The cervico-ocular reflex

Abstract

This chapter discusses that stabilization of a visual target on the fovea of the retina during target or head motion is necessary, in order to achieve through various oculomotor mechanisms. Target or head velocity, and frequency are the variables that the brain uses to determine the oculomotor system that is recruited. For individuals with vestibular hypofunction, additional compensatory mechanisms may be used to enhance gaze stability when the head is moving. The chapter focuses on the studies of the cervico-ocular reflex (COR) involving rotation of the trunk beneath a still head. It is important to know that in natural circumstances the COR may be influenced differently by active versus passive head rotation. Investigations of the COR using the methods of non-rotation are also discussed in the chapter. The chapter reviews the methods, and discusses the findings that are based on electrophysiological and behavioral evidence for the COR. It emphasizes on the methods that are used to record the COR.

Full text

Cervico-Ocular Reflex in Normal Subjects and
Patients with Unilateral Vestibular Hypofunction
*Michael C. Schubert, †Vallabh Das, †‡Ronald J. Tusa, and †§Susan J. Herdman
*Laboratory of Vestibular Neurophysiology, Johns Hopkins University School of Medicine, Baltimore,
Maryland, †Departments of Otolaryngology, ‡Neurology, and §Rehabilitation Medicine, Emory University,
Atlanta, Georgia, U.S.A.
Objective: To determine whether the cervico-ocular reflex
contributes to gaze stability in patients with unilateral vestib-
ular hypofunction.
Study Design: Prospective study.
Setting: Tertiary referral center.
Patients: Patients with unilateral vestibular hypofunction (n ⳱
3) before and after vestibular rehabilitation and healthy subjects
(n ⳱7).
Interventions: Vestibular rehabilitation.
Main Outcome Measures: We measured the cervico-ocular
reflex in patients with unilateral vestibular hypofunction before
and after vestibular rehabilitation and in healthy subjects. To
measure the cervico-ocular reflex, we recorded eye movements
with a scleral search coil while the trunk moved at 0.3, 1.0, and
1.5 Hz beneath a stabilized head. To determine whether the
head was truly stabilized, we measured head movement using
a search coil.
Results: We found no evidence of cervico-ocular reflex in any
of the seven healthy subjects or in two of the patients with
unilateral vestibular hypofunction. In one patient with chronic
unilateral vestibular hypofunction, the cervico-ocular reflex
was present before vestibular rehabilitation only for leftward
trunk rotation (relative head rotation toward the intact side).
After 5 weeks of placebo exercises, there was no change in the
cervico-ocular reflex. After an additional 5 weeks that included
vestibular exercises, cervico-ocular reflex gain for leftward
trunk rotation had increased threefold. In addition, there was
now evidence of a cervico-ocular reflex for rightward trunk
rotation, potentially compensating for the vestibular deficit.
Conclusion: The cervico-ocular reflex appears to be a highly
inconsistent mechanism. The change of the cervico-ocular re-
flex in one patient after vestibular exercises suggests that the
cervico-ocular reflex may be adaptable in some patients. Key
Words: Cervico-ocular reflex—Vestibular hypofunction.
Otol Neurotol 25:65–71, 2004.
The cervico-ocular reflex (COR) has been proposed as
a mechanism of gaze stability for subjects with bilateral
(1–5) and unilateral vestibular hypofunction (6,7). This
hypothesis is based, in part, on the finding that COR gain
in subjects with vestibular hypofunction is greater than
COR gain in healthy subjects (3,8,9). Studies of subjects
with healthy vestibular systems report COR gain values
that vary from 0 to 0.4 (2,8–10). Of all these studies
investigating COR in healthy individuals, only one has
measured head movements and eye movements to con-
trol for head movement producing a vestibulo-ocular re-
flex during testing (10). In that study of eight healthy
subjects, methods were used to reduce head movements
to less than 0.04 degree, and COR gain was always less
than 0.07. Given the variability among separate studies
and the question of adequate stabilization of the head
during measurement of COR, it is not clear whether the
COR is present in healthy subjects or in patients with
unilateral vestibular hypofunction (UVH).
