MOTOR SYSTEMS NEUROREPORT

Gait-dependent integration of neck muscle

aerent input
Gregoire Courtine,CA Charalambos Papaxanthis, Davy Laroche and Thierry Pozzo

ERM 207 INSERM Motricite and Plasticite, UFR STAPS, Universite de Bourgogne, 21708 Dijon, France
CA
Corresponding Author: gregoire.courtine@u-bourgogne.fr

Received17 June 2003; accepted 22 July 2003

DOI: 10.1097/01.wnr.0000091300.11924.89

We investigated the integration of neck muscle aerents during
walking and running. Subjects walked or ran straight ahead, with
or without an additional mass (20% of body weight). They per-
formed all trials without vibration and with continuous vibration
(80 Hz) applied to the lateral aspect of the neck.Vibration system-
atically caused body deviation toward the side opposite to the stim-
ulation. The amplitude of vibration-induced body deviations was
dramatically larger for walking (21.6 7 4.61) than for running
(8.0 7 2.51). The additional mass marginally straightened body
trajectory (average 5.6%), indicating that the gait-dependent eect
of neck vibration cannot solely be attributed to dierences in body
inertia between walking and running. We concluded that neck
muscle aerences are selectively gated according to the gait per-
formed. NeuroReport 14:2365^2368
c 2003 Lippincott Williams
& Wilkins.

Key words: Neck muscle; Running; Sensory gating; Vibration; Walking

INTRODUCTION MATERIALS AND METHODS

Sensory inputs arising from visual, vestibular, and muscular
receptors play a determinant role in successful locomotion.
Abnormal or faulty information from any of the above-
mentioned sensory receptors induces direction-specific
deviations of gait as well as side-specific imbalances [17].
In addition, the set of sensory inputs regulating the motor
output strongly depends on the nature of the movement.
For instance, Brandt and collaborators reported that an
acute [1] or transient [4] vestibular tone imbalance causes
fewer deviations from the intended path when running than
when slowly walking. The authors concluded that the
automatic locomotor programme for running suppresses the
destabilizing vestibular input.
Sensory information originating in neck muscles is also
critical for ensuring efficient locomotion. Neck muscle
afferents provide essential information to the brain for
computing head position with respect to the rest of the body
so as to adapt postural control [8,9]. Muscle vibration
delivered to the lateral aspect of the neck during walking [3]
or stepping in place [2] induces a deviation or a rotation of
the whole body, respectively, toward the side opposite to the
vibration.
Whether neck proprioception is subjected to a gait-
dependent modulation is not yet known. We hypothesized
that perturbation of proprioceptive sensory information
may induce similar effects on locomotor control as
perturbation of vestibular sensory input. Such a finding
would highlight a general sensorimotor strategy by which
the CNS modulates sensory input according to the require-
ments of the locomotor task.
Participants: Ten healthy subjects (five males and five
females, 2030 years old), without any neurological dis-
orders, participated in the present experiment after giving
their written consent in accordance with the declaration of
Helsinki. Eight subjects performed the whole protocol
because two of them (one male and one female) were
insensitive to neck muscle vibration (see above).
Motor tasks: In the no vibration condition, the subjects,
without vision (blindfolded), were asked to walk or run
straight ahead (from an erect body position) while carrying
or not carrying an additional mass (which was 20% of the
total body weight). The additional mass was located near
the bodys center of gravity by means of a weighted vest
worn by subjects. They started and stopped walking
according to a verbal signal provided by the investigator.
They performed five trials in each condition. All trials of
gait and mass conditions were randomized.
After completion of the no vibration condition, partic-
pants performed the same locomotor task but with a
vibratory stimulation applied to the right aspect of the neck
(see [2] for details). The vibration (80 Hz) was triggred by
photocell detectors located 2 m from the starting point.
Before performing the vibration condition we determined
(control condition) whether subjects were sensitive to neck
muscle vibration: two subjects who were not were elimi-
nated from the trial. We also ascertained that muscle
vibration had the same effects when applied to either the
right or left side by applying the stimulus alternately to

0959- 4965

c Lippincott Williams & Wilkins Vol 14 No 18 19 December 2003 2365

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 NEUROREPORT G. COURTINE ETAL.

right and left sides of the neck during running and walking
(three times for each condition). All subjects showed
bilateral effects for both walking and running. Figure 1
shows mean (7 s.d.) locomotor trajectories during bilateral
neck muscle vibration from a single subject. Control and
vibration conditions were separated by an interval of
20 min.

