.5° vergence experiments with a visual background only. A 40° horizontal change in eye position never influenced the axis of eye rotation produced by the VOR during pitch head rotation. Eye position did not affect the VOR eye rotation axes, which stayed aligned with the yaw and roll head rotation axes, when torsional gain was high. If torsional gain was low, eccentric eye positions produced yaw and roll VOR eye rotation axes that tilted somewhat in the directions predicted by Listing's law, i.e., with or opposite to gaze during yaw or roll. These findings were seen in both visual conditions and in both vergence experiments. During yaw and roll head rotations with a 40° vertical change in gaze, torsional eye position followed on average the prediction of L2: the left eye showed counterclockwise (ex-) torsion in down gaze and clockwise (in-) torsion in up gaze and vice versa for the right eye. In other words, the left and right eye's position plane rotated temporally by about a quarter of the horizontal vergence angle. Our results indicate that torsional gain is the central mechanism by which the brain adjusts the retinal image stabilizing function of the VOR both in far and near vision and the three dimensional eye positions during yaw and roll head rotations in near vision follow on average the predictions of L2, a kinematic pattern that is maintained by the saccadic/quick phase system.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
When animals move their head,
without compensatory eye movements the image of the visual surround
would shift across the retina and greatly degrade vision. To avoid
retinal slip, the rotational vestibulo-ocular reflex (VOR) spins the
eyes around the same axis as the head but in the opposite
direction. This theoretically optimal VOR behavior would stabilize the
image on the entire retina. In the absence of vergence, i.e., for
distant target viewing, the optimal VOR strategy exists in the monkey
(Misslisch and Hess 2000
) but not in humans with the
latter showing a compromise between optimal VOR and Listing's law
behavior (Misslisch and Tweed 2001; Misslisch et al. 1994, 1996; Solomon et al.
1997; Thurtell et al. 1999; Tweed et al.
1994a). What happens with the optimal VOR strategy observed in
monkeys during far viewing when using near targets, i.e., when the
vergence system, which converges or diverges the eyes to allow fixation
of targets at different depths, plays a part? The first goal of this
study was to determine the effect of eccentric eye position on the yaw,
pitch, and roll VOR when the eyes were converged to fixate eccentric
targets placed on a 0.1- and 0.2-m distant isovergence screen
(requiring 17 and 8.5° vergence). In addition, we examined whether
the near VOR depended on peripheral visual input by testing subjects
with and without presentation of a structured visual background.
Listing's law is a kinematic constraint on eye movements,
which has been shown to be valid for saccades, fixations (e.g., Ferman et al. 1987b; Minken et al.
1993; Tweed and Vilis 1990), and smooth
pursuit (Haslwanter et al. 1991; Tweed et al.
1992) when subjects were refixating between or tracking far
targets (with the head upright and stationary). When describing eye
positions as rotations (around fixed axes) away from a special
reference position called primary position, Listing's law implies that
the eye rotation axes are confined to a roughly fronto-parallel
head-fixed plane, called Listing's plane, which is orthogonal to the
gaze direction in primary position (Tweed and Vilis
1990; von Helmholtz 1867). Listing's law is
modified by the degree of horizontal vergence. When fixating near
targets, the position planes of each eye (best called primary planes)
(see Tweed 1997) rotate temporally by a certain angle.
Dividing the experimentally determined angle of the primary planes by
the vergence angle yielded a ratio of 0.21 in monkeys (Misslisch
et al. 2001) and ratios ranging between 0.17 and 0.25 in humans
(Bruno and van den Berg 1997; Kapoula et al.
1999; Mikhael et al. 1995; Minken and van
Gisbergen 1994; Mok et al. 1992; Somani
et al. 1998; Steffen et al. 2000; Tweed 1997; van Rijn and van den Berg 1993).
Theoretical considerations suggested that the optimal ratio is 0.25:
this value rotates the eyes such that images of the visual plane
the
plane containing both lines of sightin the two retinas are perfectly
aligned (Tweed 1997; van Rijn and van den Berg
1993). This kinematic pattern has been termed the binocular
extension of Listing's law or L2 (Tweed 1997). Do eye
positions during head rotations lie on temporally rotated primary
planes when the eyes are converged? As a second aim of this study, we
examined the influence of L2 on three-dimensional (3D) eye positions
during yaw and roll head rotations.