The purpose of this study was to measure COR while
head movement is monitored in patients with UVH and
in healthy subjects. We also investigated the effects of
vestibular adaptation exercises on COR gain in subjects
with vestibular hypofunction. This article presents evi-
dence that COR is not found in healthy subjects or in the
majority of patients with unilateral vestibular loss. One
patient with UVH was found to have a COR that in-
creased after vestibular adaptation exercises.
PATIENTS AND METHODS
Subject Characteristics
Three patients with UVH and seven healthy controls pro-
vided informed consent in compliance with the Emory Univer-
sity Institutional Review Board. The patient subjects were
grouped on the basis of diagnoses. Patients with UVH had
greater than or equal to 25% unilateral weakness between the
right and left sides on caloric and rotary chair tests (constant
This research was supported by grant DC-03196 (S.J.H., R.J.T.,
V.D.) and the Foundation for Physical Therapy (M.C.S.).
Address correspondence and reprint requests to Dr. Susan J. Herdman,
Department of Rehabilitation Medicine, Emory University, 1441 Clifton
Road N.E., Atlanta, GA 30322, U.S.A. Email: sherdma@emory.edu
Otology & Neurotology
25:65–71 © 2004, Otology & Neurotology, Inc.
65

velocity rotation, 240 deg/s). Two of the patients with UVH
were diagnosed as having vestibular neuritis on the basis of
history. Magnetic resonance imaging (MRI) of the head was
performed to rule out an acoustic neuroma in two of the pa-
tients with UVH. All normal subjects had normal caloric tests.
Each of the subjects was screened for cervical abnormality.
Measurement of COR and Vestibulo-Ocular Reflex
During COR and vestibulo-ocular reflex (VOR) testing, the
subject was seated in the rotary chair and the trunk was stabi-
lized using a dynamic air splint (Kramer; Cramer Products,
Gardner, KS, U.S.A.). Velcro bands were used to further sta-
bilize the body in the chair. Eye and head position were mea-
sured using search coils. The COR was measured by recording
eye movements during trunk oscillation while the subject’s
head was stabilized using a bite bar wrapped in dental wax
(positioned in front of the subject). During COR testing, trunk
and knees moved in the direction of the rotating chair. The
VOR was measured by recording eye movements during
whole-body sinusoidal rotation. Caution was taken to ensure
comfort of the subject and that the head was positioned in 30
degrees of flexion (11).
Protocol
Cervico-ocular reflex and VOR were recorded at 0.3, 1.0,
and 1.5 Hz (peak velocities, 24, 31, and 34 deg/s, respectively).
All tests were performed in complete darkness. The subjects
were not given specific instructions during COR testing, al-
though an investigator was present in the room to ensure alert-
ness and to monitor the comfort of the subjects. Subjects were
asked to perform naming tasks to ensure alertness during VOR
testing. Subjects with vestibular hypofunction were tested at the
time of the initial assessment, after 4 weeks of placebo exer-
cises, and after 5 weeks of vestibular adaptation exercises.
Protocol to Increase COR Gain with
Mental Imagery
In addition to the above-described protocol, three of the
healthy control subjects performed tasks to determine whether
the COR could be enhanced through mental imagery. Specifi-
cally, the subjects were asked to 1) imagine a target on their
knees; 2) concentrate on the motion of their trunks; and 3)
imagine the direction of the relative head rotation.
Data Capture and Analysis
Eye and head position was measured using 6-foot magnetic
field coils (CNC Engineering, Seattle, WA). Search coils
(Skalar, Delft, The Netherlands) were precalibrated by measur-
ing changes in voltages occurring during known rotations. The
system was 99% linear over an operating range of ±25 degrees.
Horizontal and vertical eye and head position as well as chair
tachometer and chair position were converted to digital signals
using software written in LabVIEW (National Instruments,
Austin, TX).
All data from the coil experiments were stored on the hard
drive of a Dell Dimension Pentium III (Dell Computer Corpo-
ration, Austin, TX) desktop computer for offline analysis. Gaze
and head position signals were filtered at 200 Hz using six-pole
Bessel antialiasing filters before digitization at 1 kHz with 16-
bit precision (Krohn-Hite Corp., Avon, MA). Horizontal and
vertical eye and head velocities were differentiated from posi-
tion signals using a two-point central difference algorithm.
