Neurophysiologie Clinique/Clinical Neurophysiology (2008) 38, 391398

REVIEW/MISE AU POINT

Ocular versus extraocular control of posture and

equilibrium
Contrle oculaire versus extraoculaire de la posture
et de lquilibre
M. Guerraz a,, A.M. Bronstein b

a
CNRS UMR 5105, laboratoire de psychologie et neurocognition, universit de Savoie, 73376 Le-Bourget-du-lac, France
b
Academic Department of Neuro-otology, Division of Neuroscience and Mental Health, Imperial College London,
Charing Cross Hospital, Fulham Palace Road, London W6 8RF, UK

Received 17 September 2008; accepted 17 September 2008

Available online 16 October 2008

KEYWORDS Summary Vision has been shown for almost a century to be heavily involved in postural con-
Vision; trol. However, the mechanism by which it operates is still an open debate. The purpose of this
Postural control; manuscript is to review the evidence supporting the view that there are two modes of visual
Motion parallax; detection of body sway: ocular and extraocular. The former is based on the characteristics of
Efference copy the visual ow (retinal slip), the second one is based on either the copy of the motor command
(efference copy) or the extraocular muscle afferents (re-afferences) consecutive to eye move-
ments. Results from the literature indicate that these two modes of visual detection of body
sway are effective and can operate congruently. For sufciently large body sway with respect
to eyetarget distance, the ocular and the extraocular perception systems could provide two
sources of visual information about body displacements. However, the afferent system might
remain the only one used for small lateral body sway.
2008 Elsevier Masson SAS. All rights reserved.

Rsum Le rle de la vision dans la rgulation posturale a t mis en vidence depuis plus dun
MOTS CLS sicle. Toutefois, les mcanismes par lesquels elle opre restent lobjet de dbats. Le but de ce
Vision ; manuscrit est de faire le point sur les travaux qui supportent lide selon laquelle la dtection
Contrle postural ; visuelle des oscillations corporelles pourrait se faire selon deux modes : partir dinformations
Parallaxe de soit purement oculaires, soit extraoculaires. La premire est base sur les caractristiques
mouvement ; du ux visuel (glissement rtinien), alors que la seconde est base soit sur la copie de la
Copie effrente commande motrice oculaire (copie effrente), soit sur les affrences musculaires conscutives
aux dplacements de lil. Les rsultats de la littrature font apparatre que ces deux modes
de dtection sont effectifs et peuvent oprer de concert dans la rgulation posturale. Lorsque

Corresponding author.
E-mail address: Michel.guerraz@univ-savoie.fr (M. Guerraz).

0987-7053/$ see front matter 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.neucli.2008.09.007
392 M. Guerraz, A.M. Bronstein

les oscillations posturales sont sufsamment importantes relativement la distance ilcible,

les systmes perceptifs oculaires et extraoculaires savrent tre deux sources information-
nelles pertinentes permettant au systme dtre inform sur les oscillations corporelles. Seul
le systme affrent, oculaire, peut encore vhiculer des informations pertinentes lorsque les
oscillations sont de faible amplitude.
2008 Elsevier Masson SAS. All rights reserved.

