Neurophysiologie Clinique/Clinical Neurophysiology (2013) 43, 237—242

Disponible en ligne sur

www.sciencedirect.com

ORIGINAL ARTICLE/ARTICLE ORIGINAL

Achilles tendon vibration shifts the center

of pressure backward in standing and
forward in sitting in young subjects
Un stimulus vibratoire sur les tendons d’Achille déplace le
centre de pression en arrière en position debout et en avant
en position assise chez des sujets jeunes

G. Barbieri a,∗, A.-S. Gissot b, V. Nougier a, D. Pérennou a,b

a
Équipe Santé-Plasticité-Motricité, Laboratoire TIMC-IMAG, UMR UJF CNRS 5525, Université Grenoble 1,
38041 Grenoble, France
b
Clinique de Médecine Physique et Réadaptation, Institut de Rééducation, Hôpital Sud, CHU Grenoble, BP
338, avenue de Kimberley, 38434 Echirolles cedex, France

Received 14 September 2012; accepted 23 June 2013

Available online 17 July 2013

KEYWORDS Summary
Postural control; Objective. — An Achilles tendon (AT) vibration is known to disrupt the postural control in stand-
Verticality; ing (‘‘vibration-induced falling’’, VIF) and to backward tilt the postural vertical in restrained
Force platform; sitting position, suggesting a link between AT vibration and internal representations involved
Vibration; in postural control. A recalibration of some troubles in body orientation by oriented sensory
Proprioception manipulations could be of great clinical interest. In order to use the VIF paradigm in a procedure
suitable for a rehabilitation context, AT vibration deserves to be more investigated in sitting,
for security reasons, and first in young participants.
Method. — In 12 healthy participants (6 men/6 women; 23.3 ± 1.9 years), posturographic data
to AT vibrations (85 Hz) were recorded over 30 s standing and 40 s sitting trials.
Results. — Surprisingly, four types of differences were found between standing and sitting AT
vibrations: presence/absence of VIF and kinesthetic illusion, opposite directions of the centre of
foot pressure (CoP) displacements, different temporal profiles. In standing: VIF without kines-
thetic illusion, rapid backward shift of CoP with a peak of −54.6 ± 11.3 mm (with respect to
baseline P < 0.001) 2.8 ± 0.2 s after vibration onset. In sitting: surprising systematic forward shift
of the CoP throughout the duration of AT vibration with a maximum of 27.9 ± 18.9 mm (P < 0.05
with respect to the baseline) 19.6 ± 0.3 s after vibration onset, associated with a kinesthetic
illusion in most subjects.

∗ Corresponding author.
E-mail address: guillaume.barbieri@imag.fr (G. Barbieri).

0987-7053/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.neucli.2013.06.001
238 G. Barbieri et al.

Conclusion. — The present study invalidates our idea to extend in sitting the VIF paradigm but
the unexpected results open a new window about the basic mechanisms underlying muscle
vibration effects.
© 2013 Elsevier Masson SAS. All rights reserved.