Horizontal and vertical eye and head accelerations were gen-
erated by differentiating velocity arrays. Eye in orbit velocity
and position was determined by subtracting head velocity and
position signals from gaze velocity and position signals.
Mean peak COR gain was calculated as the ratio of peak eye
velocity to peak trunk (chair) velocity. In determining the mean
peak COR gain, we corrected (because of unwanted head mo-
tion) for the contribution of VOR (ratio of peak eye velocity to
peak head velocity) by multiplying the slow component eye
velocity (SCEV) on the basis of the VOR gain measured
at each of the three frequencies (0.3, 1.0, and 1.5 Hz). We
then subtracted that value from the SCEV measured during
COR testing.
Right and left hemicycles were analyzed individually. Only
COR gain values greater than 0.06 were considered significant
and used for statistical analysis. The value of 0.06 was chosen,
as this represents the limit at which we could accurately detect
the eye position signal.
Exercises
Placebo exercises consisted of saccadic eye movements with
the head stationary while viewing a Ganzfeld (blank surface).
Vestibular exercises included adaptation exercises and eye-
head exercises to targets (Table 1) (12). All exercises were to
be performed four to five times daily for a total of 30 to 40
TABLE 1. Progression of vestibular rehabilitation to improve gaze stability
Exercise weekly progression Duration Frequency
Week 1: X1 with target held in hand and also with target at distance, horizontal and
vertical head movements
1 min each exercise Five times daily
Week 2: X1 with target held in hand and also with target at distance, horizontal and
vertical head movements; also eye head movements between two targets with
emphasis on seeing clearly
1–2 min each exercise Five times daily
Week 3: X1 with target held in hand and also with target at distance, horizontal and
vertical head movements; X1 with checkerboard with target placed in center held in
hand, horizontal head movements; eye head movements between two targets;
imaginary target paradigm
1 min each exercise Four times daily
Week 4: X1 with checkerboard with target placed in center held in hand; X2 with
target held in hand, horizontal and vertical head movements; eye head movements
between two targets; imaginary target paradigm
1 min each exercise Four times daily
Week 5: X1 with target held in hand; horizontal and vertical head movements;X1
with checkerboard with target placed in center held in hand, horizontal and vertical
head movements; X2 with target held in hand, horizontal and vertical head
movements; eye head movements between two targets; imaginary target paradigm
1 min each exercise Four times daily
X1, head rotates horizontally or vertically while subject views a stationary target; X2, head and target rotate in opposite directions (horizontal or
vertical) while subject attempts to view target.
66 M. C. SCHUBERT ET AL.
Otology & Neurotology, Vol. 25, No. 1, 2004

min/d. Individuals were provided with daily calendars to mark,
ensuring compliance. Subjects brought the calendars to their
weekly visits. Compliance for all of the subjects ranged from
50 to 100%, based on the calendars.
RESULTS
Subject Characteristics
Table 2 shows the individual characteristics for all
subjects. Note that the age range for the normal subjects
and the subjects with UVH are similar (tstatistic ⳱1.04,
p⳱0.16). None of the subjects had neck abnormality.
Two of the three patient subjects had MRI or computed
tomographic scans. There were no intracranial abnor-
malities noted. One patient (Subject UVL35) had an
acoustic neuroma identified. That tumor was removed
before participation in the study.
COR in Healthy Controls and in Other Subjects
with UVH
We found no evidence of COR in any of the normal
subjects. This is illustrated in Figure 1 A for one subject.
Table 3 summarizes the data from all controls. Attempts
to enhance the COR in the healthy control subjects
through mental imagery did not produce a COR. We
found no COR in the other two patients with UVH. An
example of this is shown in Figure 1 B. Note that any
SCEV can be attributed to the small amount of head
movement during testing.
