Do sensorial manipulations affect subjects differently
depending on their postural abilities?
Thierry Paillard, Riadh Bizid, Philippe Dupui
............................................................... ............................................................... .....
See end of article for
Dr T Paillard, Department of
Sports Science, Universite ´ de
Pau et des Pays de l’Adour,
6500 Tarbes, France;
Accepted 21 January 2007
Published Online First
20 February 2007
Br J Sports Med 2007;41:435–438. doi: 10.1136/bjsm.2006.032904
Objectives: To examine whether sensorial manipulation affects subjects differently according to their postural
performance and the strategies used. The literature showed that the level of competition of soccer players
influences their postural performance and strategy.
Methods: Eight high-level (HL) professional soccer players and nine regional-level (RL) soccer players were
tested (1) in a reference condition and (2) in a manipulated sensorial condition (MAN). The MAN condition
consisted of perturbing the proprioceptive and exteroceptive information. For each postural condition,
balance was assessed by measuring the centre of foot pressure using a force platform during a test of bipedal
Results: The postural control was less perturbed in the HL than in the RL players in the two postural conditions.
Moreover, the group–condition interaction showed that the postural control was less disturbed in the HL than
in the RL players when the sensory information was manipulated.
Conclusions: The HL soccer players probably possessed a better internal model of verticality than the RL
players. Subjects who had a better postural control level were less disturbed by sensorial manipulation than
the others in postural regulation.
Indeed, cutaneous sensory disturbance by hypothermic anaes-
thesia,1myotendinous interference by muscle vibration,2pain
by stimulation of skin thermoreceptors,3electromyostimula-
tion,4galvanic vestibular stimulation5or visual manipulation by
stroboscopic light6affects postural stability in healthy subjects.
Keshner et al7showed that combined disturbances (ie, visual,
proprioceptive, vestibular) caused greater disturbance than any
one presented alone. However, combined perturbations might
not disturb all subjects in the same way. We do not know
whether subjects who present a better postural performance
level counteract differently to the effects of manipulating
proprioceptive and exteroceptive information in postural
regulation than the others. Moreover, the postural strategy—
that is, preferential involvement of short (visuovestibular
contribution) or long (myotatic participation) neuronal loops
in postural regulation—can also influence the effects of
sensorial perturbation. To study this, sportsmen at different
levels of competition should be studied.
Indeed, the level of competition influences the postural
performance: the higher the level of competition, the more
stable the posture.8 9The level of competition also influences
the preferential use of the different neuronal loops involved in
balance regulation as Paillard et al9showed with soccer players.
Hence, one can wonder (1) whether high-level soccer players
can preserve better postural control than players at a lower level
when their proprioceptive and exteroceptive informations are
disturbed and (2) whether these two categories of players are
differently affected by sensorial manipulation.
The aim of this work was to compare the postural behaviour
between high-level soccer players and players at a lower level in
two different postural conditions. The first condition was the
reference (no sensorial perturbation), whereas in the second
condition proprioceptive and exteroceptive information was
ostural regulation requires the integration of propriocep-
tive and exteroceptive information. When this information
is perturbed or manipulated, postural control is altered.
Eight male professional soccer players who played at a national
championship (high level or HL group) and nine male amateur
soccer players who played at a regional level (regional level or
RL group) participated in the study. The subjects’ morpholo-
gical characteristics (table 1) showed no difference between the
two groups (one-factor analysis of variance (ANOVA)). All the
subjects had practised soccer for at least 6 years. None of the
subjects had ankle, knee or hip injuries in the past 4 months.
HL players train almost every day and RL players twice a week.
The experiment was conducted at the middle of the competition
season. The subjects signed a written informed consent
according to the declaration of Helsinki.
We asked the subjects to stand as still as possible, with their
arms along the body, on two legs barefoot on a force platform
(PostureWin, Techno Concept, Cereste, France; 40 Hz fre-
quency, 12 bit A/D conversion), which recorded the displace-
ments of the centre of foot pressure (COP) with three strain
gauges. The subjects were placed according to precise marks.
Their legs were tended and their feet formed a 30˚angle relative
to each other (inter-malleolar distance of 5 cm). They were
tested (1) in the reference condition (REF) and (2) in a
manipulated sensorial condition (MAN). For the two condi-
tions, the test lasted 51.2 s first with eyes open (EO) then with
eyes closed (EC). In the EO situation, the subjects looked at a
fixed level target 2 m away. In the EC situation, they were
asked to keep their ‘‘gaze’’ in a straight-ahead direction.