Résumé
MOTS CLÉS Objectif. — Un stimulus vibratoire appliqué sur les tendons d’Achille (TA) perturbe le contrôle
Contrôle postural ; postural en position debout (vibration-induced falling/VIF, dans la littérature anglo-saxonne)
Verticalité ; et entraîne un biais vers l’arrière de la perception posturale de la verticale en position assise
Plateforme de force ; sans possibilité de mouvement, suggérant un lien entre vibration appliquée sur les TA et mod-
Vibration ; ulation des représentations internes impliquées dans le contrôle postural. La possibilité de
Proprioception recalibrer certains troubles de l’orientation corporelle par des manipulations sensorielles ori-
entées représente un grand intérêt en rééducation clinique. La vibration appliquée sur les TA
doit donc être davantage analysée, en position assise pour des raisons de sécurité, et d’abord
avec des sujets jeunes.
Méthode. — Les données posturographiques de 12 participants sains (6 femmes/6 hommes ;
23,3 ± 1,9 ans) chez lesquels des stimuli vibratoires (85 Hz) furent appliqués au niveau des
tendons d’Achille ont été enregistrées en position debout et assise.
Résultats. — Quatre différences principales et inattendues ont été trouvées entre la position
debout et assise : présence/absence du VIF et d’illusion kinesthésique, déplacements opposés
du centre de pression (CdP), profils temporels différents. En position debout : VIF sans illusion
kinesthésique, rapide et puissant recul du CdP avec un pic à −54,6 ± 11,3 mm (comparé à la
moyenne sans vibration, p < 0,001) 2,8 ± 0,2 s après le début de la vibration. En position assise :
surprenante et systématique avancée du CdP durant toute la période de vibration avec un
maximum à 27,9 ± 18,9 mm (p < 0,05, comparé à la moyenne sans vibration) 19,6 ± 0,3 s après
le début de la vibration, associée à une illusion kinesthésique chez certains participants.
Conclusion. — L’idée d’étendre à la position assise le paradigme du VIF est invalidée. Toutefois,
les résultats inattendus obtenus ouvrent de nouvelles perspectives concernant la compréhension
des mécanismes et effets de la vibration musculaire.
© 2013 Elsevier Masson SAS. Tous droits réservés.