Subject (UVH81) with COR
An 81-year-old individual had a sudden onset of ver-
tigo 11 months before testing (Subject UVH81). The
vertigo resolved over the course of several days, but she
continued to have complaints of disequilibrium and os-
cillopsia. In this patient, bithermal and ice water irriga-
tion showed no response on the left side and normal
response on the right side. Rotary chair testing showed
lower gains for 60 and 240 deg/s constant velocity rota-
tions to the left (0.49 and 0.21) compared with rotation to
the right (0.59 and 0.34). Magnetic resonance imaging
scans were normal. Audiography showed mild to mod-
erately severe, symmetric sensorineural hearing loss at
high frequencies bilaterally. The patient was diagnosed
as having left vestibular neuronitis on the basis of history
and examination.
Table 4 summarizes COR and VOR gain for this pa-
tient at initial assessment, after placebo exercises, and
after vestibular adaptation exercises. Cervico-ocular re-
flex was evident during the initial assessment only for
trunk rotation to the left at 0.3 Hz (relative head rotation
to the right) (Fig. 2A). The mean gain of the COR was
0.10 ± 0.04 (range, 0.06–0.16). After 4 weeks of placebo
exercises, there were no appreciable changes in COR.
After 5 weeks of vestibular rehabilitation, during 0.3-Hz
trunk rotations, mean COR gain had increased threefold
(0.32 ± 0.13). Peak slow eye velocity during trunk rota-
tion to the left had increased to 7.8 ± 3.0 deg/s, whereas
the mean head velocity was only 0.6 ± 0.4 deg/s (Fig.
2B). COR was also identified at 1.0- and 1.5-Hz trunk
rotation to the left (COR gain ⳱0.13 ± 0.04 and 0.13 ±
0.05, respectively). There was an indication of COR for
rightward trunk rotation (relative head rotation to the
left), but only in a few trials (Table 4).
DISCUSSION
COR in Healthy Individuals
We found no measurable COR in seven healthy indi-
viduals. This differs from previous reports (2,8,9). We
believe the explanation for this difference is the degree
of head stabilization during COR testing. We measured
head stabilization using a search coil attached to the
TABLE 2. Subject characteristics: initial assessment
Subject Age (yr)
Time from
onset (mo)
Head thrust
test
Caloric asymmetry
(% loss)
Rotary chair
VOR gain
a
Tc (s)
b
UVH 35 2 + Right 100 Left ⳱0.565 Left ⳱3.8
Right ⳱0.362 Right ⳱5.4
UVH 52 5.5 + Left 100 Left ⳱0.205 Left ⳱10.1
Right ⳱0.427 Right ⳱6.5
UVH 81 11 + Left 100 Left ⳱0.209 Left ⳱7.9
Right ⳱0.336 Right ⳱10.3
Normal 31 NA –Bil 20 Not tested
c
Not tested
c
Normal 36 NA –Bil 4 Not tested
c
Not tested
c
Normal 40 NA –Bil 16 Not tested
c
Not tested
c
Normal 83 NA –Bil 4 Not tested
c
Not tested
c
Normal 40 NA –Bil 3 Not tested
c
Not tested
c
Normal 34 NA –Bil 7 Not tested
c
Not tested
c
Normal 27 NA –Bil Not tested
d
Left ⳱0.408 Left ⳱13.5
Right ⳱0.424 Right ⳱24.4
a
VOR gain for 240-deg/s step rotations to the right and left.
b
Time constant for 60-deg/s step rotations to the right and left.
c
In normal subjects, vestibular function was assessed usually with caloric testing.
d
Caloric could not be performed adequately because of scarring of tympanic membrane.
+, Side of unilateral vestibular hypofunction; –Bil, negative head thrusts in both directions; VOR, vestibulo-ocular reflex; UVH, unilateral
vestibular hypofunction; NA, not applicable.
67CERVICO-OCULAR REFLEX
Otology & Neurotology, Vol. 25, No. 1, 2004

head. In our pilot studies with healthy control subjects
and with subjects with vestibular hypofunction, eye
movements attributed initially to COR were all related to
excessive head movements. Because of this, we took
several steps to eliminate head movements. First, we
found it necessary to wrap the lower trunk and hips in a
dynamic air splint to stabilize the body. Second, we had
to teach subjects how to bite down on the bite bar to
prevent unwanted head movements. Finally, training ses-
sions were performed to ensure the head remained stable.