We found that VOR eye rotation axes were not affected by eccentric eye
position when torsional gain was highindependent of the degree of
vergence or the visual condition. During yaw and roll head rotations
with different near target elevations, the 3D positions of the left and
right eye lay on average in temporally rotated planes, i.e., they
followed the kinematic pattern predicted by L2. While the former
finding excludes an influence of Listing's law or L2 on VOR
velocity, the latter finding indicates an adherence of 3D
eye position to L2, maintained by the quick phase system.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Animal preparation
We implanted a head-holding device as well as dual search coils
on both eyes (Hess 1990) in two female rhesus monkeys
(Macaca mulatta; abbreviated SU and
JU, who participated in our previous study on the influence
of eye position on the VOR in far viewing) (see Misslisch and
Hess 2000). Surgical procedures were performed under sterile
conditions with the animals in deep anesthesia. Animals were treated
with antibiotics and analgesics postsurgically. All procedures were in
accordance with the National Institutes of Health Guide for the Care
and Use of Laboratory Animals and approved by the Veterinary Office of
the Canton of Zurich.
Measurement and representation of 3D eye position and eye velocity
We applied the magnetic field search coil technique
(Robinson 1963) to measure the 3D angular position of
the two eyes (Skalar, eye position meter 3000). Coil voltages, head
position, and head velocity signals, as well as a photodiode signal
indicating full-field illumination, were sampled at 833 Hz (Cambridge
Electronic Design, Model 1401plus) and stored on hard disk for off-line analysis.
The measured eye position was calibrated as described in detail
elsewhere (Hess et al. 1992). 3D eye positions were
expressed as rotation vectors, and the eye's orientation while looking
at a straight-ahead target was chosen as reference position
(Haustein 1989; Hess et al. 1992). Using
these eye position recordings, we computed the eye angular velocity
vector, 
, as described in Hepp (1990). We expressed
the angular eye position and eye velocity vectors in a head-fixed,
right-handed coordinate system with the x, y, and
z axes pointing along the nasooccipital, interaural and
longitudinal head axis. By definition, positive directions of the
coordinate axes represented clockwise, downward and leftward components
(as seen from the subject's point of view) of eye position and eye velocity.
Experimental set-up and protocols
Animals were seated in a primate chair with the head restrained in an upright position so that the anterior side of the lateral semicircular canals was elevated by roughly 15°. The primate chair was fixed within the inner frame of a vestibular rotator, equipped with three motor-driven axes (Acutronic, Jona, Switzerland). The rotator was surrounded by a lightproof sphere of 0.8-m radius. Onset and profiles of sinusoidal chair rotation were computer-controlled.
Animals were used to fixating distant targets during vestibular
stimulation for water reward (Misslisch and Hess 2000).
In the present experiments, we trained animals to fixate near targets placed either eccentrically at 20° up, down, left, and right or at
straight ahead. Two sets of experiments were done, with targets located
on an isovergence surface at a distance of 0.1 m (required horizontal vergence of 17°) or 0.2 m (horizontal vergence of
8.5°). The quality of fixation and convergence was controlled with
behavioral windows.
We tested the VOR in eccentric eye positions using sinusoidal yaw, pitch, and roll chair rotation at 1 Hz (amplitude ±5°; peak velocity: 31.4°/s). Near targets were visible throughout the 4-s test period. All experiments were performed either with (optokinetic random dot pattern visible: 8.5 and 17° vergence experiments) or without a structured visual background (with only the near target visible: 17° vergence experiments). Because the elevation of the arc containing the near target LEDs had to be fixed manually, experiments were carried out in blocks of 10 repeated trials (e.g., 10 yaw head rotations while looking at the LED located 20° up). Vestibular stimulation started when a monkey kept fixation on a close target for 500 ms. Animals were then asked to fixate the earth-stationary near target, keeping the gaze lines converged, during the 4 s of the subsequent head rotation. A trial was aborted when the monkey interrupted target fixation. Animals were rewarded with water for successful trials. For each visual condition (with/without structured background), the total number of trials was 110 (10 trials each for gaze 20° up, 20° down, and center during yaw; 10 trials each for gaze 20° left, 20° right, and center during pitch; and 10 trials each for all 5 gaze directions during roll).
