Expectation and the Vestibular Control of Balance
Michel Guerraz1 and Brian L. Day 2
Abstract
& Recent experiments have shown that the visual channel of
balance control is susceptible to cognitive influence. When a
subject is aware that an upcoming visual disturbance is likely to
arise from an external agent, that is, movement of the visual
environment, rather than from self-motion, the whole-body
response is suppressed. Here we ask whether this is a principle
that generalizes to the vestibular channel of balance control.
We studied the whole-body response to a pure vestibular
perturbation produced by galvanic vestibular stimulation (GVS;
0.5 mA for 3 sec). In the first experiment, subjects stood with
vision occluded while stimuli were delivered either by the
subject himself (self-triggered) or by the experimenter. For the
latter, the stimulus was delivered either without warning
(unpredictable) or at a fixed interval following an auditory cue
INTRODUCTION
Any unplanned whole-body movement inevitably per-
turbs the position of the head in space. The visual and
vestibular sensory systems detect the ensuing head
motion and feed information back to the brain where
compensatory action is organized. Thus, if a person is
to any extent unstable, for instance, when standing
freely, unexpected changes in either of these sensory
channels evoke powerful balance responses. Such re-
sponses can be evoked either by simply moving the visual
environment relative to the person (e.g., Guerraz, Thilo,
Bronstein, & Gresty, 2001; Guerraz et al., 2000; Bronstein
& Buckwell, 1997; Lestienne, Soechting, & Berthoz, 1977;
Lee & Lishman 1975) or by electrically stimulating the
vestibular system (e.g., Day, Séverac Cauquil, Bartolomei,
Pastor, & Lyon, 1997; Fitzpatrick, Burke, & Gandevia,
1994; Britton et al., 1993; Lund & Broberg, 1983; Nashner
& Wolfson, 1974). However, in both cases the response is
entirely inappropriate because the change in sensory
input is caused by an external agent and not by true
self-motion. Nevertheless, the balance system appears to
interpret the sensory input as having arisen from an
unplanned body movement to which it must respond.
These compensatory responses have been well docu-
mented and are usually considered to occur automati-
1
Université du Sud Toulon–Var, France, 2University College
London
D 2005 Massachusetts Institute of Technology
(predictable). Results showed that GVS evoked a whole-body
response that was not affected by whether the stimulus was
self-triggered, predictable, or unpredictable. The same results
were obtained in a second experiment in which subjects had
access to visual information during vestibular stimulation. We
conclude that the vestibular-evoked balance response is
automatic and immune to knowledge of the source of the
perturbation and its timing. We suggest the reason for this
difference between visual and vestibular channels stems from a
difference in their natural abilities to signal self-motion. The
vestibular system responds to acceleration of the head in space
and therefore always signals self-motion. Visual flow, on the
other hand, is ambiguous in that it signals object motion and
eye motion, as well as self-motion. &
cally and at a low level. However, there is evidence,
at least for the visual channel, that under certain cir-
cumstances we are able to suppress the inappropriate
balance response. Guerraz et al. (2001), in a visual
perturbation study, used a motor to move a false wall
sideways, to the left or right. When the wall was moved
unpredictably, it evoked a whole-body response in the
same direction. When the subject controlled the direc-
tion and timing of the wall movement with a joystick, the
response was strongly suppressed. In separate experi-
ments they showed that for response suppression to
occur it was not necessary for the subject to initiate the
wall movement, provided they were given some other
knowledge of when the wall would move; neither was it
necessary that subjects knew the direction of wall move-
ment. Response suppression appeared to result simply
from the explicit prior knowledge that an upcoming
sensory perturbation is likely to be caused by an external
agent rather than by a real, unplanned body movement.
These data suggest that cognitive processes interact
with, and are able to suppress, the visual channel of
balance control. In the present experiments, we investi-
gate whether this is a principle that generalizes to the
vestibular channel.
To study the vestibular analogue of visual-response
suppression we have used the technique of galvanic
vestibular stimulation (GVS) to evoke a whole-body
response in standing subjects. The response to GVS is
not unlike that evoked by the moving wall stimulus used
Journal of Cognitive Neuroscience 17:3, pp. 463–469
by Guerraz et al. (2001). It, too, consists of a laterally
directed body motion to the left or right depending on
the polarity of the stimulus. The body leans and bends in
a direction toward the anodal ear (Day et al., 1997) with
a time course and amplitude similar to the visual-evoked
response (Guerraz et al., 2001). However, there is an
important difference between the two sensory channels.