Introduction tilted representation of the vertical. These troubles might

Mechanical vibration has been widely used to modulate mus-
cular proprioception via Ia afferent fibers [11,16], and to
investigate its role in the control of erect posture.
In a freely standing participant, vibration is well known
to elicit a whole-body tilt, oriented forwards for tibialis
vibration and backwards for the Achilles tendons (AT) vibra-
tion [1,6,9,13,14,18,21], the so-called ‘‘vibration-induced
falling’’ or VIF [9]. VIF was initially interpreted as the conse-
quence of a gastrocnemius muscle contraction in response to
an illusion of lengthening of this muscle evoked by AT vibra-
tion [18]. This interpretation refers to a modulation of body
geometry around the ankles. An alternative interpretation
evoking a modulation of internal representations involved
in postural control has been suggested from experiments in
weightlessness [19], and recently confirmed [2,7,20]. The
study by Barbieri et al. [2] showed that AT vibration mod-
ulates the postural perception of vertical (PV). Using a
specific device wherein participants were strapped in a sit-
ting position to control body geometry, they showed that AT
vibration induced an immediate and systematic backward PV
tilt. Recently, tendinomuscular vibration has also been used
to improve the spatially oriented domain of balance control,
which is impaired in amputees [8] and stroke patients [4].
Impaired lateral (lateropulsion) or sagittal (retropulsion)
body orientation are major postural problems after a stroke
or in elderlies [3,15,17]. They may be due to a biased spatial
coordinates system referred to the vertical, and results in a
be eligible to a recalibration by oriented sensory manipu-
lations in order to improve patients’ postural balance. AT
vibration might be of great clinical interest for that purpose,
especially in elderly fallers exhibiting a retropulsion due to a
PV tilted backward [15]. One idea suggested by the study of
Barbieri et al. [2] was that the VIF paradigm could be clini-
cally used to modulate the severity of the retropulsion. This
implies a methodological approach to design a procedure
suitable for a rehabilitation context for frail people show-
ing a great postural instability. For security reasons, these
procedures must be implemented in sitting.
Although the ultimate goal is to improve frail or elderly
patients’ postural instability, a first step is to investigate
the VIF paradigm in sitting in young participants. Indeed
the VIF paradigm has never been tested in sitting. This was
the main objective of the present study. We hypothesized
that in young participants, with eyes closed, AT vibration
should entail a systematic backward displacement of the
body, associated to a backward shift of the centre of pres-
sure (CoP).
Patients and methods
Participants
Twelve healthy participants (23.3 ± 1.9 years; six females
and six males) naive with regard to the paradigm and
purpose of the study participated in this experiment. All
Achilles tendon vibration shifts the center of pressure in young subjects
participants signed an informed consent as required by the
Helsinki declaration (1964) and the local ethical committee.
Set up
Postural sway was measured by two rectangular (21 × 32 cm)
force platforms (PF02, Equi, France) placed side by side.
Each foot was placed on one of the two platforms (heels
separated by 9 cm, toe out at 30 degrees). The trajectory
of the CoP under each foot was calculated from the ver-
tical reaction forces measured by the four load cells of
each platform. The mean CoP displacement was then com-
puted from the CoP displacements under each foot and
from the body weight distribution between the two supports
[10].
Proprioceptive stimulation was delivered by two mechan-
ical inertial vibrators (VB 115, Techno concept, Cereste,
France) attached to the participant’s AT with elastic bands.
Vibration characteristics (frequency: 85 Hz and amplitude:
0.85 mm) were those proposed by Roll et al. [18]. Two
seconds were necessary to reach the selected frequency.
Procedure
We decided to design this study with a unique AT vibra-
tion in each posture, i.e. standing and sitting, in every
participant, for several reasons. The main reason was the
clinical perspective we wanted to explore. The possibility
of an interesting clinical effect implies an obvious effect for
each trial. Furthermore, due to the powerful and system-
atic effect of vibration on AT in standing, on the one hand,
and to the exploratory goal of this study in sitting, on the