Data analysis
Quick phases of vestibular nystagmus were removed from 3D eye position and eye velocity data by means of a semi-automatic computer program that used thresholds for the second derivative of 3D eye velocity (jerk). Each trial was inspected visually and incorrectly placed markers were identified and corrected interactively. By setting markers, data were limited so that the first (last) sample corresponded to the onset (offset) of vestibular stimulation.
We determined 3D VOR velocity as a function of head velocity and eye
position by performing the same multivariable function fitting
(Press et al. 1988) as used in our previous studies
(Misslisch and Hess 2000; Misslisch et al. 1994,
1996) in which the details of this method are given. Briefly,
we fitted a 3 × 13 generalized gain matrix, which yielded the
minimal least-squared error to fitted 3D eye velocity. The columns of
this matrix represent the influence of ocular drift, torsional,
vertical, and horizontal head velocity, and the product of the three
components of head velocity with each of the components of eye position
on 3D eye velocity. Using this matrix, we then determined, for each
animal and each visual condition, the orientation of 3D VOR eye
velocity for each combination of vestibular stimulation and eccentric
eye position. This method accurately described the experimental data as
can be seen in the examples of VOR responses derived from the best-fit
generalized gain matrices (black curves in Figs.
1 and 2).
To quantify the effect of eye position, we computed how far the eye
rotation axis tilted for a 20° change in vertical (yaw, roll) or
horizontal (pitch, roll) gaze direction, away from straight ahead
(swing angle). For example, when the orientation of the eye
rotation axis during yaw head rotation was 1° up when gaze was center
and 3° up when gaze was 20° up, then the swing angle amounted to
3 
1 = 2° (up). In this case (yaw, gaze 20° up), the
prediction of Listing's law is that the axis of eye rotation swings
10° up (in the direction of gaze). In the case of roll rotation with
gaze 20° up, the prediction of Listing's law is that the eye
rotation axis swings 20° down opposite to the direction of gaze (see
Misslisch and Hess 2000) (parallel projection model with
torsional gain = 0.5). Thus we defined that positive or negative swing angles during yaw or roll were in the direction in accordance with Listing's law.
|
|
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Orientation of binocular eye rotation axis during yaw, pitch, and roll VOR when viewing near eccentric targets
Figure 1 shows an example of VOR eye velocities during sinusoidal yaw head rotation at 1 Hz (4 cycles) with the monkey fixating near targets in different positions, i.e., 20° up, center, and 20° down (top to bottom). Data obtained from the left and right eye are plotted in the left and right column, respectively. In each panel, gray thick curves are horizontal (main component) and torsional eye velocity, black curves are derived from the best-fit generalized gain matrix and the dotted curve is horizontal head velocity.
The close match between data and best-fit curves demonstrate that the
fitted matrices accurately describe the data (see METHODS). More importantly, if vertical eye position influenced the orientation of the VOR eye rotation axis as predicted by Listing's law or L2, we
should see a considerable change in the torsional velocity component as
gaze changes from 20° down to 20° up (Misslisch et al.
1994). As revealed by inspecting Fig. 1, this is not the case: torsional eye velocity is only slightly modulated (in both eyes) when
gaze is 20° up, less so when gaze is center and insignificantly when
gaze is 20° down, suggesting that the axis of slow phase eye velocity
is hardly effected by a very large vertical change in eye position.
This result is much more readily seen when plotting the horizontal
versus the torsional component of eye velocity (Fig. 2). As in Fig. 1,
left and right eye velocities are shown in the left and
right column, and data obtained for fixating near targets at
20° up, center or 20° down are drawn in the top, middle,
and bottom panels. The gray thick curves are the tips of the
eye angular velocity vectors and the superimposed black thin curves
represent the best-fit to the data. By definition, the angular eye
velocity vector points along the eye rotation axis, with its length
being proportional to the speed of eye rotation and the direction of rotation is determined by the right-hand rule: when the thumb of the
right hand points along the rotation axis then the fingers curl round
in the direction of the eye's movement. For instance, eye velocity
vectors pointing upward along the ordinate (z axis of our
coordinate system) represent leftward eye motion and vectors pointing
rightward represent clockwise (cw) eye motion. Examining the
orientation of the eye velocity vectors and their best-fit curves for
the various gaze directions reveals that the VOR eye rotation axis of
both eyes is tilted a little back when the eyes are looking 20° up
(top); when gaze is center, this tilt becomes less; and when
gaze is 20° down, there is almost no tilt. That is, the axes are
approximately aligned with the ordinate, which means that the eye
oscillates almost exclusively horizontally around a head-vertical axis.