Even with all of this preparation, there was still some
head movement present, which caused a VOR response.
By subtracting the VOR eye response during trunk rota-
tion, we feel we have been able to reliably measure COR
alone. Our results of no recordable COR in healthy sub-
jects is in agreement with Sawyer et al. (10), who also
measured head movements in eight normal subjects.
Effects of Mental Set
Some studies of normal subjects and of subjects with
bilateral vestibular hypofunction have reported that men-
tal set enhances the COR (9,13). We attempted to repro-
duce these results in a series of experiments on three of
the healthy subjects. The conditions we used included
asking the subjects to imagine a target on their knees, to
concentrate on the motion of their trunks, and to imagine
the direction of the relative head rotation. These three
conditions were used in an effort to enhance COR. For
example, imagining fixation on the knees would result in
SCEV of COR in the direction of the knees. However,
we were not able to identify measurable eye movements
that could be COR in any of the conditions. Our results,
therefore, are similar to those of Sawyer et al. (10), who
also could not identify a COR, even when manipulating
mental set. Again, we think that differences among these
studies are related to the degree of head stabilization
during COR testing.
COR in Individuals with UVH
We found no COR in two of three patients with UVH
before and after vestibular exercises. The single patient
in whom we did find COR was an 81-year-old subject
who developed sudden onset of vertigo 11 months before
testing. We believe the SCEV generated by rotation of
the trunk while the head was stabilized in Subject
UVH81 was due to the COR. The SCEV (2.8 deg/s)
generated was more than 10-fold the velocity of head
movement (0.23 deg/s) generated during trunk on head
rotation.
It is unlikely that other mechanisms such as spontane-
ous or gaze-evoked nystagmus or certain orienting strat-
egies are responsible for the SCEV we found in Subject
TABLE 3. Mean values for healthy controls
Test frequency
(Hz)
Eye velocity (deg/s)
(mean ± 1 SD) during COR
Head velocity (deg/s)
(mean ± 1 SD) during COR
COR gain
(mean ± 1 SD)
VOR gain
(mean ± 1 SD)
0.3
a
0.81 ± 0.82 0.89 ± 0.3 0.006 ± 0.04
b
0.66 ± 0.1
1.0 1.2 ± 1.3 1.17 ± 1.19 0.02 ± 0.02 0.68 ± 0.003
1.5 1.5 ± 1.7 1.94 ± 1.95 0.01 ± 0.01 0.84 ± 0.002
a
COR was measured in all seven subjects at 0.3 Hz but in only four subjects at 1.0 and 1.5 Hz.
b
This value is below the resolution of position signal from eye coil.
COR, cervico-ocular reflex; VOR, vestibulo-ocular reflex.
FIG. 1. No evidence of a COR is seen in a healthy control (A)
subject. Chair velocity peaks at 24 deg/s. Neither eye nor head
velocities are greater than 1.6 deg/s. All traces are of motion in
the horizontal plane. Eye velocity trials have been desaccaded.
Slow component eye velocity trials are not inverted because the
head is intended to be stable. Positive numbers along the ordi-
nate indicate rightward velocity rotation, whereas negative num-
bers indicate leftward velocity rotation. Dashed line placed at
zero velocity is for reference. (B) Subject with complete right
UVH. No evidence of a COR in a subject with complete right
UVH. Neither eye nor head velocities are greater than 1.4 deg/s
(see Fig. 1 A for legend).