For the visual channel, it is a simple matter to convey to
the subject the direction of an impending visual per-
turbation. One simply informs the subject that the wall
will move to the right or left. For the vestibular chan-
nel, it is not possible to give directional information by
externalizing the stimulus in the same way because the
vestibular system, unlike the visual system, responds ex-
clusively to motion of the head in space and not to
external phenomena. The direction of an impending
vestibular perturbation could be indicated by referring
to its perceptual effects. For instance, one could tell the
subject that the stimulus will make him feel that his
body is tilting, say, in a particular direction. However,
the perceptual effects of GVS can be quite subtle and for
a given polarity can range from an apparent movement
in one direction to one diametrically opposite, depend-
ing on both the circumstances and subject (Wardman,
Taylor, & Fitzpatrick, 2003). We therefore did not give
any directional information to our subjects in the pres-
ent experiments. However, this should not interfere
with our experimental aim because, for the visual chan-
nel, directional information is not essential for suppres-
sion to occur. Knowledge of stimulus timing is sufficient.
Part of these data has been communicated to the Phys-
iological Society (Guerraz & Day, 2001).
RESULTS
Experiment 1
The postural response to GVS (0.5 mA for 3 sec) was
similar to that described previously (Day et al., 1997).
During stimulation the body became bent and tilted
towards the side of the anode. Thus, when averaged
across all trials in which GVS was applied, the steady-
state tilt-in-space was significantly, F(2,10) = 25.5, p <
.01, greater for the head than for the trunk, mean (SD)
1.31 (0.42)8 and 1.07 (0.41)8, respectively, both being of
greater amplitude than for the pelvis, 0.71 (0.34)8.
The mean time courses of the responses are shown in
Figure 1. It illustrates, for the three conditions, the aver-
age lateral position of both the midline marker placed
over C7 and the center of foot pressure (COP). These
traces represent mean deviations in the direction of
the anode averaged across both polarities of stimula-
tion (see Methods). For all conditions, the galvanic stim-
ulus induced a displacement of the top of the body
(C7) towards the anode, with a latency of 250 msec
and a peak amplitude of 25 mm (see Figure 1). The

Figure 1. Grand mean lateral

translation of both C7 (top)
and COP (bottom) to
unpredictable, predictable,
and self-triggered GVS in
Experiment 1. Period of
stimulation is shown by the
thickened portion of the time
axis. Upward def lections
indicate translations toward
the anode (either right or left).
On the right of the figure is
depicted the effect of galvanic
stimulation, that is, an overall
tilt of the body with greater
tilts in space recorded from
higher body segments.

464 Journal of Cognitive Neuroscience Volume 17, Number 3


Figure 2. Setup of Experiment 2. The two factors, visual conditions
(top) and stimulus type (bottom) are represented. The six IREDs
placed on the subject are shown.
behavior of the visual channel when tested under similar
conditions (Guerraz et al., 2001). However, this could be
explained by one crucial difference between the experi-
ments. In the visual experiment, subjects could have
suppressed the visual information and relied upon their
remaining two channels (vestibular and somatosensory)
of self-motion information. In the present Experiment 1,
if subjects suppressed vestibular information they only
had a single channel (somatosensory) remaining be-
cause vision was occluded. This restriction of nonves-
tibular information may have altered the predictability
effect. Therefore, in Experiment 2 we included a condi-
tion in which subjects had access to visual information
(see Figure 2) while they were stimulated with GVS.
Figure 3 shows the time course of the mean lateral
displacement at the level of C7 and the COP for the
different experimental conditions. The GVS-evoked re-
sponse was reduced by approximately 40% when vision
was available in comparison with complete darkness. At
the level of C7, the availability of visual information had
a significant effect on both the slope of the dynamic
response, vision, F(1,11) = 17.3, p < .01, and the steady-
state response, vision, F(1,11) = 17.8, p < .01. In con-
trast, the response was not affected by whether the
stimulus was unpredictable or self-triggered: stimulus
type, dynamic response: F(1,11) = 0.17, p > .05; steady-

response took approximately 1.5 sec to build up after

which the body position was more or less maintained
while the stimulus was held on. This response ceased
shortly after the offset of stimulation (250 msec) as the
body returned towards its initial baseline position. The
COP showed a similar time course except for an oppo-
sitely directed transient 200 msec after stimulus onset/
offset as the muscles actively pushed the body toward/
away from the anode. The whole-body center of mass
(COM) must lie vertically above the COP position when
the body is stationary. Therefore, the mean steady-state
COP displacement of 16.2 mm represents a good esti-
mate of COM displacement.