Introduction Further insight into the importance of visual ow on

A person moving or standing upright can usually sense
how he is moving or being moved relative to his environ-
ment. Detection of changes that accompany body movement
is essential for controlling body equilibrium as well as
locomotion. There are mechanical changes within and at
the surface of the body (somatosensory proprioceptive
information), head motion information (vestibular input),
and optical changes at the eye (optic ow pattern; visual
information). Somatosensory afferents include cutaneous,
musculo-articular, and muscular receptors. Since these
receptors are distributed throughout the whole body, these
are critical for determining body conguration, i.e. the
relation of each segment relative to the adjacent ones.
In contrast, vestibular and visual receptors are located on
the head, which can move independently of the trunk.
Therefore, in order to contribute to body conguration or
orientation, vestibular and visual signals require the inte-
gration of the position of the head on the trunk and also,
specically for visual signals, the integration of the position
of the eye in the head [80].
Vision is not necessary for normal balance one can stand
in the dark. However, from as early as the studies of Travis
[74], it appeared that body stability was improved by vision,
a situation known to clinicians since the 19th century [49].
Deformation of the retinal image due to ego-motion or trans-
position of objects in the environment (called optic ow)
is therefore an important source of information concerning
the world and the relation of the subject into this world
[33].
Simple ways to understand the nature of the visual
basis of postural control consist in modulating the physical
visual parameters available and then examine the resultant
spontaneous body sway. The most popular protocol is the
Romberg test [9,26]. It involves the comparison of an indi-
viduals spontaneous sway with eyes closed and eyes open.
Under these conditions, a reduction of body oscillations with
eyes open can be observed in most people, across the differ-
ent frequency components of body sway [26]. The degree of
postural improvement with eyes open is variable and, among
other factors, depends on :
the stance width [19];
the support surface [10,54];
the availability of other sensory inputs.
For instance, when either the vestibular [8,41,59,62] or
the somatosensory [47,55,58] system is impaired, giving rise
to a signicant postural instability (or even falls), then visual
stabilization becomes particularly effective.
postural control has been gained by manipulating articial
optical ow in order to evoke postural disturbance. The
most popular paradigm refers to the moving room paradigm.
In their pioneer work, Lee and Lishman [50] showed ele-
gantly that vision was not purely an exteroceptive sense, but
was an integral component of the control system for main-
taining stance. In the moving room paradigm, the subject
assumes a quiet upright stance within a visual environment
(room) that can move relative to the xed frame of refer-
ence (oor) on which subjects are standing. The movement
of the room creates an optical ow eld pattern that is sim-
ilar to the pattern experienced when translating back and
forth through the environment. The result of such exper-
iments showed directionally specic postural response: in
response to linear forward and backward motion of the
whole visual environment, subjects sway in the direction of
motion [15,16,50,69,70]. This response is usually considered
to arise from a consequence of a misinterpretation of the
visual ow as due to self-motion instead of object-motion.
A wide variety of moving visual stimuli has been employed to
investigate the phenomenon of visually-induced body sway.
These include:
tilting rooms [10];
projected displays simulating a moving visual wall, tun-
nel, oor or ceiling [29,31,34,52,75];
visual roll rotations [17,25,28,76].
Consistently, it has been reported that in rhythmic motion
of the visual scene (sinusoidal motion) with either a lin-
ear or a rotational stimulus, the largest effect of vision is
observed at fairly low frequencies (0.2 Hz) at which the
body is phase locked with the moving scene. At higher fre-
quencies (>0.3 Hz), this locking is lost and the overall effect
of visual motion is reduced [29,35,75]. The amplitude of
postural sway, at least in the antero-posterior direction, has
been shown to be particularly affected by visual properties
such as the density of texture [52,75] and the velocity of
motion [52,53].
Spontaneous body oscillations cause the image of the
visual environment to move on the retina. It has gener-
ally been considered that this retinal slip, which is directly
related to subjects displacements, is used by the CNS as
feedback for compensatory sway. Although many experi-
mental results are in agreement with such hypothesis, it
must be noted that compared to complete darkness, spon-
taneous lateral body oscillations are largely reduced when
standing subjects xate a small LED in an otherwise dark-
ened environment [56]. In these circumstances, the eyes
track the LED while the head is moving sideways. If we
Ocular versus extraocular control of posture and equilibrium
assume that the tracking is correct during self-motion, lit-
tle or no retinal shift of the xated target would occur
on the retina. Since spontaneous body sway is reduced in
this minimal visual condition, we must assume that head
displacement can be sensed on the basis of the amplitude
of the eye movement (extraretinal motion perception) and
not retinal slip. These results have led to the idea that
two different mechanisms subserve visual stabilization of
posture [38,45,57,65]: afferent and efferent motion per-
ception. The former is based on the characteristics of the
visual ow (retinal slip), the second one is based on either
the copy of the motor command (efference copy) or the
extraocular muscle afferents (re-afferences) consecutive to
eye movements. In order not to infer about the underlying
neurophysiological mechanisms (see under for a discussion),
the so-called efferent motion perception will be renamed
in the present manuscript as extraocular motion percep-
tion.
Extraocular motion perception and posture
The processing of visual information during eye movements
has been largely investigated in the context of percep-
tion or world constancy. The central question is how the
world is maintained stable despite eye movements. Two not
mutually exclusive hypotheses seek to explain the origin of
these extraocular information (see [78] for a review): the
inow and outow hypothesis. The inow theory saccadic eye movements induces some reduced amplitude
holds that proprioceptive receptors (e.g. muscle spindles)
of extraocular muscles provide the necessary information
about the position and displacement of the eyeballs in the
orbits. The main limitation originating from this theory is
that extraocular afferents (inow) cannot be transduced
until the eyes begin to move. The outow theory pro-
poses that neural outow or a branch of the neural outow
such as a corollary discharge [67] or an efference et al. [45] recorded body displacements in such patients in
copy [77] is the information used by the CNS to main-
tain visual consistency. The particular interest of this theory
is that the sensory consequences of the action (visual ow
due to the displacement of the eyes) can be anticipated
in a feed-forward manner. The most notorious results in
favor of the outlfow hypothesis stems from the work of
Stevens et al. [68]. These authors observed that, in subjects
whose eyes were paralyzed with curare, at the time of an
attempted saccade, the world seemed to be in a different
position when sustained eccentric gaze was unsuccessfully
attempted.
If extraretinal cues are critical for stabilisation of pos-
ture, the distance of the visual target appears as an
important parameter. Indeed, when a subject is moving side-
ways while tracking an earth-xed LED, the displacement of
the eyeballs in their orbit (angle ) is directly related to the
distance between the eyes and the target. The value of this
angle is a trigonometric function of eyetarget distance:
= arctan (d/x), where d represents lateral head movement
and x represents eyetarget distance. As shown by several
authors, xation of a tiny light spot (LED) in an otherwise
completely dark surround stabilises posture up to a limit
of 11.5 m [11,56,57]. Then, as can be seen in Fig. 1, the
amplitude threshold for effective extraocular postural sta-
bilisation is an angle of approximately 0.3 for 1 cm head
393