Overall, the effect of a 40° change in vertical eye position on the
binocular eye rotation axes during yaw stimulation is negligible
compared with the predictions of Listing's law (half-angle
rule: 10° tilt of eye rotation axis, in the direction of gaze)
and L2 (see Mok et al. 1992).
This result was seen in all trials and in all other types of vestibular stimulation (Fig. 3). In other words, the effect of eye position on the VOR rotation axis is insignificant not only during yaw (Fig. 3A) but also during pitch (Fig. 3B) and roll (Fig. 3, C and D) head rotations. The three panels in each subplot of Fig. 3 show examples of the curves derived from the best-fit generalized gain matrices computed for the left (black) and right (gray) eye data. Figure 3A plots the same best-fit velocity vectors for the left and right eye as in the left and right column of Fig. 2. As mentioned in the preceding text, here the rotation axes of the left and right eye are closely aligned and basically unchanged by a large change in vertical eye position. Figure 3B shows the best-fit vertical (ordinate) and torsional (abscissa) eye velocity components obtained during pitch head rotation. The rotation axes of the two eyes are very little affected by a 40° change in horizontal eye position (Fig. 3B), i.e., the eye motion produced by the pitch VOR in near viewing does not depend on gaze azimuth. Plotting horizontal (Fig. 3C) or vertical (Fig. 3D) versus torsional eye velocity shows that also during head roll the orientation of the left and right eye's rotation axes are invariant for large changes in vertical (Fig. 3C) or horizontal (Fig. 3D) eye position.
|
The orientation of the VOR's eye rotation axis ("swing angle," see
METHODS) depended on torsional gain and on the type of vestibular stimulation. Figure 4
summarizes the swing angles averaged over both monkeys as a function of
torsional gain. Data were obtained during yaw with vertical (Fig.
4A), pitch with horizontal (Fig. 4B), and roll
with vertical (Fig. 4C) or horizontal (Fig. 4D) gaze changes (indicated by 
/
or 
/
pointing arrows).
Positive angles denote a tilt of the eye rotation axis in the direction of gaze, e.g., upward tilt when gaze was 20° up in yaw, and negative angles represent a tilt opposite to the direction of gaze. Because for
all types of vestibular stimulation the swing angles were highly
symmetrical for up versus down and left versus right gaze, we computed
an average swing angle (±SD) from all 20 trials obtained for any
condition (e.g., 10 trials each during yaw with gaze 20° up and 20°
down). The data shown in Fig. 4 were collected while monkeys were
fixating a near target requiring a horizontal vergence angle of 17°,
either with (
and 
) or without (
and 
) a full-field visual
background.
|
Two main results are seen in Fig. 4. First, the amount of the averaged
swing angles was typically smaller than a few degrees, i.e., close to
zero, when torsional gain was large (monkey JU, and
), but somewhat larger when torsional gain was low (monkey SU, and ). The dependence of the swing angles on torsional gain was similar as for the VOR in far vision (- - - and · · · in Fig. 4) (obtained from Eq. A1 in
Misslisch and Hess 2000). An exception to that rule is
the pitch VOR, where swing angles were invariantly very small and
independent of torsional gain. Second, the small swing angles were
invariantly positive during yaw or pitch and negative during roll
stimulation. This pattern is qualitatively consistent with a VOR whose
function is a compromise between optimal (full-field) retinal image
stabilization and Listing's law (far-viewing condition:
Misslisch and Hess 2000; Misslisch and Tweed
2001; Misslisch et al. 1994). However,
quantitatively the swing angles are far from the predictions of
Listing's law or L2 (see preceding text).
The same pattern was observed when horizontal vergence was smaller
(8.5°), i.e., when the near target was 20 cm away from the eyes (Fig.