Butterworth filter with a 6 Hz low-pass cut-off frequency. The
COP surface area (90% confidence ellipse) evaluates the
Abbreviations: ANOVA, analysis of variance; COP, centre of foot
pressure; EC, eyes closed; EO, eyes open; HL, high level; LF, low frequency;
MAN, manipulated sensorial condition; REF, reference condition; RL,
subject’s postural performance: the smaller the area, the better
the performance.10The COP velocity (sum of the cumulated
COP displacements divided by the total time) evaluates the
subjects’ postural control.10COP excursions were also investi-
gated in the frequency domain to assess the preferential
involvement of short or long neuronal loops in balance
regulation.11Fast Fourier transforms were applied to COP
displacements from 0 to 20 Hz. Hence, the total spectral energy
was calculated and distributed in three frequency bands: low
frequencies (LF), 0–0.5 Hz; medium frequencies, 0.5–2 Hz; and
high frequencies, .2 Hz.12Low frequencies mostly account for
visuovestibular regulation, medium frequencies for cerebellar
regulation and high frequencies for proprioceptive regulation.12
Before the postural tests (EO and EC) in the MAN condition,
subjects’ feet were placed in a metal bucket filled with ice-cold
water (5˚C) to mid-calf for 20 min. The aim was to anaesthetise
the sensitivity of the cutaneous receptors.1Immediately after
cooling, the postural tests were performed in the MAN
condition. During each test, the gastrocnemius medialis and
lateralis, and vastus medialis and lateralis of both legs were
electrically stimulated (Cefar Rehab 4 Pro, Malmo, Sweden).
One electrode (Cefar Stimtrode, 50689 mm, Axelgard, Sweden)
was placed over the motor point on each of the four muscles
with a biphasic symmetric square wave (continuous pulse
350 ms, 25 mA, 80 Hz). The aim was to disturb the myotatic
proprioceptive information.4The subjects wore a cervical collar
to jointly render the head with the trunk in order to limit
information from the cervical articulations (fig 1).
The effects of condition (REF and MAN), group (HL and RL)
and vision (EO and EC) were tested using three-factor ANOVA
with repeated measures on three factors. When significant
treatment effect occurred, Newman–Keuls post hoc analyses
were used to test the difference among means. The level of
significance was set at p,0.05.
For the REF and MAN conditions, the COP velocity was
significantly greater in the RL group than in the HL group with
EO and EC (table 2). For both groups (HL and RL), in the two
conditions (REF and MAN), the COP velocity was significantly
greater in the EC situation than in the EO situation (table 2).
Furthermore, the condition–vision interaction was significant
for the COP velocity (table 2). The COP surface area was not
different between the two groups in all the conditions (table 2).
The MAN condition significantly altered the COP surface area
and the COP velocity for both groups with EO and EC (table 2).
In addition, a significant group–condition interaction was
observed for the COP velocity (table 2).
The spectral analysis showed that the energies of the low-
frequency band and the medium-frequency band were greater
for the HL players than for the RL players irrespective of the
postural condition (REF and MAN; table 3). Statistical analysis
did not reveal any other significant effects or interactions.
Irrespective of the condition (REF or MAN), COP velocity was
slower in the HL soccer players than in the RL players, which
showed that the HL players demonstrated better postural
control than the RL players. This confirms the results of Paillard
et al,9which showed that national players’ postural control was
better than that of regional players in the REF condition.
Moreover, Era et al8had already also observed that there is a
close relationship between the level at which the sport is played
and the effectiveness of postural regulation in rifle shooters.
characteristics between the two groups (one-factor analysis
Comparison of the subjects’ morphological
HL group (n=8)RL group (n=9)Statistics
HL, high level; NS, not significant; RL, regional level.
Values are mean (SD).
situation) in the manipulated sensorial condition immediately after cooling
of the subjects’ feet in a metal bucket filled with ice-cold water. During this
test, the gastrocnemius medialis and lateralis and vastus medialis and
lateralis were electrically stimulated and the subject wore a cervical collar.
Informed consent was obtained for publication of this figure.
Illustration of the postural test (eyes wrapped for eyes-closed
regional-level and high-level soccer players with eyes open
or closed in the two conditions
Comparison of postural parameters between
GroupConditionVision Surface areaVelocity
EC, eyes closed; EO, eyes open; HL, high level; MAN, manipulated
condition; REF, reference condition; RL, regional level.
*Significant group effect (RL or HL).
?Significant vision effect (EO or EC).
`Significant condition effect (REF or MAN).
1Significant group–condition interaction (p,0.05).
?Significant condition–vision interaction.
Mean (SD) values of the centre of foot pressure (COP) surface area (mm2)
and the COP velocity (mm/s).
436 Paillard, Bizid, Dupui
Two phenomena could explain why the HL players demon-
strated better postural control than the RL players. Firstly, the
HL players presented probably a better neuromuscular activa-
tion responsiveness (they modulated better muscle activity) at
the level of ankle extensors than the RL players as the COP
velocity is inversely correlated to the ankle torque.13Secondly,
according to Perrin et al,14high-level athletes develop a higher
sensitivity of sensory receptors than lower level athletes.