other hand, the choice was made to study exclusively the
primary participants’ reaction to the vibratory stimulation
and not a mean behavior of successive trials. This allowed
eliminating possible post-effects induced by repeated AT
vibration [14,21]. Thus, only one posturographic assessment
per condition (one in standing and one in sitting) and per
participant was recorded in a randomized order between
participants.
In the standing condition, participants were asked to
remain upright without resisting the possible effect of vibra-
tion with the force platform placed under the feet, barefoot,
eyes closed, and arms relaxed and hanging freely along the
body. They were instructed that the experimenter would
protect them from falling should their balance become dis-
turbed. The data were recorded over a period of 30 s,
including 10 s with no vibration, 8 s with vibration and 12 s
with no vibration.
In the sitting condition, participants were seated on a
stool with the force platforms placed under the feet, bare-
foot, eyes closed and arms relaxed and hanging freely along
the body. The participants could adjust the seat height to
feel comfortable and with instruction of having their feet
entirely in contact with the force platforms. They were
instructed that the experimenter would protect them from
falling should their balance become disturbed. They were
asked to not resist the possible effect of the vibration. The
data were recorded over a continuous period of 40 s, includ-
ing 10 s before vibration onset, 20 s of vibration and 10 s after
the vibration stopped.
239
After each posturographic assessment, participants were
asked about any possible kinesthetic illusion they had expe-
rienced.
Data analysis
Data were analyzed in MATLAB® and statistical analy-
sis was performed using SPSS 14 software. Since vibratory
stimulation of the AT induces body sway mostly in the
anterior—posterior direction, only this axis was considered
for data analysis. According to the experiment, different
posturographic variables (expressed in millimeters) were
calculated for each condition and participant: in stand-
ing, the mean position of the CoP during the 10 s before
vibration onset, the backward peak reached with vibra-
tion (calculated between the 2nd and the 3rd second of
vibration, due to the 2 s latency to reach the selected
frequency), the forward peak reached immediately after
vibration was stopped (calculated between the 1st and
the 2nd second after vibration stopped), and the mean
position of the CoP during the 12 s after vibration; in
sitting, the mean position of the CoP before vibration
(10 s), the forward peak reached with vibration (calcu-
lated in the last second of vibration), and the mean
position of the CoP after vibration (10 s). All these values
were then adjusted to a normalized posturographic sig-
nal with a mean position before vibration onset taken as
a baseline equal to 0 mm. By convention, negative values
indicated a backward displacement of the CoP and posi-
tive values a forward displacement. Analyses of variance
(ANOVAs) were used for statistical comparison of the dif-
ferent sequences. The level of significance was set at 0.05.
Post-hoc analyses (Tukey tests) were used whenever nec-
essary. Data are given in the form of means ± standard
deviation.
Results
All participants were sensitive to vibration. The average dis-
placement along the anterior—posterior axis of the CoP of
the 12 participants in standing and sitting are represented
in Fig. 1.
As expected in standing condition, a rapid back-
ward tilt of the body was found, visually obvious and
quantified by a backward shift of the CoP (peak vibra-
tion = −54.6 ± 11.3 mm; P < 0.001). This peak was reached
in average 2.8 ± 0.2 s after the onset of vibration. Imme-
diately after the vibration was stopped, a rapid forward
shift of the CoP (peak after vibration = 31.5 ± 26.5 mm) was
observed, significantly different from the mean CoP after
vibration (4.5 ± 17.1 mm; P < 0.001). No significant differ-
ence was found between the CoP position before and after
vibration (P = 0.939), indicating that no post-effect lasted.
In two participants, the magnitude of the VIF obliged the
experimenter to slightly touch the participant in order to
avoid a backward fall. These trials were however taken into
account because the study was designed to analyze one trial
per participant and because this condition was analyzed just
to confirm what is well known in the literature. No partic-
ipant did not report any illusion but only the perception of
the backward falling.
240 G. Barbieri et al.