The factor stimulus type (unpredictable, predictable,
or self-trigger) had no effect on any of the dynamic or
steady-state response measures. Thus, the slope of the
initial lateral response did not differ for the marker at
C7, F(2,10) = 0.3, p > .05, or for the COP, F(2,10) = 0.6,
p > .05. The steady-state displacement was similarly
unaffected: C7, F(2,10) = 0.17, p > .05; COP, F(2,10) =
0.2, p > .05. Comparable results were obtained from
measures of body segment tilts.

Experiment 2 Figure 3. Grand mean lateral translation of both C7 (top) and

The results of Experiment 1 showed that the spatiotem-
poral characteristics of the GVS-evoked postural re-
sponse were not affected by the predictability of the
forthcoming event. This is in marked contrast to the
COP (bottom) to unpredictable and self-triggered GVS in two different
visual conditions (vision/darkness) in Experiment 2. Period of
stimulation is shown by the thickened portion of the time axis.
Upward def lections indicate translations towards the anode (either
right or left).

Guerraz and Day 465

state response: F(1,11) = 0.32, p > .05. Neither was
there a significant interaction between the two factors:
Stimulus Type  Vision, dynamic response: F(1,11) =
0.63, p > .05; steady-state response: F(1,11) = 0.06, p >
.05. Similar results were obtained from statistical anal-
ysis of body-segment tilts and COP displacement.
To investigate the possibility of a learning effect we
divided the above data into two blocks consisting of
the average responses of the initial and the final 50% of
trials for each subject per condition. There was no sig-
nificant main effect of learning for either the dynamic
response, Block1: 9.1 mm/sec; Block2: 10.2 mm/sec;
block: F(1,11) = 2.19, p > .05, or for the static response,
Block1: 11.9 mm; Block2: 13.3 mm; block: F(1,11) =
1.16, p > .05. In addition, for both of these measures
there were no significant interactions between the block
factor and the vision or stimulus type factors.
DISCUSSION
Here we have shown that the vestibular channel of
balance control differs fundamentally from the visual
channel. Whereas the response to a visual perturbation
is suppressed if the subject is aware that an upcoming
disturbance is likely to be caused by an external agent
rather than self-motion (Guerraz et al., 2001), the same
is not true of the vestibular channel. Expectation of a
vestibular perturbation, either through a voluntary ac-
tion that initiates it (self-trigger condition) or through
prior knowledge of the event and timing cues (predict-
able condition), had no consistent effect on the evoked
whole-body response. Suppression did not occur al-
though subjects knew when the stimulus would happen,
that it was artificial, and that it had the potential to make
them sway sideways. For the first experiment, because
vision was occluded, suppression of the vestibular chan-
nel would have left the subject with only somatosensory
information with which to balance the body. Such a
large reduction in total sensory information for balance
control may have inhibited the action of a vestibular-
suppression mechanism. However, the same results
were obtained in the second experiment when vision
was made available. This experimental situation is di-
rectly comparable to that in which visual suppression is
known to occur (Guerraz et al., 2001). Our data there-
fore suggest that the vestibular channel of balance
control is not susceptible to cognitive modulation in the
same way as the visual channel.
Despite its apparent automaticity, the GVS-evoked
response involves complex processing and is readily
modifiable by many different factors. These include head
and trunk position (Pastor, Day, & Marsden, 1993; Lund
& Broberg, 1983), standing posture (Marsden, Blakey, &
Day, 2003; Day et al., 1997), support surface properties
(Fitzpatrick et al., 1994), and the availability of non-
vestibular sensory information (Day & Cole, 2002; Day,
466 Journal of Cognitive Neuroscience
Guerraz, & Cole, 2002; Horak & Hlavacka, 2001; Britton
et al., 1993; Njiokiktjien & Folkerts, 1971). Imaging
studies have shown that GVS is capable of activating
cortical areas (Lobel, Kleine, LeBihan, Leroy-Willig, &
Berthoz, 1998), and projections from cortex to brain-
stem nuclei (Wilson et al., 1999; Kably & Drew 1998)
provide possible routes for higher centers to exert
cognitive influence on vestibulospinal pathways. Why
then should vestibular responses differ from visual re-
sponses and not be modifiable by cognitive input? We
suspect the reason for this difference arises from the
natural ability of the two channels to signal self-motion.