For the two groups in the two conditions, the COP velocity
being significantly greater in the EC situation than in the EO
situation means the suppression of vision disturbed postural
control in all circumstances. Moreover, the condition–vision
interaction was significant for the COP velocity, which
illustrated that the suppression of vision disturbed the postural
control more in the MAN condition than in the REF condition
for the two groups. Without sensorial manipulations (REF
condition), the proprioception partially compensated the loss of
visual information, whereas with sensorial manipulations
(MAN condition), the proprioception compensated less the loss
of visual information. Therefore, sensorial manipulations can
disturb the proprioceptive information and thus the postural
control. However, Vuillerme et al15showed that after a
proprioceptive perturbation delivered by tendon vibration of
ankle muscles, the contribution of visual information was not
altered by the reintegration of proprioceptive information.
These authors only disturbed proprioception at a very local
level, whereas in the present study the manipulation concerned
proprioceptive information at a more extended level (electrical
stimulation of calf and thigh muscles, and reduced neck
movements due to the cervical collar) and the plantar
cutaneous information. Thus, the wider the manipulation of
sensory information, the more visual information would
compensate the deterioration of the efficiency of the disturbed
sensory receptors to regulate balance.
Moreover, the COP velocity and the COP surface area were
greater in the MAN condition than in the REF condition for the
two groups. These results showed that the sensory manipula-
tion deteriorated the postural abilities. In addition, the group–
condition interaction showed that the COP velocity increase
was weaker in the HL soccer players than in the RL players in
the MAN condition. This means that the postural control was
less disturbed in the HL than in the RL players when the
sensory information was manipulated. With the vestibular
system not stimulated, two reasons could explain why the HL
soccer players compensated the effects of erroneous proprio-
ceptive information better than the RL players. Either the
otolithic inputs are more efficient and allow better detection
and control of the body orientation or the interoceptive inputs
are more precise and induce a better internal model of
verticality. Bringoux et al16have already shown that the
relevance of otolithic and/or interoceptive inputs increases with
increasing sporting expertise. The results of the spectral
analysis confirmed this as the visuovestibular (LF band)
contribution was greater in the HL players than in the RL
players. However, the LF band showed neither a significant
condition effect nor a group–condition interaction. Hence, the
vestibular system could not lead to a better compensation of
otolithic information in the HL group than in the RL group in
the MAN condition. Moreover, in the static postural condition,
the frequencies of body oscillations are below the detection
threshold of movement of the vestibular system.17Therefore, in
the MAN condition the vestibular system probably did not
bring any additional information in the HL soccer players than
in the RL soccer players to regulate their balance. The results of
Balter et al18confirm this hypothesis as they showed that
professional gymnasts did not present a higher sensitivity of the
vestibular system than control subjects. Therefore, the HL
soccer players probably possessed a better internal model of
verticality (a better knowledge of the body axis and verticality)
than the RL players.
The postural regulation of subjects with a better sports
performance level was less disturbed by sensorial manipulation
than that of sportsmen with a lower level of performance.
Spectral analysis of postural sways in the reference and manipulated conditions
GroupCondition Vision LF bandMF bandHF band
6274.2 (2795.4 )
EC, eyes closed; EO, eyes open; HF, high frequency; HL, high-level; LF, low frequency; MAN, manipulated condition;
MF, medium frequency; REF, reference condition; RL, regional level.
*Significant group effect (RL or HL; p,0.05). No significant condition effects (REF or MAN) or vision effects (EO or EC)
and no significant interactions were observed.
Distribution of normalised values of the total spectral energy (V2/Hz) in three frequency bands is as follows:
LF (0–0.5 Hz); MF (0.5–2 Hz); and HF (.2 Hz). Values are mean (SD)
What is already know on this topic
There is a close relationship between the level at which the sport
is played and the effectiveness of postural regulation. However,
no study has analysed (1) whether high-level sportsmen can
preserve better postural control than sportsmen at a lower level
when their proprioceptive and exteroceptive information are
disturbed and (2) whether these two categories of sportsmen
are differently affected by sensorial manipulation.
What this study adds
The present study shows that the high-level sportsmen conserve
a better postural control than the sportsmen at a lower level
when their proprioceptive and exteroceptive information is
disturbed. Furthermore, it also shows that the postural control is
less disturbed in the high-level sportsmen than in the sportsmen
at a lower level when the sensory information was manipulated.
Sensorial manipulation and postural regulation 437
We thank Daniel Boschat from Techno-Concept for graciously lending
us the force platform.