5): the amount of swing angles was close
to zero when torsional gain was large () and the amount of swing
angles was somewhat larger when torsional gain was low (); and
virtually no tilt of the eye rotation axis during pitch, independent of torsional gain.
|
Interaction of vergence (binocular extension of Listing's law) and yaw or roll VOR
The monkeys that participated in this study kept horizontal
vergence fairly stable throughout the 4 s of near target fixation. Figure 6 shows two examples of this
general observation, plotting horizontal vergence during the 10 trials
each of yaw (left) and roll stimulation (right)
while looking 20° up, center or 20° down (top to
bottom). In these examples, the near target was 10 cm in
front of the monkey's eyes (17° vergence) and the structured background was not visible. As can be seen in the graphs, horizontal vergence was somewhat smaller than the expected ideal value of 17°.
For all three experiments17° vergence condition with and without
visual background, 8.5° vergence condition with visual backgroundthe average gain of horizontal vergence in subject JU or SU was 0.92 ± 0.03, 0.87 ± 0.04, and 0.85 ± 0.03 or 1.03 ± 0.06, 1.01 ± 0.07, and 0.76 ± 0.08.
|
As mentioned in the INTRODUCTION, if 3D eye position during head rotations with near target viewing adhered to the binocular extension of Listing's law, L2, then the slow phase torsional position of the left and right eye should systematically depend on their vertical position. For instance, when gaze is 20° down, one should see counterclockwise (negative) torsion in the left eye and clockwise (positive) torsion in the right eye, and vice versa when vertical eye position is 20° up. We performed two experiments in which we had the subjects change vertical eye position over a 40° range: yaw and roll head rotation with gaze 20° up, center, or 20° down. In the pitch experiment, we varied gaze not in elevation, but in azimuth (20° left, center or 20° right), to determine the effect of eye position on the axis of eye rotation (3D eye velocity) so that we did not evaluate the adherence of 3D eye position to L2 during pitch head rotations.
Figure 7, A and B,
illustrates that during yaw (top) or roll
(bottom) head rotation with gaze 20° up, center, and 20°
down (gray data) 3D eye position, on average, follows L2. More
specifically, when the monkey was looking 20° down, mean torsional
slow phase eye position (see best-fitted black line in Fig.
7A), was negative (counterclockwise) in the left eye
(left) and positive (clockwise) in the right eye
(right). The opposite pattern was seen when the animal was
looking 20° up. In this example, quantifying the temporal tilt of the
best-fit eye position planes yielded 3.6 and 3.9° for the left and
right eye during yaw, 4.6 and 3.5° for the left and right eye during
roll. 3D eye position was kept around the temporally tilted planes by
the VOR quick phases. This is illustrated in Fig. 7B, which
plots torsional versus vertical position of quick phase endpoints of
the same data during yaw (top) or roll (bottom)
head rotation with gaze 20° up, center, and 20° down (gray dots).
When the monkey was looking 20° down, mean torsional quick phase end
position (see best-fitted black line in Fig. 7B), was
negative (counterclockwise) in the left eye (left) and
positive (clockwise) in the right eye (right). The opposite
pattern was seen when the animal was looking 20° up. The temporal
tilt of the best-fit straight lines through quick phase endpoints
yielded (mean ±SD) 3.4 ± 2° and 4.4 ± 1.9° for
the left and right eye during yaw, 4.9 ± 5.4° and 3.7 ± 5.5° for the left and right eye during roll (goodness-of-fit for
straight line fit with 1 df: 
2 = 0.988/0.623
for the left/right eye during yaw and 2 = 0.948/0.791 for left/right eye during roll). A similar pattern was
found when plotting torsional versus vertical position of quick phase
onsets. These results suggest that it is the quick phases of VOR that
keep 3D eye position around tilted planes. Computing the quotients
between these angles of temporal plane rotation and the amount of
vergence (see Fig. 6), i.e., calculating the L2 factors, yielded values
close to 0.25.
|
Figure 8 summarizes the L2 factors for
all yaw and roll conditions where the targets were placed over a 40°
vertical range. In both monkeys and in all conditions17 or 8.5°
vergence, visible or nonvisible backgroundthe L2 factors were near
0.25, the value expected if 3D eye position during yaw and roll head
rotation in near vision perfectly adhered to the binocular extension of Listing's law (Tweed 1997).