Data acquisition and analysis: We analyzed the displace-

ments of the head for all the experimental conditions. We
recorded the motion of the head for a distance of 4 m by
means of the 3D video ELITE system (six TV cameras,
100 Hz, BTS, Italy). The beginning of the recording field was
set at 5 m from the starting point. The X-axis and the Z-axis
of the spatial reference frame, respectively, coincided with
the direction of the straight line and with the direction
perpendicular to it (lateral direction). Subjects wore a
specially designed helmet equipped with two reflective
markers (8 mm diameter) placed in the head sagittal plane
and aligned with the naso-occipital axis at the level of the
arch of the eyebrows. The helmet was carefully placed so
that the mid-point between the two markers, labeled head
mid-point (xB, yB, zB; average of x, y, z coordinates of both
markers) was in the head/body yaw rotation axis, approxi-
mately at the center of the cephalic segment, and well
described the motion of head and body through space [10].
The head mid-point trajectories were calculated and
analyzed.
The gait cycle was defined as the interval between two
successive toes-off. During walking and running, the head
moves up and down with the movement of the body,
thereby describing a sinusoid whose troughs roughly
correspond with toes-off [11]. Successive toes-off were set Fig. 1. Average locomotor trajectories (mean path 7 s.d.) during walk-
at the trough of head up-down motions. ing and running while muscle vibration is delivered to the left (light) or
Length, duration and speed of each gait cycle were right (dark) lateral aspect of the neck. Trajectories were obtained from
control data of a typical subject who performed three trials in each
computed. The angle of the instantaneous velocity vector of condition.
head mid-point (xB,zB) with respect to the straight line (X
axis) in the horizontal plane defined heading of the body
(FB) at each point of the trajectory [12]: from the instructions given to subjects, body speed, reflected
FB tan 1 _zB =_xB . The orientation of the locomotor trajec- by the mean velocity of head mid-point, significantly
tory was measured as the mean heading value over the whole changed according to the gait style (gait effect,
gait cycle. This value exactly corresponds to the orientation of F(1,7) 113, p o 0.0001), i.e. subjects progressed faster when
the mean line on which the body moves during the gait cycle. running compared to walking. Nevertheless, neither the
The change in heading was defined positive for a left turn, i.e. mass nor the vibration conditions significantly influenced
the direction of vibration-induced deviations. the body speed (p 4 0.5 for both conditions).
Figure 2b shows average (all subjects considered) loco-
Statistical analysis: A within-subject ANOVA for re- motor trajectories performed in each gait, mass and
peated measures was used to evaluate significant differ- vibration conditions. Without neck muscle vibration, and
ences between experimental conditions. Post hoc differences for all the experimental conditions, blindfolded subjects
were assessed by means of Sheffes test. showed minor deviations from a straight line as revealed by
the near-zero mean heading values (mean average
0.74 7 0.831, all conditions intermingled, see Fig. 2c). On
RESULTS the other hand, when neck muscles were vibrated (right
The results of the control condition (Fig. 1) showed bilateral side), a systematic locomotor deviation toward the side
effects of neck muscle vibration on body steering. These effects opposite to the stimulation (leftward deviation) was
were more or less pronounced according to the gait performed, observed for all subjects, whatever the gait and mass
i.e. walking vs running. We elucidated the origin of these conditions. All subjects perceived a force pushing them
observations further by studying the effects of right neck leftward. Despite vibration-induced locomotor disorienta-
muscle vibration under different conditions of locomotion and tions during both walking and running, the amplitude
mass. Figure 2 depicts the principal results of this report. of body deviations appreciably depended on the gait
Figure 2a illustrates mean values of body speed under the performed. The statistical analysis yielded an interaction effect
different experimental conditions. As could be expected of vibration and gait conditions (F(1,7) 20.9, p o 0.005). A post

2366 Vol 14 No 18 19 December 2003

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
NECK MUSCLE VIBRATION DURING WALKING AND RUNNING NEUROREPORT

Fig. 2. (a) Average values (mean 7 s.d.) of mean body speed according to the experimental condition. (b) Average (all subjects and trials) locomotor
trajectories recorded when subjects crossed the recording eld of the motion capture system.The inset shows the conguration of the experimental set-
up (see Materials and Methods for further details). (c) Mean values of heading direction averaged over the time interval of a complete gait cycle. Hori-
zontal lines joined conditions between which signicant dierences were found.