Figure 1 Geometrical relationship between the eye and a
xation target during lateral head sway over a distance d
(for these plots d = l cm). The amplitude of eye displacement
(angle ) is plotted as a function of eye-target distance. Adapted
from Guerraz et al. [38].
sway amplitude that is in the physiological range of postural
sway in the lateral direction.
Recently, Glasauer et al. [36] showed that pursuing a
LED in darkness was accompanied with body oscillations in
phase with the horizontal displacement of the eyes in their
orbit. In contrast, when subjects were required to pursue
that LED over a stable background, that generates an impor-
tant visual ow on the retina, they were much more stable.
Rougier et Garin [64] also reported recently that performing
of body displacements. Ocular motor disorders also pro-
vide valuable arguments in favour of extraocular postural
control. Patients with acute unilateral partial labyrinthine
failure due to vestibular neuritis show vestibulo-ocular
(spontaneous nystagmus) as well as vestibulo-spinal function
disorders (postural imbalance). In order to investigate the
relationship between these two vestibular functions, Jahn
complete darkness when either the nystagmus was present
or transiently suppressed. Suppression (reduction of 70%
of mean peak slow phase velocity) was achieved by ask-
ing subjects to xate a head-xed bright LED (presented
via a vido-oculographie [VOG] mask) providing no visual
afferent cue. When standing on foam in complete darkness,
patients showed excessive postural sway. Interestingly, sup-
pression of the nystagmus was accompanied by an important
reduction of postural sway.
Altogether, these results indicate that the extraocular
signal is a critical indicator of head or body motion in space,
which is largely considered by postural control mechanisms.
However, the nature of this extraocular information remains
open. In the next section, we will see that both extraocular
afferents (proprioception) and efferent signals likely play a
signicant role in postural control.
Role of the efference copy in postural control?
Jahn et al. [45] suggested that the important imbal-
ance observed in patients with acute unilateral partial
labyrinthine failure was likely the direct consequence
of their ocular motor disorders (spontaneous nystagmus).
Observations of subjects with congenital nystagmus (CN)
394
suggest, however, that spontaneous nystagmus per se
cannot be directly incriminated. CN is an ocular motor
disorder characterized by involuntary oscillatory eye move-
ments (46 Hz) that disrupt foveal target xation [21,37].
Despite almost constant eye movement and, therefore,
large amount of retinal image slip, subjects with CN rarely
complain of either oscillopsia (i.e. they retain spatial con-
stancy) or impairment of visuo-motor coordination [51]. In
agreement, they were repeatedly found:
not to sway more than normals in complete darkness;
to use vision quite efciently to stabilise their body
[23,30,39].
We showed for instance that evoked body response by
manipulation of the visual ow was of similar amplitude in
subjects with CN and controls [39]. DellOsso et al. [20]
suggested that the appearance of oscillopsia depends on
whether the site of eye oscillation was either within or out-
side of the efference copy loop. Such explanation could be
extended to visual control of posture in patients with either
acquired or CN, i.e. postural imbalance does not depend
on the nystagmus or the retinal slip per se but depends
on whether the retinal consequences of the nystagmus can
be anticipated and cancelled (CN) or not (acquired nys-
tagmus). This phenomenon is similar to that of saccadic
suppression and reminds us that the retinal slip (visual feed-
back) inherent to voluntary saccade in normal subjects is
not accompanied by postural destabilization [79]. In light
of these results, we can hypothesize that efferent copy sig-
nals about eyes movements (outow theory) are used for on illusory self-motion has been challenged, and several
postural stabilisation.
Role of the extraocular afferents in postural
control
Although the inow theory has been largely debated,
experimental data attests of the involvement of re-
aferrences from extraocular muscles in perceptual and
postural functions. By means of vibrators, Roll and Roll [63]
applied mechanical vibration to various points on the periph-
ery of the eyeballs in standing subjects with eyes closed.
Vibration of sufcient amplitude and frequency applied to a
muscle or tendon activates mainly the primary spindle end-
ings connected to the large Ia afferent bres, whose ring
rate seems to be interpreted by the CNS as an elongation of
that muscle. Application of such vibration to the extraoc-
ular muscles evoked in standing subjects induces whole
body shifts whose direction depended strictly on which eye
muscle was vibrated. These data conrm that extraocular
signals are critical for postural control and unambiguously
point at a contribution of extraocular proprioception in such
control.
Afferent motion perception
Although extraretinal cues appear to be critical, they
cannot embrace the whole contribution of vision on pos-
tural control. Indeed, xation of a single LED in darkness
reduces body sway less effectively than does xation of
the same target in an illuminated room for instance. All
M. Guerraz, A.M. Bronstein
in all, postural stability increases with the enrichment of
the visual environment [56]. Hence, many physiological
and physical parameters affect visual control of pos-
ture (see [13]: we note the object size and localization
[10,54,56], binocular disparity [32], visual motion [3] but
also visual acuity [56] and spatial frequency [48] to name
a few). Hence, the functionality of the retina and the
geometrical structure of the visual ow deserve to be
considered.
Functional specialization of the peripheral retina?
In spatial orientation, two visual systems, a focal one (cen-
tral vision) and an ambient one (peripheral vision), are
usually distinguished. The rst one, the central area of the
retina, is usually considered as being specialized in object
motion perception and object recognition. In contrast, the
peripheral region of the visual eld is particularly sensi-
tive to moving scenes, and is therefore thought to dominate
both perception of self-motion and postural control [24].
Such a differential involvement of the central versus the
peripheral retina in spatial orientation derives from research
on the illusory perception of self-motion (vection). Brandt
et al. [12] showed that circular vection, around the lon-
gitudinal body axis (in yaw) can be elicited with circular
visual stimuli (drum) but only if the diameter of the stim-
uli sustained more than 30 in diameter. Similar results
have been observed both with roll vection [44] and linear
vection [7]. However, this dominance of peripheral vision
studies have shown that vection could be elicited in cen-
tral vision [4,60]. Post [60] showed that the magnitude
of vection was greater with a larger stimulus area, but
differences disappeared when an equal area stimulus was
presented either centrally or in the periphery of the visual
eld.
The hypothesis of a dominance of the peripheral vision
on postural control has also received an important support
[27,52,2,22]. As for illusory self-motion perception, a cen-
tral eld 30 in diameter reduces spontaneous body lateral
and foreaft sway less effectively than does the periph-
eral eld (200 ) in which the central area is masked [56].
However, when the angular size of the visual eld is strictly
comparable in both the central and the peripheral part of
the retina (30 ), then the eccentric visual eld exhibits a
signicantly smaller contribution than does the central one.
Finally, when the peripheral eld size is corrected by the
cortical magnication factor of the retina in the primary
cortex (V1), spontaneous body sway is stabilized by the
peripheral retina to the same extent [71]. Similarly, postural
sway evoked by the displacement or rotation of a visual envi-
ronment is of smaller amplitude when the stimulus covers
only the central visual eld rather than the peripheral eld.
However, when the area is of equal size, visually-induced
body sway is more potent in the central eld than in the
periphery [69]. Therefore, the peripheral dominance often
reported in either visual stabilization of spontaneous body
sway or visually-induced body sway is more likely due to
the size of the stimulated eld manipulated than to a func-
tional specialization of the peripheral vision for postural
control.
Ocular versus extraocular control of posture and equilibrium 395

