REVIEWS

PROPRIOCEPTION AND
LOCOMOTOR DISORDERS
Volker Dietz
Advances in our understanding of movement control allow us to define more precisely the
requirements for the rehabilitation of patients with movement disorders. Most purposeful,
complex movements are programmed in the central nervous system (CNS) and adapted by
proprioceptive feedback. The selection of and interaction between different sources of afferent
input is task dependent. Simple stretch reflexes are thought to be involved primarily in the control
of focal movement. For more complex motor behaviours such as locomotion, afferent input
related to load and hip-joint position probably has an important role in the proprioceptive
contribution to the activation pattern of the leg muscles. There is increasing evidence that
movement disorders such as spasticity and Parkinsons disease involve the defective use of
afferent input in combination with secondary compensatory processes. This has implications for
therapy, which should be directed to take advantage of the plasticity of the CNS.

CENTRAL PATTERN GENERATOR
A neural circuit that produces
self-sustaining patterns of
behaviour independently of
sensory input.
ParaCare, Institute for
Rehabilitation and Research,
University Hospital Balgrist,
Forchstrasse 340, CH 8008,
Zurich, Switzerland.
e-mail:
dietz@balgrist.unizh.ch
doi:10.1038/nrn939
The study of movement control has relevance to our
general understanding of brain function. But it also has
implications for specific fields, such as neurology,
cognitive neuroscience, rehabilitation medicine and
robotics. Our understanding of movement disorders
and their appropriate treatment depends on knowledge
of the neuronal mechanisms that underlie functional
movements. Movement disorders are the focus of one of
the most rapidly expanding fields in medicine, leading
to increasing costs of treatment and rehabilitation. This
review focuses on the role of proprioception during
human locomotion, which can serve as a paradigm for
functional movements.
In a more general sense, locomotion is representative
of movement control. It is a subconsciously performed,
everyday movement that is highly reproducible. It is
adapted automatically to existing conditions, such as
ground irregularities, within a large safety margin.
Knowledge about the neuronal control of human loco-
motion is also of broad interest for clinical reasons.
Characteristic disorders of locomotion are often the first
sign of a central or peripheral lesion of the motor system.
However, impaired movement is not only the direct
consequence of a central lesion, reflected in defective
programming or reflex functioning. Rather, a move-
ment disorder also reflects secondary compensatory
processes that are induced by the primary lesion. In
many cases, the altered motor behaviour can be consid-
ered as the optimal outcome for a given lesion of the
motor system1. The complexity of primary and sec-
ondary effects of a lesion means that detailed analysis of
a movement disorder is required to define the target
of any treatment.
Basic aspects of locomotion
It is generally accepted that locomotion in mammals
depends on neuronal circuits (networks of interneu-
rons) in the spinal cord (the CENTRAL PATTERN GENERATOR,
or CPG) that can act in the absence of any afferent
input2. Afferent information influences the central
(spinal) pattern and, conversely, the CPG selects appro-
priate afferent information according to external
requirements2. In addition, proprioceptive information
provides the basis for a conscious representation of our
body in space, which becomes severely disturbed in
deafferented individuals3. Both the spinal locomotor
centre (CPG) and the reflexes that mediate afferent
input to the spinal cord are under the control of the

NATURE REVIEWS | NEUROSCIENCE VOLUME 3 | OCTOBER 2002 | 7 8 1

REVIEWS

Table 1 | Proprioceptive reflexes suggested to be involved in locomotion

Appropriate Receptor Afferent Reflex Suggested function
stimulus pathway connection
Dynamic muscle Muscle spindles Group I Monosynaptic Compensation for ground
stretch (small (nuclear bag) stretch reflex irregularities; running/
amplitude) hopping (?)
Muscle stretch Muscle spindles Group II (III) Polysynaptic spinal Compensation for
(large amplitude; (nuclear chain) reflex perturbations of gait
static)
Change in bodys Golgi tendon Group Ib Polysynaptic; Control of bodys centre
centre of mass organs convergence of mass
spinal interneurons
Joint-movement Muscles around Group I, II Polysynaptic; Influence on locomotor
position joint; mechano- convergence pattern (hip); local
sensors of joint spinal interneurons compensation (other leg
capsule joints)
Skin Mechanosensors Group II (III) Polysynaptic; Adaptation to actual
deformation of skin convergence ground conditions
spinal interneurons
Noxious stimulus; Free endings; Group III, IV Spinal interneurons Withdrawal reflex
pressure Pacini corpuscles (CPG); flexor reflex
CPG, central pattern generator.