Thierry Paillard, Riadh Bizid, Department of Sports Science, Universite ´ de
Pau et des Pays de l’Adour, Tarbes, Cedex, France
Philippe Dupui, School medicine, Universite ´ de Toulouse 3, Toulouse,
Competing interests: None declared.
Informed consent was obtained for publication of figure 1.
1 Stal F, Fransson PA, Magnusson M, et al. Effects of hypothermic anesthesia of the
feet on vibration-induced body sway and adaptation. J Vestib Res
2 Uimonen S, Sorri M, Laitakari K, et al. A comparison of three vibrators in static
posturography: the effect of vibration amplitude on body sway. Med Eng Phys
3 Blouin JS, Corbeil P, Teasdale N. Postural stability is altered by the stimulation of
pain but not warm receptors in humans. BMC Musculoskelet Disord 2003;4:23.
4 Hoffman MA, Koceja DM. Dynamic balance testing with electrically evoked
perturbation: a test of reliability. Arch Phys Med Rehabil 1997;78:290–3.
5 Popov KE, Kozhina GV, Smetanin BN, et al. Postural responses to combined
vestibular and hip proprioceptive stimulation in man. Eur J Neurosci
6 Isableu B, Ohlmann T, Cremieux J, et al. Differential approach to strategies of
segmental stabilisation in postural control. Exp Brain Res 2003;150:208–21.
7 Keshner EA, Kenyon RV, Langston J. Postural responses exhibit multisensory
dependencies with discordant visual and support surface motion. J Vestib Res
8 Era P, Konttinen N, Mehto P, et al. Postural stability and skilled performance-a
study on top-level and naive rifle shooters. J Biomech 1996;29:301–6.
9 Paillard T, Noe ´ F. Effect of expertise and visual contribution on postural control in
soccer. Scand J Med Sci Sports 2006;16:345–8.
10 Caron O, Gelat T, Rougier P, et al. A comparative analysis of the centre of gravity
and centre of pressure trajectory path lengths in standing posture: an estimation
of active stiffness. J Appl Biomech 2000;16:234–47.
11 Dietz V, Mauritz KH, Dichgans J. Body oscillations in balancing due to segmental
stretch reflex activity. Exp Brain Res 1980;40:89–95.
12 Golomer E, Dupui P, Bessou P. Spectral frequency analysis of dynamic balance in
healthy and injured athletes. Arch Int Physiol Biochim Biophys 1994;102:225–9.
13 Masani K, Popovic MR, Nakazawa K. et al, Importance of body sway velocity
information in controlling ankle extensor activities during quiet stance.
J Neurophysiol 2003;90:3774–82.
14 Perrin P, Deviterne D, Hugel F, et al. Judo, better than dance, develops
sensorimotor adaptabilities involved in balance control. Gait Posture
15 Vuillerme N, Teasdale N, Nougier V. The effect of expertise in gymnastics on
proprioceptive sensory integration in human subjects. Neurosci Lett
16 Bringoux L, Marin L, Nougier V, et al. Effects of gymnastics expertise on the
perception of body orientation in the pitch dimension. J Vestib Res
17 Fitzpatrick R, McCloskey DI. Proprioceptive, visual and vestibular thresholds for
the perception of sway during standing in humans. J Physiol 1994;478:173–86.
18 Balter SG, Stokroos RJ, Akkermans E, et al. Habituation to galvanic vestibular
stimulation for analysis of postural control abilities in gymnasts. Neurosci Lett
EDITORIAL BOARD MEMBER ............................................................... .............
Professor (Docent) of Sports Medicine at the University of
Jyva ¨skyla ¨, Finland and Associate Professor (Docent) of Exercise
Physiology at the University of Tampere, Finland. Professor
Kannus is a specialist physician in sports and exercise medicine
with clinical experience at the Tampere Research Center of
Sports Medicine for more than 20 years.
In 1999, Professor Kannus was nominated to be responsible
for sports and exercise medicine education at the University of
Tampere. His scientific work has focused on basic and applied
research of the musculoskeletal system of the human body—
that is, its anatomy, biomechanics, physiology, and pathology,
as well as prevention, treatment and rehabilitation of injuries
and disorders. He has over 900 publications to his name, of
which 320 are scientific articles in peer-reviewed international
journals, with 28 international book chapters and two interna-
tional textbooks. Professor Kannus’ pedagogic abilities are
widely recognised. He has attended many foreign international
scientific congresses as a keynote speaker or invited lecturer,
giving 75 presentations altogether. These visits have been not
only in Scandinavia and Europe but also in the USA, Canada,
Japan, Argentina, and Australia.
ekka Kannus, MD, PhD (born 29 July 1959) is the Chief
Physician of the Injury & Osteoporosis Research Center,
UKK Institute, Tampere, Finland. He is also Associate
438Paillard, Bizid, Dupui