Structure of the visual ow

Lamellar versus radial visual ow

Visual ow generated during self-motion is not uniform
throughout the visual eld. When a subject is walking or
swaying back and forth in a 3-dimensional (3D) environment,
looking straight ahead, a radial ow having a focus of expan-
sion in the direction of motion is projected on the center
of the retina. At the peripheral edges of the eld of view,
the visual ow is nearly parallel to the line of motion. This
ow structure at the periphery has been termed lamellar
ow. The differential effect of radial versus lamellar ow
has been investigated in relation to the retinal location.
Results of such experiments showed that in central vision
(1520 ), both lamellar and radial ow can induce direction-
ally specic antero-posterior body sway [4,69]. In contrast, Figure 2 Relative retinal displacements for an observer mov-
in the retinal periphery, only the lamellar ow (horizontal ing from right to left, xating an object F at an intermediate
ow) can provoke body sway [69]. The interaction between distance. The further away the object is from the xation point,
the retinal location and the structure of the visual ow was the greater its angular displacement () on the retina is during
interpreted by Stoffregen as reecting a differential sensi- lateral displacement of the observer.
bility of the peripheral and central retina to the different
structures of the visual ow [69]. Irrespective of retinal
area, regulation of antero-posterior body sway appears to readjustments in the direction of motion. However, system-
be mainly controlled by radial ow (expansion/contraction) atic lateral postural adjustments in the opposite direction
when the observer is looking in the direction of self-motion of display motion were evoked when a stationary point of
and by lamellar ow (pure translation) when looking in the xation (window frame) was placed in the foreground (see
direction orthogonal to self-motion [5,69,75]. Conversely, Fig. 3). In these circumstances, displacement of the visual
when looking in the perpendicular direction of motion, lat- background behind a stationary window (foreground) mim-
eral body sway is largely regulated on the basis of lamellar
ow.