hoc analysis revealed that the amplitude of vibration-induced
body deviations was significantly larger (p o 0.005) for walking
(21.6 7 4.61) than for running (8.0 7 2.51). All subjects consis-
tently showed the same gait-dependent behaviour. It is worth
noting that the deviations of the locomotor trajectories were
almost similar when the body mass increased ( + 20% of body
mass) or when the body mass remained unchanged (compare
grey with black traces in Fig. 2c) for both gait and vibration
conditions (the average difference between loaded and
unloaded conditions was 5.4%; p 4 0.4).
DISCUSSION
In the current study, we showed that the amplitude of
deviations induced by continuous neck muscle vibration
dramatically depends on the gait performed. Body devia-
tions decreased during running in comparison with walking
by 50% or more. These results expand those obtained in
previous studies, which found a sharp decrease in body
deviations when perturbing vestibular or visual sensory
systems during running with respect to walking [1,4,5].
A biomechanics-related hypothesis may account for the
differential effect of neck muscle vibration on walking vs
running [13]. Inertia of the moving body linearly increases
with speed, and consequently is much larger when running
than when walking. Increase in body inertia causes a
mechanical stabilisation of balance (gyroscopic effect) that
might explain why the amplitude of body deviations
decreased when running. This hypothesis, however, can
only partially explain the previous observation. Indeed,
when subjects walked or ran with an additional load (20% of
subjects body mass) while maintaining the same speed as
without load, their body inertia substantially increased but
their locomotor paths were marginally straightened (5.6%).
Our results indicate that gait-dependent effects of neck
muscle vibration during locomotion cannot solely be
attributed to biomechanical factors.
Neural mechanisms are more likely to explain the
experimental results of our study. During standing, neck
muscle vibration causes activation of lower leg muscles
70100 ms after stimulation onset, suggesting that a short-
latency integrative system mediates cervical influences on
postural control [14]. The differential motor effect elicited by
vibration during walking vs running is a strong suggestion
that integration of neck muscle afferents is substantially
modulated according to the gait style. This idea is consistent
with neurophysiological data which showed a down-
modulation of Ia axon terminals from leg muscles during
running with respect to walking [15]. Since an increase in
body velocity is accompanied by a change of gait, we cannot
definitively rule out that the speed of locomotion rather
than the gait style per se determines gating of neck input
effects on postural control. Indeed, subjects ran about 2.5
times faster than they walked, and showed almost the same
decrease in the amplitude of their body deviations. The
effect of body speed and gait style could be assessed by