Detection of self-motion is of primary importance for
the balance control system. Both the vestibular and
visual systems have the potential for signaling self-
motion but with an important difference. The vestibu-
lar system detects unambiguously the acceleration of
the head in space and therefore always signals self-
motion. The visual channel is different in that visual
flow carries information regarding both self-motion and
object motion; a displacement of either the body or the
visual scene can yield similar patterns of retinal motion
stimuli. A postural response would be appropriate for
the former situation but not the latter. Similarly, postural
responses to the visual flow arising from eye movements
would be inappropriate. In this regard, White, Post, and
Leibowitz (1980) showed that visual flow accompanying
saccades have little or no influence on body sway,
whereas similar externally produced visual flow induces
a systematic postural adjustment.
Balance responses that are evoked by artificial means,
such as those studied here, usually are inappropriate
and interfere with stability. Cognitive suppression of
these would therefore be a desirable feature of the bal-
ance system. The same reasoning does not apply to real
disturbances of the body, which evoke functional re-
sponses that help preserve stability. For example, sud-
den displacements of the support surface on which
subjects stand evoke fast muscle responses with a
latency between the simple segmental reflex and voli-
tional latencies (Horak, Diener, & Nashner, 1989). These
early, automatic muscle responses, driven in part by
somatosensory inputs, represent a first line of defense
to keep the body upright and are functionally impor-
tant. They too seem relatively immune to cognitive
modulation. Thus, precuing subjects with partial or com-
plete information about the forthcoming support sur-
face disturbance has either no effect (Diener, Horak,
Guschlbauer, & Dichgans, 1991) or only subtle effects
(Maki & Whitelaw, 1993) on the automatic responses.
However, it is unclear whether this lack of modulation
is due to the functional nature of the response or due to
an absence of cognitive input.
Responses to real disturbances of the vestibular sys-
tem produced by movement of the head in space are
similarly unaffected by knowledge of the upcoming
disturbance. This can be observed at a single neuron
Volume 17, Number 3
level by recording from cells in the vestibular nuclei. Roy
and Cullen (2001) showed that a class of vestibular
neurons (VO neurons) in the monkey responds vigor-
ously to passive rotation of the head in space. Interest-
ingly, when the monkey made similar but active head
movements, by turning the head on the trunk, the
responses of these neurons were suppressed. However,
this suppression is unlikely to be due to cognitive
influence arising from knowledge of perturbation tim-
ing. Thus, the neuronal responses were not suppressed
when the monkey ‘‘drove’’ the perturbation by using a
steering wheel or a button push to control the motor
that turned the head in space. As Roy and Cullen con-
cluded, the suppression is more likely to be due to a
mechanism that involves an efferent copy of the motor
command to the neck musculature.
Although we have been unable to find modulation of
vestibular-evoked responses based on internal knowl-
edge of stimulus timing, we do not rule out other types
of cognitive modulation, for instance, by changes in
cognitive load. This can be manipulated by asking sub-
jects to do a secondary task (dual task paradigm) such as
mental arithmetic (Lestienne et al., 1977) or reaction
time tasks (Vuillerme, Nougier, & Teasdale, 2000). How-
ever, the effects of such tasks on balance control are
rather inconsistent. Indeed, some authors report no dif-
ference in postural sway (Yardley et al., 2001), whereas
others report either a small improvement (Hunter &
Hoffman, 2001; Vuillerme et al., 2000) or a decrement
of postural stability (Pellecchia, 2003). Similarly, when
stance is challenged by sensorial disturbances such as a
disturbing visual flow, authors report either an increase
of evoked responses (Lestienne et al., 1977) or no effect
(Ehrenfried, Guerraz, Thilo, Yardley, & Gresty, 2003).
These results would suggest that the effect of cognitive
load on balance control, if any, is rather modest.
In conclusion, we have shown a fundamental differ-
ence between the vestibular and visual channels of
balance control in terms of cognitive modulation
through knowledge of the source of the perturbation
and its timing. We suggest the basis of this difference
arises because the origin of visual flow information is
unavoidably ambiguous. This has allowed a process to
evolve that is able to suppress the balance response to a
visual perturbation if the perturbation is deemed likely
to result from object motion or eye motion, rather than
self-motion. In contrast, there is no equivalent require-
ment for the vestibular channel and therefore no neces-
sity for such cognitive control.