68 M. C. SCHUBERT ET AL.
Otology & Neurotology, Vol. 25, No. 1, 2004

UVH81 during trunk on head rotation. Spontaneous nys-
tagmus in a patient with a left UVH should generate an
SCEV to the right. In Subject UVH81 the direction of the
SCEV was to the left, but the quick phase component
changed directions, which is not congruent with sponta-
neous nystagmus (Fig. 3). Gaze-holding nystagmus can
occur when the eyes are positioned as little as 15 to 30
degrees eccentrically, and is absent when the eyes are
centered in the orbit (14). In addition, in gaze-holding
nystagmus, the direction of the slow component eye
movements is dependent on eye in orbit position. In our
patient, eye position during trunk rotations did not ex-
ceed 6 degrees eccentrically, which is not sufficient to
elicit gaze-holding nystagmus. Furthermore, in our pa-
tient, the direction of the slow component eye move-
ments was always to the left, regardless of eye in orbit
position (Fig. 3). Thus, gaze-holding nystagmus does not
explain the slow eye velocity we identified during trunk
on head rotation in our patient. It is also unlikely that the
SCEV in Subject UVH81 is due to an orienting strategy.
One orienting strategy involves the generation of antici-
patory smooth eye movements in response to target mo-
tion on the retina (15,16). Anticipatory smooth eye
movements have not been found in the absence of a
visual target, however (16), such as in our paradigm. In
another orienting strategy, the eyes predict the eventual
head/chair position and “jump”ahead, in the direction of
the eventual head/chair position (17). As can be seen in
Figure 3, just as the chair changes from rotating to the
left to rotating to the right, the eyes quickly jump to the
right. Similarly, quick-phase eye movements to the left
occur just as the chair begins rotating to the left. The
presence of the quick-phase eye movements of this par-
ticular orienting response, however, would not explain
the slow-phase eye velocities during trunk on head rota-
tion that we believe is COR.
It is interesting that the COR was present initially only
during ipsilesional trunk rotation (head relative right)
and that initially we did not find a COR for relative head
rotations left (toward the side of the lesion). If the COR
serves a compensatory role as a substitute for the VOR,
one would expect a COR for relative head rotations to-
ward the left, the side of this subject’s vestibular lesion.
One possibility is that the COR was present in this pa-
tient before the onset of her vestibular deficit. The COR
was then lost for one direction with the occurrence of the
unilateral vestibular loss.
COR Adaptation
Our results are the first demonstration of COR adap-
tation in a patient with UVH. Mean COR gain toward the
unaffected side did not change during a 4-week period of
placebo exercises, but increased from 0.1 to 0.32 (0.3
Hz) after 5 weeks of vestibular adaptation exercises. In
addition, COR was now present for 1.0 and 1.5 Hz. Be-
fore the initiation of vestibular exercises, a COR was not
present for ipsilesional head movement (relative). This
increased to a gain of 0.1 after rehabilitation. Heimbrand
et al. (13) demonstrated adaptation of the COR using
magnifying lenses in patients with bilateral vestibular
loss. Their adaptation paradigm is similar to the stimuli
inherent in the exercises performed by our patient which,
like magnifying glasses, are designed to produce retinal
slip. It is possible therefore that the exercises designed to
enhance adaptation of the vestibular system may induce
adaptation of the COR. Our findings of a COR at fre-
quencies of 1.0 and 1.5 Hz indicate that the COR may be
useful with some activities.