SPINALIZATION
Surgical separation of the spinal
cord from the brain.
SPINAL STRETCH REFLEX
Also known as the short-latency
reflex, this is the simplest reflex
known. Muscle stretch is
detected by muscle spindles, the
afferent (Ia) fibres of which
monosynaptically (and
oligosynaptically) excite the
motor neurons that innervate
the same muscle, leading to
muscle contraction.
brainstem4. In addition, there is phase-linked cortico-
spinal control of locomotion in humans57 and in other
mammals8,9. Voluntary commands have to interact with
the spinal locomotor generator to change, for example,
the direction of gait or to avoid an obstacle10,11. For most
other rhythmic elementary motor behaviours, such as
hopping or swimming, CPGs are also assumed to exist2.
Any disturbance of this finely coordinated interaction
between afferent inputs and pattern generation after a
central lesion, such as stroke or spinal cord injury, leads
to a movement disorder.
Adequacy of animal models. The nature of motor-
control mechanisms in humans can usually be deduced
only by indirect methods; such mechanisms are inferred
from knowledge obtained from animal experiments.
Consequently, it is not surprising that new treatments in
neurorehabilitation are frequently founded on basic
research in quadrupeds12. For example, research on
walking in cats has led to new therapies, such as loco-
motor training for patients with spinal cord injuries13,14.
For most of the basic mechanisms that underlie
locomotion, there seems to be no fundamental differ-
ence between bipeds and quadrupeds1,15,16. Essential
spinal neuronal mechanisms, such as the afferent
inputs that determine the locomotor pattern (includ-
ing hip-joint-related and load-receptor-related
inputs) are probably similar for quadrupedal and
bipedal locomotion.
Nevertheless, there are also differences in the quanti-
tative relationships between central mechanisms and
peripheral input. For example, the regulation of bipedal
gait requires specific neuronal mechanisms to maintain
the body in an upright position. Furthermore, the ability
of the isolated spinal cord to generate locomotor move-
ments is considerably greater in the SPINALIZED cat or rat7
than in the monkey17 or in humans14. These differences
might be due to a greater dominance of supraspinal
control over spinal neuronal circuits in humans.
Control of body equilibrium. A basic aspect of the
neuronal control of locomotion in both the quadru-
pedal cat and the bipedal human concerns the
anti-gravity function of the leg extensors. It has been
suggested that the SPINAL STRETCH REFLEX adapts the pre-
programmed motor patterns of leg muscles to the ter-
rain, and compensates for unexpected changes in
ground level18. Whereas this neuronal mechanism
explains quick, unilateral patterns of reflex activity in
leg extensor muscles, more complex bilateral coordi-
nation of leg muscle activation is needed to maintain
body equilibrium when gait is disturbed by an obsta-
cle. Irrespective of the conditions under which stance
and gait are investigated, the neuronal pattern that is
evoked during a particular task is always directed to
hold the bodys centre of mass over the base of support.
All control mechanisms must therefore be considered
and discussed in this respect. One consequence is that
the selection of afferent input by central mechanisms
must correspond to the requirements for body stabi-
lization. Neuronal signals of muscle stretch or length
are insufficient for the control of bipedal posture.
Only a combination of afferent inputs can provide
the information that is needed to control the bodys
equilibrium during locomotion.
Interactions between afferent inputs. The control of
locomotion involves the use of afferent information
from a variety of sources in the visual, vestibular and
proprioceptive systems. As a rule, spinal reflex pathways
and descending pathways converge on common spinal
interneurons to integrate these inputs19. For example,
visual feedforward information reduces the activity that
arises from the length sensors of muscles (the muscle
spindles)20. Furthermore, the amount of proprioceptive
feedback from the legs during various locomotor activi-
ties determines the influence of vestibulospinal input on
the stabilization of body movement21,22. Conversely,
somatosensory loss increases vestibulospinal sensitivity 23.