3D visual cues and motion parallax

A natural 3D environment provides a pattern of differen-

tial horizontal movements between elements in the eld
related to the displacement of the observer. These rela-
tive movements are referred to as motion parallax. Since
lateral sway is supposed to be mainly controlled by lamel-
lar cues, motion parallax is likely a relevant cue involved in
its regulation. The term motion parallax or relative motion
parallax is commonly used to denote the (horizontal) opti-
cal motion of a discontinuity produced by a near object
against that produced by a far object due to a change in
the observers position. (Fig. 2). It does also refer to change
in the projective relations among elements in the visual
eld in condition of simulated motion while the observer is
stationary and the scene moving. Motion parallax contains
information about both the direction of motion and the mag-
nitude of depth relative to the xation point as illustrated in
Fig. 2.
As the observer moves to the left while tracking an object
F (xation point), the direction of motion of the images of
other objects in the environment depends on their depth
relative to the xation point. Objects behind the xation Figure 3 Sample records of both head and centre des pres-
point (background) move on the retina in the direction of sions (COP) lateral translation evoked by the displacement of
self-motion while objects in front of that xation point the background while looking straight at the background (back-
move in the opposite direction. Bronstein and Buckwell [14] ground xation: left panel) or xation of the foreground (right
and Guerraz et al. [39,42] investigated whether this optical panel). Upward deections indicate deviation in the direction
geometry would determine the direction of visually-evoked of stimulus motion while downward deections indicate devia-
postural response. As mentioned above, movement of the tion in the opposite direction to motion. Adapted from Guerraz
whole visual surrounding (either 2D or 3D) induces postural et al. [42].
396
ics an optic ow (motion parallax) that a standing subject
would experience during spontaneous quiet lateral sway in
a 3D environment. As shown in Fig. 2, when an observer x-
ates a near object, the images of distant stationary objects
move in the same direction as the head. Therefore, pos-
tural adjustment in the opposite direction to background
motion is consistent with the direction of image movement
on the retina that a moving subject would experience in a
stationary 3D environment.
The role of motion parallax was further conrmed in an
experiment on spontaneous body sway in a stable 3D envi-
ronment [38]. In that experiment, standing subjects were
required to xate a LED placed at 45 cm from their eyes
in a dark room. Motion parallax was manipulated by intro-
ducing a second LED either at 170 cm (maximum parallax),
85 cm away from the xed LED (medium parallax) or at
the same plane (no parallax). Motion parallax is propor-
tional to the distance between the different stimuli and
inversely proportional to the distance between the xated
object and the observer. Consistently, postural stability (lat-
eral displacement) was proportionally improved with the
amount of motion parallax. Interestingly, power spectral
analysis revealed that the stabilizing effect of motion par-
allax was restricted to low frequency components of body
sway (<0.5 Hz). The distance between the different LED had
no effect on sagittal displacements since such displacements
mainly generate disparity and image size cues and not par-
allax [57].
The various experimental results presented above
therefore conrm the presence of two modes of visual sta-
bilization in the control of posture. The improvement of
stability when xating a single LED in darkness attests prin-
cipally to the extraocular motion perception mechanism.
The decreasing sway observed with increasing parallax, for
instance, attests to the afferent (ocular or retinal) motion
perception mechanism in action. It should be noted that, for
geometrical reasons, the relative motion of the two objects,
say angle (relative movement of the two objects on the
retina), is always smaller than (eye displacement). The
amplitude of is related to the distance between the two
objects (D) and to the distance between the xation point
and the subject (x) and is given by = arctan [dD/(x(x + D)]
(Fig. 4).
However, the threshold of perception for relative motion
between two objects is lower than that of absolute motion of
a single object [1,46,66]. Therefore, we can postulate that
the threshold of visual detection of body sway is lower for
afferent motion perception (angle ) than that for extraoc-
ular motion perception (angle ). One might envisage that
when eyetarget distance is large enough not to provide the
required amount of extraretinal information, the relative
movement between the foreground and a background far
away could still provide pure retinal information about body
sway. Experimental results tend to conrm such hypothesis.
Indeed, Guerraz et al. [38] showed that a single LED placed
at a distance of 1 m, seen in isolation, provided only little
stabilisation as compared to the condition with eyes closed.