Contribution of COR to Gaze Stability
Although the number of patients in this study is small,
our data suggest that the COR may not contribute sig-
nificantly to gaze stability during head movements in
patients with UVH. First, it appears to be an inconsistent
response. It was present in only one of three subjects and,
even in that subject, was not found consistently. Second,
TABLE 4. Cervico-ocular reflex and vesibulo-ocular reflex gain across frequencies for subject UVH81 with
cervico-ocular reflex
Frequency
Initial
assessment
After
placebo exercises
After adaptation
vestibular exercises
COR gain
(mean ± 1 SD)
VOR gain
(mean + 1 SD)
COR gain
(mean ± 1 SD)
VOR gain
(mean + 1 SD)
COR gain
(mean ± 1 SD)
VOR gain
(mean + 1 SD)
0.01 ± 0.01 0.03 ± 0.03 0.09 ± 0.03
0.3 Hz (0 of 18 trials) 0.70 ± 0.11 (0 of 11 trials) 0.67 ± 0.05 (2 of 14 trials) 0.61 ± 0.06
VOR - WBR 0.02 ± 0.15 0.02 ± 0.02 0.08 ± 0.04
(ipsilesional) 1.0 Hz (0 of 16 trials) 0.62 ± 0.06 (0 of 7 trials) 0.78 ± 0.16 (5 of 30 trials) 0.44 ± 0.08
COR (head 0.020 ± 0.02
relative left) (0 of 20 trials) 0.01 ± 0.02 0.04 ± 0.04
1.5 Hz 0.64 ± 0.04 (0 of 17 trials) 0.59 ± 0.06 (0 of 30 trials) 0.90 ± 0.19
0.10 ± 0.04 0.10 ± 0.04 0.32 ± 0.13
0.3 Hz (8 of 18 trials) 0.97 ± 0.09 (4 of 11 trials) 0.92 ± 0.07 (11 of 18 trials) 0.78 ± 0.04
VOR - WBR 0.06 ± 0.01 0.06 0.13 ± 0.04
(contralesional) 1.0 Hz (0 of 16 trials) 0.88 ± 0.05 (1 of 6 trials) 0.89 ± 0.05 (15 of 30 trials) 0.90 ± 0.08
COR (head 0.04 ± 0.02 0.08 ± 0.01 0.13 ± 0.05
relative right) 1.5 Hz (0 of 20 trials) 0.85 ± 0.04 (0 of 16 trials) 0.98 ± 0.07 (6 of 30 trials) 0.95 ± 0.16
COR, cervico-ocular reflex; VOR, vestibulo-ocular reflex; WBR, whole-body rotation in the dark.
69CERVICO-OCULAR REFLEX
Otology & Neurotology, Vol. 25, No. 1, 2004

COR gain was quite low. The gain of the COR therefore
would not prevent significant retinal slip during most
normal head movements. Third, the velocities at which
COR was present were quite low (24–34 deg/s). The
velocity of head movements during many activities of
daily living typically exceeds 100 deg/s (18). Finally, it
is not clear in what way COR would aid gaze stability
when it was not in phase with the relative head move-
ment. It is possible, however, that COR might contribute
to gaze stability in some way during slow head movements.
CONCLUSION
The COR is difficult to elicit with passive trunk on
head rotation in normal subjects and in subjects with
unilateral vestibular hypofunction. The difference be-
tween our results and earlier studies on COR appears to
be the degree of head stabilization, confirmed by mea-
surement. We found evidence of a COR in only one
patient with a unilateral vestibular hypofunction. Ini-
tially, the COR was present only for trunk rotation to the
left (relative head rotation toward the intact side). Ves-
tibular adaptation and eye-head movement exercises ap-
pear to produce an increase in the gain of the COR for
both directions, although the asymmetry remained. If the
COR contributes to gaze stability, it would be in a very
limited way and only for lower velocity head rotation.
REFERENCES
1. Kasai T, Zee DS. Eye-head coordination in labyrinthine-defective
human beings. Brain Res 1978;144:123–41.
2. Chambers BR, Mai M, Barber HO. Bilateral vestibular loss, oscil-
lopsia, and the cervico-ocular reflex. Otolaryngol Head Neck Surg
1985;93:403–7.
3. Bronstein AM. Plastic changes in the human cervicoocular reflex.
Ann N Y Acad Sci 1992;656:708–15.
4. Bronstein AM, Morland AB, Ruddock KH, et al. Recovery from
bilateral vestibular failure: implications for visual and cervico-
ocular function. Acta Otolaryngol Suppl 1995;520:405–7.
5. Maurer C, Mergner T, Becker W, Jurgens R. Eye-head coordina-
tion in labyrinthine-defective humans. Exp Brain Res 1998;122:
260–74.
6. Sklyut IA, Likhachev SA, Sklyut MI. Role of cervical propriocep-
tive afferentation in mechanisms of compensation of vestibular
dysfunction: 1—passive cervicoocular reflex in unilateral labyrin-
thine lesion. Vestn Otorinolaringol 1999;2:34–8.