METHODS
Subjects
Six (age range, 23–33 years) and 12 (age range, 23–49
years; mean [SD]: 33.3 [8.4] years) healthy subjects
consented to participate in Experiments 1 and 2, respec-
tively, according to guidelines of the local ethics com-
mittee and to the declaration of Helsinki. All subjects
had no history of vestibular, orthopedic, or neuromus-
cular disease.
Galvanic Stimulation
A constant-current custom-built stimulator was used to
produce the galvanic vestibular stimulus. GVS was ap-
plied to the subjects via 3-cm-diameter electrodes placed
over the mastoids. GVS consisted of a rectangular cur-
rent profile with a constant intensity of 0.5 mA lasting for
3 sec. This intensity gives rise to a measurable and
reproducible body sway response (Day et al., 1997)
but produces little or no cutaneous sensation. The po-
larity of the stimulus (anode right or left) was randomly
switched between trials.
Postural Recordings
Subjects stood on a force plate (type 9281B, Kistler
Instrumente AG, CH-8408 Winterthur, Switzerland),
which registered ground reaction force in three dimen-
sions. From these data, the position of the ground
reaction force vector or the COP was calculated. The
position and tilt of each body segment (head, torso, and
pelvis) was measured in three dimensions using an
optoelectronic motion analysis system (Selspot II). As
shown in Figure 2, two markers (25 cm apart) were
mounted on a rigid helmet worn by the subject to
measure head displacement; two were stuck onto the
skin overlying the spine, one at the level of C7 and the
other at the lumbar level, and two were mounted 35 cm
apart on a frame clamped to a belt around the iliac crests
to measure the displacement of the pelvis. The position
and force data were collected with a sampling frequency
of 100 Hz.
Procedure
Experiment 1: Absence of Visual Information
Subjects were instructed to stand still and relaxed with
the head facing forward and feet placed together. A
lightweight push-button box was held by both hands at
about waist level throughout the experiment. During
trials, vision was occluded using liquid-crystal spectacles
(PLATO visual occlusion spectacles, Translucent Tech-
nologies, Toronto, Ontario, Canada).
The experiment involved three conditions.
1. Unpredictable condition: GVS was triggered by
the experimenter at a random time. Before each trial,
subjects were naive as to whether or not and when they
would be stimulated.
2. Predictable condition: Subjects were verbally
instructed before the trial that they were going to be
Guerraz and Day 467
stimulated 500 msec after they heard a tone. The volume
of the tone was set at a comfortable hearing level.
3. Self-trigger condition: Subjects were instructed
before the trial that they should trigger GVS them-
selves using the push button, but without knowing the
polarity of the stimulus. Subjects were aware that a
stimulus would be delivered as soon as they pressed
the button. They were instructed that they would have
to trigger the stimulus within 3 sec of hearing a tone
but otherwise were free to choose when to push the
button.
Before data recording, two trials of each condition
were given to the subject while seated with eyes open in
a room with normal illumination. This familiarized them
with the stimulus and allowed them to practice the
procedure. Each of the three experimental conditions
consisted of 16 trials, 8 with each polarity of stimulation.
In addition, 16 control trials without any stimulation
were also recorded. The total of 64 trials was presented
in a pseudorandom order, with each trial lasting for
17 sec. Subjects had a rest period of 10 min after the first
32 trials.
Experiment 2: Vision versus Darkness
The purpose of Experiment 2 was to investigate whether
self-triggering GVS would modulate the postural re-
sponse when subjects had visual information available
regarding self-motion. The visual scene provided in the
visual condition was a 2-D square grid (25 by 25 cm),
consisting of 121 low-intensity, colored light-emitting
diodes suspended on fine wires in a grid. The grid was
placed vertically with its center at eye height and 50 cm
in front of the subject in an otherwise blacked-out room.
This static visual display was previously used by Day et al.
(2002) and was shown not to suppress the GVS-evoked
response but to reduce it notably (40%). The experi-
mental setup is depicted in Figure 2. In the darkness
condition, the room was blacked out.