Figure 1 Experimental design and mean CoP displacements in standing and sitting conditions. With each foot positioned on one
of the two force platforms (PF02, Equi, France) placed side by side and the mechanical vibrators (85 Hz) located to the Achilles
tendon, Synoptic view of the posturographic records (upper part), mean of the main CoP positions during and after vibration for
n = 12 participants (middle part) and average displacement along the anterior-posterior axis of the CoP for n = 12 participants (lower
part) in (a) standing and (b) sitting condition. The start and the end of vibration are symbolized by downward and upward arrows,
respectively. The postural responses of all participants were adjusted so that the mean value of CoP positions before the stimulation
was considered as zero millimeter. Note that the ‘‘peak’’ values (backward and forward) of the curves do not exactly correspond
to the values of the bar diagrams. Indeed a peak value given by an average curve may be different from an average of individual
peak values pointed on individual curves.

In most cases in sitting condition, the AT vibration elicited
a kinesthetic illusion of foot dorsiflexion, more globally an
illusion of legs’ flexion or of sliding forward of the stool.
Surprisingly, the vibration induced a progressive forward
shift of the CoP (peak vibration = 27.9 ± 18.9 mm, P < 0.05,
19.6 ± 0.3 s after the onset of vibration). Once the vibration
stopped, a progressive backward displacement of the CoP
was observed and the mean position of the CoP after vibra-
tion (13.9 ± 13.5 mm) was not statistically different from
that observed before vibration (P = 0.295), indicating that
Achilles tendon vibration shifts the center of pressure in young subjects
no post-effect lasted. Interestingly, responses to AT vibra-
tion in standing and in sitting did not only spatially differ,
they also had very different temporal patterns, with a much
faster occurrence of the peak in standing.
Discussion
This study aimed at adapting the VIF paradigm to the sitting
posture in young participants. We hypothesized that both in
standing and in sitting conditions, AT vibration should entail
a systematic backward displacement of the body, associated
to a backward shift of CoP in young participants with eyes
closed.
This study confirmed a finding widely reported in the lit-
erature, that is, a rapid backward fall when AT are vibrated
in a participant standing with eyes closed (VIF), followed by
a rapid return to the reference position as soon as the vibra-
tion stopped [1,6,9,13,14,18,21,20]. This behavior without
illusion but with the perception of the backward falling may
be quantified by the displacement of CoP [1,6,14,21,20],
as we did here, with a peak quickly reached. We did not
find any VIF in sitting but conversely a surprising systematic
forward shift of the CoP throughout the duration of AT vibra-
tion, associated in most cases with a kinesthetic illusion. We
thus reveal four types of differences between standing and
sitting AT vibrations: presence vs absence of VIF, absence vs
presence of kinesthetic illusion, opposite directions of CoP
displacements, with very different temporal profiles. Differ-
ent mechanisms depending on the position might encounter
these differences.
While the present study invalidates our idea to extend
in sitting the VIF paradigm, the unexpected result opens a
new window about the basic mechanisms underlying muscle
vibration effects. At least two possible interpretations may
explain the puzzling effect. First, the forward shift of CoP
may correspond to a forward shift of the centre of mass, got
by the forward tilt of the trunk. This would be in contra-
diction with the VIF which consists in a backward whole
body tilt and not a forward body tilt. This argues for the
second interpretation: A plantar flexion induced a heel rise
and thus a forward shift of CoP, We cannot exclude that
this effect was a consequence of a stretch reflex (vibra-
tion reflex) in the calf muscles. However, it is well known
that vibration consequences are far more complex than a
low level stretch reflex, and papers of Burke et al. [5] and
Hagbarth [12] argue against this explanation. This effect
probably refers to an adjustment of the lower limbs’ geom-
etry in response to the illusion of muscular lengthening in
the vibrated muscles [18]. Indeed, it was noticeable that no
illusion was experienced by participants in standing whereas
in sitting all participants reported an illusion of foot dor-
siflexion or more globally an illusion of legs’ flexion or
of sliding forward of the stool. All together, results of a
previous paper [2] and the present one showed that AT ten-
don vibration induces two independent but complementary
effects: one dealing with the modulation of body orienta-
tion, the other one with the modulation of body geometry.
The predominance of one mechanism on the other could be
linked to the possibility for the vibrated segment to move,
which is a much easier action to perform than to move the
whole body. Ankle articulation left free, the central nervous
241
system (CNS) might interpret the new proprioceptive infor-
mation as a lengthening of the gastrocnemius muscle and
might modulate subsequently its body geometry around the
ankle in order to recover the initial position. This assump-
tion was supported in the present study by the existence
of kinesthetic illusions at the level of foot and leg when
participants were seated. Alternatively, an immobilization
of the vibrated segment might lead the CNS to update the
internal representation of the vertical resulting in a body
reorientation. From a fundamental and clinical point of
view, whether muscular vibration may modulate body geom-
etry, body orientation, or both depending on the context,
is an opened question. It could be further addressed in a
more complete study using force platforms (with feet and/or