FIG. 2. (A) Subject with complete left UVH before vestibular
rehabilitation. Evidence of COR during 0.3-Hz chair rotation to
the left in a patient with left UVH (Subject UVH81) at the time of
the initial assessment. After chair (trunk) velocity peaks to the left
at 24 deg/s (relative head rotation to the right), an SCEV occurs
to the left with a mean 2.8 ± 0.7 deg/s. Mean head velocity is
stable at 0.23 ± 0.3 deg/s. All traces are of motion in the hori-
zontal plane. Eye velocity trials have been desaccaded. Positive
numbers along ordinate indicate rightward velocity rotation and
negative numbers indicate leftward velocity rotation. Dashed line
placed at zero velocity for reference. (B) Subject with complete
left UVH after vestibular rehabilitation. Increased COR gain dur-
ing 0.3-Hz rotation to the left after 5 weeks of vestibular rehabili-
tation in Subject UVH81. Mean SCEV increased to 7.8 ± 3.0
deg/s, although head velocity was stable at 0.6 ± 0.4 deg/s (see
Fig. 2 A for legend).
FIG. 3. Position trace of subject with complete left UVH. Posi-
tion plot of eye and head in Subject UVH81 during trunk on head
COR trial. (Bold arrows) Quick-phase eye position, which
changes orientation as the trunk position changes. (Vertical ar-
rows) Slow-phase eye position, which does not change. All traces
are of motion in the horizontal plane. Positive numbers along
ordinate indicate rightward position and negative numbers indi-
cate leftward position. Dashed line are placed at zero velocity for
reference.
70 M. C. SCHUBERT ET AL.
Otology & Neurotology, Vol. 25, No. 1, 2004

7. Sklyut IA, Likhachev SA, Sklyut MI. Role of cervical propriocep-
tive afferentation in mechanisms of compensation of vestibular
dysfunction: 1—active cervicoocular reflex in unilateral lesion of
the labyrinth. Vestn Otorinolaringol 1999;5:37–42.
8. Bronstein AM, Hood JD. The cervico-ocular reflex in normal sub-
jects and patients with absent vestibular function. Brain Res 1986;
373:399–408.
9. Huygen PLM, Verhagen WIM, Nicolasen MGM. Cervico-ocular
reflex enhancement in labyrinthine-defective and normal subjects.
Exp Brain Res 1991;87:457–64.
10. Sawyer RN, Thurston SE, Becker KR, Ackley CV, Seidman SH,
Leigh RJ. The cervico-ocular reflex of normal human subjects in
response to transient and sinusoidal trunk rotations. J Vestib Res
1994;4:245–9.
11. Fetter M, Hain TC, Zee DS. Influence of eye and head position on
the vestibulo-ocular reflex. Exp Brain Res 1986;64:208–16.
12. Herdman SJ, Clendaniel RA. Assessment and treatment of com-
plete vestibular loss. In: Herdman SJ, ed. Vestibular Rehabilita-
tion. Philadelphia: FA Davis Co, 2000:424–50.
13. Heimbrand S, Bronstein AM, Gresty MA, Faldon ME. Optically
induced plasticity of the cervico-ocular reflex in patients with bi-
lateral absence of vestibular function. Exp Brain Res 1996;112:
372–80.
14. Zee DS, Yee RD, Cogan DG, et al. Ocular motor abnormalities in
hereditary cerebellar ataxia. Brain 1976;99:207–34.
15. Kowler E, Steinman RM. The effect of expectations on slow ocu-
lomotor control: I—periodic target steps. Vision Res 1979;19:
619–32.
16. Kowler E, Steinman RM. The effect of expectations on slow ocu-
lomotor control: II—single target displacements. Vision Res 1979;
19:633–46.
17. Grasso R, Prevost P, Ivanenko YP, et al. Eye-head coordination for
the steering of locomotion in humans: an anticipatory strategy.
Neurosci Lett 1998;252:115–8.
18. Black RA, Halmagyi GM, Curthoys IS, Thurtell MJ, Brizuela AE.
Unilateral vestibular deafferentation produces no long-term effects
on human active eye-head coordination. Exp Brain Res 1998;122:
362–6.
71CERVICO-OCULAR REFLEX
Otology & Neurotology, Vol. 25, No. 1, 2004