The experimental procedure was similar to that used
in Experiment 1 except the predictable condition was
omitted. Each trial lasted for 17 sec during which time
the subject was either stimulated by himself (self-trigger,
n = 24), stimulated by the experimenter at a random
time (unpredictable, n = 24), or not stimulated (control
trials, n = 24). For each trial type, vision was provided in
half the trials and the room was blacked out in the
remaining half. In stimulation trials, the polarity was
varied with the anode either on the left (n = 12) or
right (n = 12). The stimulus type, stimulus polarity, and
visual environment were presented in pseudorandom
order. The 72 trials were divided into three blocks of 24
trials. Subjects had a rest period of 10 min after each
block. Before data recording, two trials of each condi-
tion were given to the subject while seated with eyes
open in a room with normal illumination.
468 Journal of Cognitive Neuroscience
Data Analysis
For each subject, the position and force signals were
averaged for each condition. The time of stimulus onset
was used to align the recordings from each trial. Re-
sponses to the two polarities of stimulation were oppo-
sitely directed but otherwise symmetrical. They were
therefore averaged after inversion of the responses
obtained with the left anode.
From the individual average position and forces under
each condition, two parameters were computed.
1. Dynamic response: This was measured from the
slope of the mean initial response. The slope was based
on a least squares linear regression fitted over a 1-sec
period covering 100 data points from 0.5 to 1.5 sec after
the onset of galvanic stimulation. Within this 1-sec
period, the profile of the initial evoked response was
approximately linear, as suggested by the strong cor-
relation coefficient (R2 > .90) observed for each con-
dition and subject.
2. Steady-state response: This was calculated from
the difference between the mean prestimulus level
(measured over a 4-sec period) and that achieved over a
1.5-sec period from 1.5 to 3 sec after the onset of the
stimulus. This latter period was chosen to avoid the
initial dynamic response.
The variables measured were subjected to statistical
analysis using one-factor (stimulus type) or two-factor
(stimulus type and vision) repeated measures analysis of
variance (ANOVA) for Experiment 1 and 2, respectively.
An additional three-factor (stimulus type, vision, and
block) analysis was performed on the data of Experi-
ment 2 to investigate learning effects. A 0.05 significance
level was adopted throughout.
Acknowledgments
The work was funded by the Medical Research Council of
Great Britain. We are indebted to the late Mr. Richard
Bedlington for expert technical assistance.
Reprint requests should be sent to B. L. Day, MRC Human
Movement Group, Sobell Department of Motor Neuroscience
and Movement Disorders, Institute of Neurology, University
College London, Queen Square, London WC1N 3BG, UK, or via
e-mail: b.day@ion.ucl.ac.uk.
REFERENCES
Britton, T. C., Day, B. L., Brown, P., Rothwell, J. C., Thompson,
P. D., & Marsden, C. D. (1993). Postural electromyographic
responses in the arm and leg following galvanic vestibular
stimulation in man. Experimental Brain Research, 94,
143–151.
Bronstein, A. M., & Buckwell, D. (1997). Automatic control of
postural sway by visual motion parallax. Experimental Brain
Research, 113, 243–348.
Day, B. L., & Cole, J. (2002). Vestibular-evoked postural
responses in the absence of somatosensory information.
Brain, 125, 2081–2088.
Volume 17, Number 3
Day, B. L., Guerraz, M., & Cole, J. (2002). Sensory interactions
for human balance control revealed by galvanic vestibular
stimulation. Advances in Experimental Medical Biology,
508, 129–137.
Day, B. L., Séverac Cauquil, A., Bartolomei, L., Pastor, M. A.,
& Lyon, I. N. (1997). Human body-segment tilts induced
by galvanic stimulation: A vestibularly driven balance
protection mechanism. Journal of Physiology (London),
500, 661–672.
Diener, H. C., Horak, F., Guschlbauer, B., & Dichgans, J.
(1991). Direction and amplitude precuing has no effect on
automatic posture responses. Experimental Brain
Research, 84, 219–223.
Ehrenfried, T., Guerraz, M., Thilo, K. V., Yardley, L.,
& Gresty, M. A. (2003). Posture and concurrent mental task
performance of subjects confronted by moving visual fields.
Cognitive Brain Research, 17, 140–153.
Fitzpatrick, R., Burke, D., & Gandevia, S. C. (1994).
Task-dependent reflex responses and movement
illusions evoked by galvanic vestibular stimulation in
standing humans. Journal of Physiology (London), 478,
363–372.
Guerraz, M., & Day, B. L. (2001). Human body response to
galvanic vestibular stimulation is not affected when the
stimulus is self-triggered. Journal of Physiology (London),
521, 142.