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
When monkeys viewed near eccentric targets, their yaw and roll VOR rotated the eyes approximately around the same axis as the head with the axes being more or less collinear if torsional VOR gain was large or weak. The axis of eye rotation produced by the pitch VOR did not depend on (horizontal) eye position nor on torsional gain, invariantly spinning the eyes around an axis that was almost collinear with the head's rotation axis. These findings indicate that Listing's law or L2 did not play a role in determining the orientation of VOR eye velocity. 3D eye positions during yaw and roll head rotations lay on average in temporally rotated planes, meaning that mean ocular torsion in each eye depended in its own way on vertical eye position. Because slow phase eye velocity was not influenced by L2, this result suggests that it is the quick phase system that maintains adherence of 3D eye position to L2.
Influence of eye position on VOR eye velocity in near vision
In three conditionsyaw and roll while looking up or down and
roll while looking left or rightwe found that the axis of eye rotation was almost not influenced by the 40° change in eye position (Figs. 1-5). Deviations from this collinearity of eye and head
rotation axes were very small when torsional gain was high (more than
0.8) and somewhat larger when torsional gain was lower (Figs. 4 and 5).
In the fourth conditionpitch while looking left and rightthe eye
and head rotation axes were well aligned, independent of torsional gain. The visibility of a structured background did not affect these
results, although the torsional gain was consistently larger in the
conditions with visual background (Fig. 4).
A prominent effect of eye position on the axis of eye rotation, i.e.,
on eye velocity, is expected if the VOR in near vision were to follow
Listing's law (e.g., Tweed and Vilis 1990; von Helmholtz 1867) or L2 (Mok et al. 1992). In the
first case, the eye velocity vector should tilt by half the gaze change
(10° if gaze changes 20°) and similarly in the second case. Our
data indicate that both kinematic constraints do not play much of a
role in determining the orientation of slow phase eye velocity in near vision.
There is increasing evidence for the notion that the torsional gain is
involved in the control of the VOR's eye rotation axis. A previous
study on the monkey 3-D VOR in far vision (Misslisch and Hess
2000) showed that normal monkeys have torsional gains close to
one anddespite large changes in eye positionVOR eye rotation axes
aligned with the head's rotation axis. This optimal VOR behavior
stabilizes the entire retinal image. However, it is not hard-wired. For
instance, prominent swings of the VOR eye rotation axis in eccentric
eye positions can be induced by weakening the torsional gain due to
plugging of the monkey's vertical semicircular canals
(Misslisch and Hess 2000).
Moreover, the intricate behavior of the human VORwith the eye
rotation axis tilting about a quarter to a third as far as, and in the
direction of, the gaze line during yaw (Misslisch et al. 1994,
1996; Misslisch and Tweed 2001; Palla et
al. 1999; Solomon et al. 1997) or pitch
(Misslisch and Tweed 2001; Misslisch et al. 1994,
1996) and about as far as the gaze line but in the opposite direction during roll (Misslisch and Tweed 2001;
Misslisch et al. 1994, 1996)can be modeled by a single
factor: a weak VOR in the torsional dimension (Misslisch and
Tweed 2001). That the torsional VOR is weak is a well-known
fact (Berthoz et al. 1981; Collewijn et al.
1985; Ferman et al. 1987a; Misslisch and
Tweed 2000; Robinson 1982; Seidman and
Leigh 1989; Tweed et al. 1994b). Because the
human VOR favors stabilization of foveal and perifoveal retinal areas,
it can also limit the range of ocular torsion, i.e., reduce the
deviations from Listing's law. Thus human and nonhuman primates seem
to share the same common mechanism underlying the VOR's performance in
the stabilization of the retinal image: adjustment of the torsional gain.
Why does the monkey VOR deviate sometimes from the optimal VOR
behavior? Maybe our subjects sometimes paid more or less attention to
the task of stabilizing gaze. That cognitive factors influence the
human VOR performance to a great extent is a well-known fact (e.g.,
Barr et al. 1976; Moller et al. 1990a,b).
Another possible answer may lie in the observation that if there was a
significant tilt of the eye rotation axis, then this tilt was always in
the directions consistent with Listing's lawin the direction of the gaze line during yaw and pitch and opposite to the direction of the
gaze line during roll (positive/negative swing angles in Figs. 4-5).