When a second set of LED was placed 3 m behind this xation
point, body sway was reduced. Thus, for sufciently large
body sway with respect to eyetarget distance, the ocu-
lar and the extraocular perception systems would provide
two sources of visual information about body displacements
M. Guerraz, A.M. Bronstein
Figure 4 Geometrical relationship between the eye and a x-
ation target and a background object during lateral head sway
over a distance d (d = 1 cm). The angle , which represents
the angle traveled by the background object on the retina, is
plotted as a function of eye-xation target distance for three
different inter-object distances D. Adapted from Guerraz et
al. [38].
while the afferent system might remain the only one used
for small lateral body sway.
Conclusion
Although vision has been shown for almost one century to
be particularly relevant for postural control, the mecha-
nisms by which vision operates are still an open debate. The
present article reviews evidence supporting the view that
there are two modes (afferent and efferent) of visual detec-
tion of body sway. Another relevant question concerning
visual control of posture is whether visually-evoked postural
responses are secondary to visually-induced perceptual illu-
sions of body motion (vection) or rather largely automatic.
Although vection and visually-evoked postural responses are
to some extent related, the latter are not the consequence
of the former. Indeed, vection usually takes a few seconds
to build up, whereas evoked responses are often observed
within 1 s after stimulus motion [6,40,61]. However, Tana-
hashi et al. [72] recently suggested that the same visual
motion signal was used for vection and postural control
and speculated that the mechanisms underlying the process-
ing of visual motion signal for postural control and vection
perception operate using different thresholds. We recently
examined further this relation [43]. Standing subjects were
tested in two visual conditions which induce directionally
opposite postural responses, parallax-like postural response
in the opposite direction of visual background motion and
the conventional postural response in the direction of
the visual background in the absence of motion paral-
lax. Despite the opposite directions of postural responses,
vection occured systematically in the opposite direction
to background motion (and with long latencies greater
than 15 s). Thus, the different time course and incongru-
ency between direction of evoked postural and perceptual
responses suggest that perceptual and postural responses
are not causally related. However, as observed previously
[72,73], the magnitude of the postural responses induced
by the moving visual scene was signicantly increased dur-
ing periods of vection. On the basis of these results, we
Ocular versus extraocular control of posture and equilibrium
suggested that visually-induced postural responses might be
mediated by two different mechanisms, a short latency sys-
tem, driven by transient visual stimuli and sensitive to visual
geometry (parallaxno parallax), responsible for fast and
automatic postural sway adjustments, and a longer latency,
vection-enhanced postural mechanism, related to the con-
scious perception of self-motion during longer duration body
displacements. The nding that this latter, long latency
mechanism shows an exponential time course with long time
constants (>18 s) suggests that its underlying mechanism
may utilise the visuo-vestibular velocity storage integrator
[18].
Conict of interest
No conict of interest does exist.
References
[1] Abadi RC, Whittle JP, Worfolk R. Oscillopsia and tolerance to
retinal image movement in congenital nystagmus. Invest Oph-
thalmol Vis Sci 1999;40:33945.
[2] Amblard B, Carblanc A. Role of foveal and peripheral visual
information in maintenance of postural equilibrium in man.
Percept Mot Skills 1980;51:90312.
[3] Amblard B, Cremieux J, Marchand AR, Carblanc A. Lateral
orientation and stabilisation of human stance: static versus
dynamic visual cues. Exp Brain Res 1985;61:2137.
[4] Andersen GJ, Braunstein ML. Induced self-motion in central
vision. J Exp Psychol Hum Percept Perfom 1985;11:12232.
[5] Andersen GJ, Dyre B. Spatial orientation from optic ow in the
central visual eld. Percept Psychophys 1989;45:4538.
[6] Berthoz A, Lacour M, Soechting JF, Vidal PP. The role of vision
in the control of posture during linear motion. Prog Brain Res
1979;50:197209.
[7] Berthoz A, Pavard B, Young LR. Perception of linear horizontal
self-motion induced by peripheral vision (linearvection): basic
characteristics and visual-vestibular interactions. Exp Brain Res
1975;23:47189.
[8] Black FO, Nashner LM. Postural control in four classes of
vestibular abnormalities. In: Igarashi L, Black O, editors.
Vestibular and visual control on posture and locomotor equi-
librium. Karger: Basel; 1985. p. 27181.
[9] Black FO, Wall C, Rockette HE, Kitch R. Normal subject
postural sway during the Romberg test. Am J Otolaryngol
1982;3:30918.
[10] Bles W, Kapteyn TS, Brandt T, Arnold F. Mechanism of phys-
iological height vertigo. II. Posturography. Acta Otolaryngol
1980;89:53440.
[11] Brandt T, Arnold F, Bles W, Kapteyn T. Mechanism of physiolog-
ical height vertigo. I. Theoretical approach and psychophysics.