Guerraz, M., Shallo-Hoffmann, J., Yarrow, K., Thilo, K. V.,
Bronstein, A. M., & Gresty, M. A. (2000). Visual control of
postural orientation and equilibrium in congenital
nystagmus. Investigative Ophthalmology & Visual Science,
41, 3798–3804.
Guerraz, M., Thilo, K. V., Bronstein, A. M., & Gresty, M. A.
(2001). Influence of action and expectation on visual control
of posture. Cognitive Brain Research, 11, 259–266.
Horak, F. B., Diener H. C., & Nashner, L. M. (1989). Influence
of central set on human postural responses. Journal of
Neurophysiology, 62, 841–853.
Horak, F. B., & Hlavacka, F. (2001). Somatosensory loss
increases vestibulospinal sensitivity. Journal of
Neurophysiology, 86, 575–585.
Hunter, M. C., & Hoffman, M. A. (2001). Postural control:
Visual and cognitive manipulations. Gait and Posture, 13,
41– 48.
Kably, B., & Drew, T. (1998). Corticoreticular pathways in the
cat I projection patterns and collateralisation. Journal of
Neurophysiology, 80, 389 – 405.
Lee, D. N., & Lishman, J. R. (1975). Visual proprioceptive
control of stance. Journal of Human Movement Studies, 1,
87–95.
Lestienne, F., Soechting, J., & Berthoz, A. (1977). Postural
readjustments induced by linear motion of visual scenes.
Experimental Brain Research, 27, 363–384.
Lobel, E., Kleine, J. F., LeBihan, D., Leroy-Willig, A., & Berthoz,
A. (1998). Functional MRI of galvanic vestibular stimulation.
Journal of Neurophysiology, 80, 2699–2709.
Lund, S., & Broberg, C. (1983). Effects of different head
positions on postural sway in man induced by a reproducible
vestibular error signal. Acta Physiologica Scandinavica,
117, 307–309.
Maki, B. E., & Whitelaw, R. S. (1993). Influence of expectation
and arousal on center of pressure responses to transient
postural perturbations. Journal of Vestibular Research, 3,
25–39.
Marsden, J. F., Blakey, G. & Day, B. L. (2003). Modulation of
human vestibular-evoked postural responses by alterations
in load. Journal of Physiology (London), 548, 949–953.
Nashner, L. M., & Wolfson, P. (1974). Influence of head
position and proprioceptive cues on short latency postural
reflexes evoked by galvanic stimulation of the human
labyrinth. Brain Research, 67, 255–268.
Njiokiktjien, C., & Folkerts, J. F. (1971). Displacement of the
body’s centre of gravity at galvanic stimulation of the
labyrinth. Confinia Neurology, 33, 46–54.
Pastor, M. A., Day, B. L., & Marsden, C. D. (1993). Vestibular
induced postural responses in Parkinson’s disease. Brain,
116, 1177–1190.
Pellecchia, G. L. (2003). Postural sway increases with
attentional demands of concurrent cognitive task.
Gait & Posture, 18, 29–34.
Roy, J. E., & Cullen, K. E. (2001). Selective processing of
vestibular reafference during self-generated head motion.
Journal of Neurosciences, 21, 2131–2142.
Vuillerme, N., Nougier, V., & Teasdale, N. (2000). Effects of a
reaction time task on postural control in humans.
Neuroscience Letters, 291, 77– 80.
Wardman, D. L., Taylor, J. L., & Fitzpatrick, R. C. (2003). Effects
of galvanic vestibular stimulation on human posture and
perception while standing. Journal of Physiology (London),
551.3, 1033–1042.
White, K. D., Post, R. B., & Leibowitz, H. W. (1980). Saccadic
eye movements and body sway. Science, 208, 21–23.
Wilson, V. J., Zarzecki, P., Schor, R. H., Isu, N., Rose, P. K., Sato,
H., Thomson, D. B., & Umezaki, T. (1999). Cortical
influences on the vestibular nuclei of the cat. Experimental
Brain Research, 125, 1–13.
Yardley, L., Gardner, M., Bronstein, A. M., Davies, R.,
Buckwell, D., & Luxon, L. (2001). Interference between
postural control and mental task performance in patients
with vestibular disorder and healthy controls. Journal of
Neurology, Neurosurgery and Psychiatry, 71, 48–52.
Guerraz and Day 469