In other words, the brain may balance several factors such as
minimizing ocular torsion (Listing's law), image stability over
certain parts of the retina and energy expenditure. The situation becomes even more complicated in near vision. Here, the distance and
eccentricity of the target and the distance of the eyes relative to the
head's rotation axis greatly influence the kinematics of the VOR.
Disentangling the relative contribution of these factors is beyond the
scope of this study, however.
Influence of vergence on VOR eye position
This study showed that slow phase eye positions in response to yaw
and roll head rotations with vertically placed near targets oscillated
around temporally rotated planes (Fig. 7). Thus down gaze in the left
or right eye was accompanied, on average, with counterclockwise or
clockwise (ex-) torsion. Likewise, up gaze in the left or right eye
went along with clockwise or counterclockwise (in-) torsion,
respectively. The amount of temporal eye position plane rotation
equaled roughly the one predicted by the binocular extension of
Listing's law, L2, namely a quarter times the horizontal vergence
angle (Tweed 1997). But how can we explain that while slow phase eye velocity clearly violates the prediction of L2, 3D eye
position during yaw and roll head rotation does on average adhere to
L2? The obvious answer to this question is that quick phases rather
than slow phases maintain L2 in near vision.
What is the advantage of the L2 behavior observed during yaw and roll
head rotations during viewing of near targets placed at 20° up,
center and 20° down? In 1997, Tweed proposed the visual-motor theory
of binocular control in his attempt to explain the L2 pattern of eye
positions found during fixations of near targets (humans: Bruno
and van den Berg 1997; Kapoula et al. 1999;
Mikhael et al. 1995; Minken and van Gisbergen
1994; Mok et al. 1992; Somani et al.
1998; Steffen et al. 2000; Tweed
1997; van Rijn and van den Berg 1993; monkeys:
Misslisch et al. 2001). On one hand, L2 keeps the images
of the visual plane aligned; on the other hand, L2 cyclorotates the
eyes about their lines of sight to keep them near their zero-vergence
primary positions. Our data indicate that during the yaw or roll
experiments the quick phases help the stereoptic system in aligning the
visual planes of the left and right eye.
When tested with distant targets, we found earlier that the monkey VOR
does not obey Listing's law at all but optimizes a purely visual
variable, i.e., stabilization of the entire retinal image
(Misslisch and Hess 2000). As the present study shows,
the monkey 3D eye positions during yaw and roll head rotations in near
vision do show the L2 pattern, due to the action of the quick phase
system, suggesting that slow and quick phases balance visual and motor
variables in up and down gazes. In any case, it seems that (binocular)
vision is the dominating factor in determining the eye's motion during
head rotations both in far and near vision. For instance,
Misslisch et al. (2001) found that ocular counterroll is
normal in far but reduced in near vision because ocular torsion would
disrupt stereopsis in the latter case. Thus it seems that the
phylogenetically young stereoptic system takes precedence over the
phylogenetically older VOR system both in the static (counterroll) and
dynamic (rotational) VOR.
| |
ACKNOWLEDGMENTS |
|---|
We thank E. Buffone, B. Disler, and A. Züger for excellent animal care and technical assistance.
This work was supported by the Swiss National Science Foundation Grant 31-47 287.96 and by the Betty and David Koetser Foundation for Brain Research.
| |
FOOTNOTES |
|---|
Address for reprint requests: B.J.M. Hess, Dept. of Neurology, University of Zurich, Frauenklinikstrasse 26, 8091 Zurich, Switzerland (E-mail: bhess{at}neurol.unizh.ch).
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
This article has been cited by other articles:
|
|
 |
|
  B. T. Crane, J. Tian, and J. L. Demer Temporal Dynamics of Ocular Position Dependence of the Initial Human Vestibulo-ocular Reflex. Invest. Ophthalmol. Vis. Sci., April 1, 2006; 47(4): 1426 - 1438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Demer and R. A. Clark Magnetic Resonance Imaging of Human Extraocular Muscles During Static Ocular Counter-Rolling J Neurophysiol, November 1, 2005; 94(5): 3292 - 3302. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Adeyemo and D. E. Angelaki Similar Kinematic Properties for Ocular Following and Smooth Pursuit Eye Movements J Neurophysiol, March 1, 2005; 93(3): 1710 - 1717. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