Vol 14 No 18 19 December 2003 2 3 67

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 NEUROREPORT
eliciting body deviations when walking and running over a
wide range of speed, including overlapping velocities.
At this stage, it is premature to propose a neural substrate
for gait-dependent modulation. Nevertheless, a tentative
explanation can be based on the known sensory and motor
mechanisms that are involved in maintenance of body
balance. Gait-dependent modulation may occur at the level
of reticulospinal and vestibulospinal neurons since descend-
ing pathways originating in these neurons strongly con-
tribute to postural control during standing and walking
[16,17]. Visual, vestibular and neck muscle inputs widely
converge in reticular [18] and vestibular nuclei [19,20], thus
suggesting that this neural site is involved in gait-dependent
modulation of sensory input. Furthermore, we wonder
whether the cerebellum is responsible for the modulation of
sensory influence on locomotor movements. Manzoni et al.
[8] demonstrated that neck input regulates spatial and
temporal properties of corticocerebellar neurons which act
on reticulospinal and vestibulospinal neurons so as to adapt
postural control to actual body orientation with respect to
the direction of gravity [21].
Converging evidence from the present and previous
studies [1,4,5] suggests that a general motor control scheme
regulates the influence of inputs from head sensory
receptors according to the gait style. Physiological and
behavioural investigations have repeatedly shown a strong
interaction between neck proprioceptive, vestibular, and
visual sensory information in balance control. For example,
the direction of postural responses to vibration of neck
muscle depends on both gaze orientation and vestibular
activity [22]. Moreover, labyrinthine-defective patients show
no unbalance on vibration of the cervical muscles [23].
Such a multi-modal integration, which necessitates com-
plex sensorimotor transformations, may explain why a
down-modulation of sensory influences occurs when switch-
ing from walking to running. Indeed, sustained head
rotations during running compared to walking [11] generate
high-frequency discharge patterns from head sensory recep-
tors that must be integrated under short delay. Therefore, in
the condition of running, combination of neck, vestibular,
and visual inputs in a common reference frame for their
transformation into appropriate motor coordinates may
complicate neural control of locomotion [11]. We propose
that during running, the CNS suppresses potentially desta-
bilizing inputs from head sensory receptors [1] and opts for a
control mode mainly operating in feed-forward.
CONCLUSION
In the present study, we provided evidences that central
neural mechanisms down-modulate inputs from neck
muscles when running with respect to walking. This finding
highlights a general sensorimotor strategy by which the
CNS controls locomotion according to the gait performed.
We suggest that the computational complexity related to the
integration of sensory inputs originating in head receptors
Acknowledgements: This work was supported by grants from the Conseil Regional de Bourgogne, the Institute National pour la
Sante et la Recherche Medicale (INSERM), and the Centre National dEtudes Spatiales (CNES).G.C. is supported by a grant from the
French Minister of Research.
2368 Vol 14 No 18 19 December 2003
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
G. COURTINE ETAL.
when running may explain why such a gait-dependent
modulation of afferent control occurs.
REFERENCES
1. Brandt T, Strupp M and Benson J. You are better off running than walking
with acute vestibulopathy. Lancet 354, 746 (1999).
2. Bove M, Courtine G and Schieppati M. Neck muscle vibration and spatial
orientation during stepping in place in humans. J Neurophysiol 88,
22322241 (2002).
3. Bove M, Diverio M, Pozzo T and Schieppati M. Neck muscle vibration
disrupts steering of locomotion. J Appl Physiol 91, 581588 (2001).
4. Jahn K, Strupp M, Schneider E et al. Differential effects of vestibular
stimulation on walking and running. Neuroreport 11, 17451748 (2000).
5. Jahn K, Strupp M, Schneider E et al. Visually induced gait deviations
during different locomotion speeds. Exp Brain Res 141, 370374 (2001).
6. Fitzpatrick RC, Wardman DL and Taylor JL. Effects of galvanic vestibular
stimulation during human walking. J Physiol 517, 931939 (1999).
7. Ivanenko YP, Grasso R and Lacquaniti F. Neck muscle vibration makes
walking humans accelerate in the direction of gaze. J Physiol 525 pt 3,
803814 (2000).
8. Manzoni D, Pompeiano O, Bruschini L and Andre P. Neck input modifies
the reference frame for coding labyrinthine signals in the cerebellar
vermis: a cellular analysis. Neuroscience 93, 10951107 (1999).
9. Mergner T, Huber W and Becker W. Vestibular-neck interaction and
transformation of sensory coordinates. J Vestib Res 7, 347367 (1997).
10. Grasso R, Glasauer S, Takei Y and Berthoz A. The predictive brain:
anticipatory control of head direction for the steering of locomotion.
Neuroreport 7, 11701174 (1996).
11. Pozzo T, Berthoz A and Lefort L. Head stabilization during various
locomotor tasks in humans. I. Normal subjects. Exp Brain Res 82, 97106
(1990).
12. Courtine G and Schieppati M. Human walking along a curved path. I.
Body trajectory, segment orientation and the effect of vision. Eur J
Neurosci 18, 177190 (2003).
13. Dietz V, Baaken B and Colombo G. Proprioceptive input overrides
vestibulo-spinal drive during human locomotion. Neuroreport 12,
27432746 (2001).
14. Andersson G and Magnusson M. Neck vibration causes short-latency
electromyographic activation of lower leg muscles in postural reactions of
the standing human. Acta Otolaryngol 122, 284288 (2002).
15. Capaday C and Stein RB. Difference in the amplitude of the human
soleus H reflex during walking and running. J Physiol 392, 513522 (1987).
16. Orlovsky GN. Activity of vestibulospinal neurons during locomotion.
Brain Res 46, 99112 (1972).
17. Matsuyama K and Drew T. Vestibulospinal and reticulospinal neuronal
activity during locomotion in the intact cat. I. Walking on a level surface.
J Neurophysiol 84, 22372256 (2000).
18. Wilson VJ. Vestibulospinal reflexes and the reticular formation. Prog Brain
Res 97, 211217 (1993).
19. Gdowski GT and McCrea RA. Integration of vestibular and head
movement signals in the vestibular nuclei during whole-body rotation.
J Neurophysiol 82, 436449 (1999).
20. Gdowski GT and McCrea RA. Neck proprioceptive inputs to primate
vestibular nucleus neurons. Exp Brain Res 135, 511526 (2000).
21. Manzoni D, Pompeiano O and Andre P. Neck influences on the spatial
properties of vestibulospinal reflexes in decerebrate cats: role of the
cerebellar anterior vermis. J Vestib Res 8, 283297 (1998).
22. Ivanenko YP, Grasso R and Lacquaniti F. Effect of gaze on postural
responses to neck proprioceptive and vestibular stimulation in humans.
J Physiol 519, 301314 (1999).
23. Lekhel H, Popov K, Anastasopoulos D et al. Postural responses to
vibration of neck muscles in patients with idiopathic torticollis. Brain 120,
583591 (1997).