Acta Otolaryngol 1980;89:51223.
[12] Brandt T, Dichgans J, Koenig E. Differential effect of central
versus peripheral vision on egocentric and exocentric motion
perception. Exp Brain Res 1973;16:47691.
[13] Bronstein A, Guerraz M. Visual-vestibular control of posture
and gait: Physiological mechanisms and disorders. Curr Opin
Neurol 1999;12:511.
[14] Bronstein AM, Buckwell D. Automatic control of postural sway
by visual motion parallax. Exp Brain Res 1997;113:2438.
[15] Bronstein AM, Hood JD, Gresty MA, Panagi C. Visual control
of balance in cerebellar and parkinsonian syndromes. Brain
1990;1(13):76779.
[16] Bronstein AM. Suppression of visually evoked postural
responses. Exp Brain Res 1986;63:6558.
397
[17] Clement G, Jacquin T, Berthoz A. Habituation of postural read-
justments induced by motion of visual scenes. In: Igarashi M,
Black O, editors. Vestibular and visual control on posture and
locomotor equilibrium. Karger: Basel; 1985. p. 99104.
[18] Cohen B, Henn V, Raphan, Dennett TD. Velocity storage, nys-
tagmus, and visual-vestibular interactions in humans. Ann N Y
Acad Sci 1981;374:42133.
[19] Day BL, Steiger MJ, Thompson PD, Marsden CD. Effect of
vision and stance width on human body motion when stand-
ing: implications for afferent control of lateral sway. J Physiol
1993;469:47999.
[20] DellOsso LF, Averbuch-Heller L, Leigh RJ. Oscillopsia sup-
pression and foveation period variation in congenital,
latent and acquired nystagmus. Neuroophthalmology 1997;18:
16383.
[21] DellOsso LF, Leigh RJ. Foveation period stability and oscil-
lopsia suppression in congenital nystagmus: an hypothesis. J
Neuroophthalmol 1992;12:16983.
[22] Delorme A, Martin C. Roles of retinal periphery and depth
periphery in linear vection and visual control of standing in
humans. Can J Psychol 1986;40:17687.
[23] Di Girolamo S, Di Nardo W, Cosenza A, Ottaviani F, Dickmann A,
Savino G. The role of vision on postural strategy evaluated in
patients affected by congenital nystagmus as an experimental
model. J Vestib Res 1999;9:44551.
[24] Dichgans J, Brandt T. Visual-vestibular interaction: effects on
self-motion perception and postural control. In: Held R, Lei-
bowitz H, Teuber H, editors. Handbook of sensory physiology.
New York: Springer-verlag; 1978. p. 755804.
[25] Dichgans J, Held R, Young LR, Brandt T. Moving visual
scenes inuence the apparent direction of gravity. Science
1972;178:12179.
[26] Dichgans J, Mauritz KH, Allum JH, Brandt T. Postural sway in
normals and atactic patients. Analysis of the stabilizing and
destabilizing effects of vision. Agressologie 1976;17:1524.
[27] Dickinson J. A note on the role of periphral vision in static
balancing. Ergonomics 1969;12:935.
[28] Dijkstra TM, Gielen CC, Melis BJ. Postural responses to station-
ary and moving scenes as a function of distance to the scene.
Hum Mov Sci 1992;11:195203.
[29] Dijkstra TMH, Schoner G, Gielen CC. Temporal stability of the
action-perception cycle for postural control in a moving visual
environment. Exp Brain Res 1994;97:47786.
[30] Eggert T, Straube A, Schroeder K. Visually induced motion
perception and visual control of postural sway in congenital
nystagmus. Behav Brain Res 1997;88:1618.
[31] Fluckiger M, Baumberger B. The perception of an optical ow
projected on the ground surface. Perception 1988;17:63345.
[32] Fox CR. Some visual inuences on human postural equilib-
rium: binocular versus monocular xation. Percept Psychophys
1990;47:40922.
[33] Gibson JJ, Olum O, Rosenblatt F. Parallax and perspective dur-
ing aircraft landings. Am J Psychol 1955;68:37285.
[34] Gielen CC, van Asten WN. Postural responses to simulated mov-
ing environments are not invariant for the direction of gaze.
Exp Brain Res 1990;79:16774.
[35] Giese MA, Dijkstra TMH, Shoner G, Gielen CCAM. Iden-
tication of the nonlinear state-space dynamics of the
action-perception cycle for visually induced postural sway. Biol
Cybern 1996;74:42737.
[36] Glasauer S, Schneider E, Jahn K, Strupp M, Brandt T. How the
eyes move the body. Neurology 2005;65:12913.
[37] Gresty MA, Page NGR, Barratt HJ. Differential diagno-
sis of congenital nystagmus. J Neurol Neurosurg Psychiat
1984;47:93642.
[38] Guerraz M, Sakellari V, Burchill P, Bronstein AM. Inuence of
motion parallax in the control of spontaneous body sway. Exp
Brain Res 2000;131:24452.
398
[39] Guerraz M, Shallo-Hoffmann J, Yarrow K, Thilo KV, Bronstein
AM, Gresty MA. Visual control of postural orientation and equi-
librium in congenital nystagmus. Invest Ophthalmol Vis Sci
2000;41:3798804.
[40] Guerraz M, Thilo KV, Bronstein AM, Gresty MA. Inuence of
action and expectation on visual control of posture. Brain Res
Cog Brain Res 2001;11:25966.
[41] Guerraz M, Yardley L, Bertholon P, Pollak L, Gresty
MA, Bronstein AM. Visual vertigo: symptom-assessement,
spatial orientation and postural control. Brain 2001;124:
164656.
[42] Guerraz M, Gianna C, Burchill P, Gresty MA, Bronstein AM.
Effect of visual surrounding motion on body sway in a 3 D
environment. Percept Psychophys 2001;63:4758.
[43] Guerraz M, Bronstein AM. Mechanisms underlying visually
induced body sway. Neurosci Lett 2008;443:126.
[44] Held R, Dichgans J, Bauer J. Characteristics of moving
visual areas inuencing spatial orientation. Vis Res 1975;15:
35765.
[45] Jahn K, Strupp M, Krafczyk S, Schiler O, Glasauer S, Brandt
T. Suppression of eye movements improves balance. Brain
2002;125:200511.
[46] Johnson CA, Scobey RP. Effects of reference lines on displace-
ment thresholds at various duration of movement. Vis Res
1982;22:81921.
[47] Kotaka S, Croll GA, Bles W. Somatosensory ataxia. In: Bles W,
Brandt T, editors. Disorders of posture and gait. New York:
Elsevier Science; 1986. p. 17883.
[48] Kunkel M, Freudenthaler N, Steinhoff BJ, Baudewig J, Paulus
W. Spatial-frequency-related efcacy of visual stabilisation of
posture. Exp Brain Res 1998;121:4717.
[49] Lanska DJ, Goetz CG. Rombergs sign: development, adop-
tion, and adaptation in the 19th century. Neurology
2000;55:12016.
[50] Lee DN, Lishman JR. Visual proprioceptive control of stance. J
Hum Mov Stud 1975;1:8795.
[51] Leigh RJ, DellOsso LF, Yaniglos SS, Thurston SE. Oscillopsia,
retinal image stabilization and congenital nystagmus. Invest
Ophthalmol Vis Sci 1988;29:27982.
[52] Lestienne F, Schoechting JF, Berthoz A. Postural readjust-
ment induced by linear motion of visual scenes. Exp Brain Res
1977;28:36384.
[53] Masson G, Mestre DR, Pailhous J. Effects of the spatio-temporal
structure of optical ow on postural readjustments in man. Exp
Brain Res 1995;103:13750.
[54] Nougier V, Bard C, Fleury M, Teasdale N. Contribution of central
and peripheral vision to the regulation of stance. Gait Posture
1997;5:3441.
[55] Paulus W, Hawken M, Quintern J, Staube A, Krafczyk S, Botzel
K, et al. Multisensory versus monosensory stabilization of pos-
ture. Agressologie 1988;29:699703.
[56] Paulus W, Straube A, Brandt T. Visual stabilization of posture.
Brain 1984;107:114363.
[57] Paulus W, Straube A, Krafczyk S, Brandt T. Differential effects
of retinal targets displacement changing size and changing dis-
parity in the control of anterior/posterior and lateral body
sway. Exp Brain Res 1989;78:24352.
[58] Paulus W, Straube A, Brandt T. Visual postural performance
after loss of somatosensory and vestibular function. J Neurol
Neurosurg Psychiatry 1987;50:15425.
M. Guerraz, A.M. Bronstein
[59] Perterka RJ, Benolken MS. Role of somatosensory and vestibular
cues in attenuating visually induced human postural sway. Exp
Brain Res 1995;105:10110.
[60] Post RB. Circular vection is independent of stimulus eccentric-
ity. Perception 1988;17:73744.
[61] Previc FH. The effects of dynamic visual stimulation on per-
ception and motor control. J Vestib Res 1992;2:28595.
[62] Redfern MS, Furman JM. Postural sway of patients with vestibu-
lar disorders during optic ow. J Vestib Res 1994;4:22130.
[63] Roll JP, Roll R. From eye to foot: a proprioceptive chain involved
in postural control. In: Amblard B, Berthoz A, Clarac F, editors.
Posture and gait: development, adaptation and modulation.
Amsterdam, Oxford: Elsevier; 1988. p. 15564.
[64] Rougier P, Garin M. Performing saccadic eye movements
or blinking improves postural control. Motor Control
2007;1(1):21323.
[65] Rushton DN, Brandt T, Paulus W, Krafcwyk S. Postural sway dur-
ing retinal image stabilisation. J Neurol Neurosurg Psychiatry
1989;52:37681.
[66] Snowden RJ. Sensitivity to relative and absolute motion. Per-
ception 1992;21:5638.
[67] Sperry RW. Neural basis of the spontaneous optokinetic
response produced by visual inversion. J Comp Physiol Psych
1950;43:4829.
[68] Stevens J, Emerson R, Gersstein L, Kallos T, Neufeld G, Nichols
G, et al. Paralysis of the awake human: visual perceptions. Vis
Res 1976;16:938.
[69] Stoffregen TA. Flow structure versus retinal location in the
optical control of stance. J Exp Psychol Hum Percept Perform
1985;11:55465.
[70] Stoffregen TA. The role of optical velocity in the control of
stance. Percept Psychophys 1986;39:35560.
[71] Straube A, Krafczyk S, Paulus W, Brandt T. Dependence of visual
stabilization of postural sway on the cortical magnication fac-
tor of restricted visual elds. Exp Brain Res 1994;99:5016.
[72] Tanahashi S, Ujike H, Kozawa R, Ukai K. Effects of visually simu-
lated roll motion on vection and postural stabilisation. J Neuro
Eng Rehab 2007;4:39.
[73] Thurrell AE, Bronstein AM. Vection increases the magnitude and
accuracy of visually evoked postural responses. Exp Brain Res
2002;147:55860.
[74] Travis RC. An experimental analysis of dynamic and static equi-
librium. J Exp Psychol 1945;35:21634.
[75] Van Asten WN, Gielen CC, Diener van der Gon JJ. Postural
adjustments induced by simulated motion of differently struc-
tured environments. Exp Brain Res 1988;73:37183.
[76] Van Asten WN, Gielen CC, Diener van der Gon JJ. Postural
movements induced by rotations of visual scenes. J Opt Soc
Am A 1988;5:17819.
[77] von Holst E, Mittelstaedt H. Das Reafferenzprinzip: Wech-
selwirkungen Zwischen Zentralnervensystem und Peripherie.
Naturwissenschaften 1950;37:4647.
[78] Weir CR. Spatial localisation: does extraocular muscle pro-
prioception play a role? Graefes Arch Clin Exp Ophthalmol
2000;238:86873.
[79] White KD, Post RB, Leibowitz HW. Saccadic eye movements and
body sway. Science 1980;208:6213.
[80] Wolsley CJ, Sakellari V, Bronstein AM. Reorientation of visually
evoked postural responses by different eye-in-orbit and head-
on-trunk angular positions. Exp Brain Res 1996;111:2838.