seat CoP information) and movement and muscular activity
analysis systems in standing and sitting, in parallel to mea-
surement of verticality perception with or without muscular
vibration, in order to answer the following questions: which
muscles have to be vibrated? In which position? With key
body segments being free or strapped? Is there any addi-
tional effect when vibration is prolonged or repeated? What
is more interesting: using the effect induced in one direc-
tion or using a possible after effect induced in the opposite
direction?
Regarding the experimental procedure, two points have
to be discussed. It is well known that in standing, a few
seconds are necessary to reach the limits of stability and
to entail a backward fall. So, a prolonged time of vibra-
tion in this condition would not be useful. Concerning the
sitting condition, our experiment was not designed to ana-
lyze the optimal time of vibration. This experiment was
firstly performed in sitting in two pilot participants with a
time of vibration equivalent to that in standing (8 s) and
had revealed neither a backward shift of the CoP nor a
VIF but a progressive forward shift of the CoP, with no
steady state reached. We decided to set the experiment
in sitting with a longer duration of vibration (20 s). Inter-
estingly, after 20 s of vibration, the CoP carried on its
forward shift. This might be further analysed in another
experiment. As for the relatively large standard deviations
reported throughout the different CoP positions in stand-
ing as in sitting, they were partly linked to the obvious
individual sensibility of participants to the vibration and
also partly due to the differences in seat height and in
height, weight and feet size of the participants, which had
not been controlled in this study. It was however notice-
able that there was little variability regarding the time of
occurrence of the maximum CoP displacement in stand-
ing.
Conclusion
The present study invalidated the use of the VIF paradigm in
sitting, but provided interesting tracks for a better under-
standing of the effects induced by muscular vibration.
Disclosure of interest
The authors declare that they have no conflicts of interest
concerning this article.
242
Acknowledgement
The authors would like to express their gratitude to the
‘French Postural, Balance and Gait Research Society — SOF-
PEL’ for its grant (GB) to perform this study. Authors are
grateful to Claudine Baker for her helpful English correction.
References
[1] Abrahamova D, Mancini M, Hlavacka F, Chiari L. The age-related
changes of trunk responses to Achilles tendon vibration. Neu-
rosci Lett 2009;467:220—4.
[2] Barbieri G, Gissot AS, Fouque F, Casillas JM, Pozzo T, Perennou
D. Does proprioception contribute to the sense of verticality?
Exp Brain Res 2008;185:545—52.
[3] Barbieri G, Gissot AS, Perennou D. Ageing of the postural ver-
tical. Age (Dordr) 2010;32:51—60.
[4] Bonan IV, Marquer A, Eskiizmirliler S, Yelnik AP, Vidal PP.
Sensory reweighting in controls and stroke patients. Clin Neu-
rophysiol 2013;124(4):713—22.
[5] Burke D, Eklund G. Muscle spindle activity in man during stand-
ing. Acta Physiol Scand 1977;100:187—99.
[6] Capicikova N, Rocchi L, Hlavacka F, Chiari L, Cappello
A. Human postural response to lower leg muscle vibra-
tion of different duration. Physiol Res 2006;55(Suppl. 1):
S129—34.
[7] Ceyte H, Cian C, Zory R, Barraud PA, Roux A, Guerraz M. Effect
of Achilles tendon vibration on postural orientation. Neurosci
Lett 2007;416:71—5.
[8] Duclos C, Roll R, Kavounoudias A, Roll JP, Forget R.
Vibration-induced post-effects: a means to improve postural
asymmetry in lower leg amputees? Gait Posture 2007;26(4):
595—602.
[9] Eklund G. General features of vibration-induced effects on bal-
ance. Ups J Med Sci 1972;77:112—24.
G. Barbieri et al.
[10] Genthon N, Rougier P, Gissot AS, Froger J, Pelissier J, Perennou
D. Contribution of each lower limb to upright standing in stroke
patients. Stroke 2008;39:1793—9.
[11] Goodwin GM, McCloskey DI, Matthews PB. Proprioceptive illu-
sions induced by muscle vibration: contribution by muscle
spindles to perception? Science 1972;175:1382—4.
[12] Hagbarth KE. The effect of muscle vibration in normal man
and in patients with motor disorders. In: Desmedt J, editor.
New developments in electromyography and clinical neuro-
physiology, 3. Basel: Karger; 1973. p. 428—43.
[13] Ivanenko YP, Solopova IA, Levik YS. The direction of postural
instability affects postural reactions to ankle muscle vibration
in humans. Neurosci Lett 2000;292:103—16.
[14] Kavounoudias A, Gilhodes JC, Roll R, Roll JP. From balance regu-
lation to body orientation: two goals for muscle proprioceptive
information processing? Exp Brain Res 1999;124:80—8.
[15] Manckoundia P, Mourey F, Pfitzenmeyer P, Hoecke JV, Perennou
D. Is backward disequilibrium in the elderly caused by an abnor-
mal perception of verticality? A pilot study. Clin Neurophysiol
2007;118:786—93.
[16] McCloskey DI. Differences between the senses of movement
and position shown by the effects of loading and vibration of
muscles in man. Brain Res 1973;61:119—31.
[17] Perennou DA, Mazibrada G, Chauvineau V, et al. Lateropul-
sion, pushing and verticality perception in hemisphere stroke:
a causal relationship? Brain 2008;131:2401—13.
[18] Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive mes-
sages induced by tendon vibration in man: a microneurographic
study. Exp Brain Res 1989;76:213—22.
[19] Roll R, Gilhodes JC, Roll JP, Popov K, Charade O, Gurfinkel V.
Proprioceptive information processing in weightlessness. Exp
Brain Res 1998;122:393—402.
[20] Thompson C, Bélanger M, Fung J. Effects of bilateral Achilles
tendon vibration on postural orientation and balance during
standing. Clin Neurophysiol 2007;118:2456—67.
[21] Wierzbicka MM, Gilhodes JC, Roll JP. Vibration-induced postural
posteffects. J Neurophysiol 1998;79:143—50.