782 | OCTOBER 2002 | VOLUME 3 www.nature.com/reviews/neuro


H-REFLEX
Also known as the Hoffmann
reflex. The H-reflex results from
electrical stimulation of sensory
(Ia) fibres, which arise from
muscle spindles, leading (as in
the spinal stretch reflex) to a
monosynaptic excitation of
motor neurons that innervate
the same muscle.
NATURE REVIEWS | NEUROSCIENCE
REVIEWS
Reflex mechanisms the polysynaptic reflexes, integrates afferent inputs
The adaptation of the locomotor pattern to external from different sources.
demands is achieved by proprioceptive input that con-
tinuously modulates the programmed pattern during Monosynaptic reflexes. The monosynaptic reflex was
locomotion according to information from peripheral described in detail as early as 1922 (REF. 25) and repre-
sensors. Proprioceptors include receptors of the loco- sents the most extensively investigated proprioceptive
motor system that are located in the muscles and ten- reflex system. Monosynaptic reflex responses can be rec-
dons, as well as other mechanoreceptors in the joints ognized clearly by the characteristic short-onset latency
and the skin. The impulses of these receptors which of the EMG responses after muscle stretch in different
signal, for example, muscle stretch or tension are motor conditions.
conveyed to the spinal cord by afferent nerve fibres of The potential significance to locomotion of group
different diameters, and hence of different excitability Ia afferent input lies in the fact that its gain can be
and conduction velocity. On the basis of these dif- modulated by presynaptic inhibition26,27 (that is, type
ferences, the input from various types of receptor can Ia excitatory input to the motor neurons can be atten-
be separated into fibre types of group I to group IV uated, for example, by supraspinal influences) and by
(TABLE 1). For example, impulses from sensors of changes in muscle-spindle sensitivity28,29. This sensi-
dynamic muscle length (spindles) are mediated by tivity is controlled by the fusimotor system, which can
group Ia afferents, whereas those representing static vary the strength of activation of intrafusal muscle
muscle length are carried by group II fibres, and infor- fibres in muscle spindles by gamma motor neurons.
mation about tension developed at the tendons (mea- Muscle-spindle activity probably also contributes to
sured by Golgi organs) is transmitted by group Ib fibres an internally modelled reafference that combines with
to the spinal cord. Only the group Ia fibres have direct
excitatory connections to the motor neurons of the same
muscle they are part of the monosynaptic stretch
reflex, which has a characteristic short latency (~40 ms). a b
The other afferent fibres converge on spinal interneu-
rons that project in a more complex way to the motor
neurons of leg muscles; consequently, their afferent
input usually leads to responses in synergistic muscle
groups with a longer latency (starting at 7080 ms). This
pathway represents the polysynaptic, or long-latency,
reflex mechanism (TABLE 1).
For methodological reasons, there has been a bias
towards the simplest reflex system, the monosynaptic
+
stretch reflex, in experimental studies of the afferent +
sources that contribute to the regulation of human
+ +
gait. Most studies have focused on the contribution of
type Ia afferents using the H-REFLEX technique (which
measures the monosynaptic reflex response of the
muscle to low-intensity electrical stimulation of group
Ia afferents). However, as I discuss below, the contri-
bution of this reflex system to the regulation of loco-
motion is limited. The significance of, for example,
type Ib, II or III afferents to locomotor movements has
been underestimated.
In fact, there is multisensory afferent input during
locomotion both in the cat 2 and in humans8,9. This
proprioceptive input arises from muscles, skin, joints
and tendons. One of the primary functions of proprio-
ceptors is to detect unexpected events and to initiate
rapid compensatory electromyographic (EMG)
responses (TABLE 1). Recently, it has become clear that
proprioceptors have further roles for example, in Figure 1 | The stretch reflex of the calf muscles is
the regulation of motor output during unperturbed differentially gated during locomotion. Presumed action of
movements24. the short-latency stretch reflex during the stance (a) and swing
Three reflex systems that are involved in locomotion phase (b) of locomotion. a | This reflex system is facilitated by
are described in the following sections. For two reflex supraspinal influences on the pre- and postsynaptic elements
during the stance phase for the rapid compensation of ground
systems, the source of afferent input is clearly defined: irregularities. b | During the swing phase, the stretch-induced
the monosynaptic reflexes, which are mediated by reflex activity in the triceps surae, evoked by active dorsiflexion
group Ia afferents; and the cutaneous reflexes, which of the foot, is inhibited pre- and postsynaptically to allow proper
are mediated by skin afferents. The third reflex system, placing of the foot.
VOLUME 3 | OCTOBER 2002 | 7 8 3
REVIEWS
a Supraspinal b compensation for small ground irregularities. This is
Spinal
Figure 2 | Schematic drawing of the neuronal mechanisms involved in human gait.
a | Physiological condition. Leg muscles become activated by a programmed pattern that is
generated in spinal neuronal circuits (turquoise pathway). This pattern is modulated by
multisensory afferent input, which adapts the pattern to meet existing requirements. Both the
programmed pattern and the reflex mechanisms are under supraspinal control. In addition, there
is differential neuronal control of leg extensor and flexor muscles. Whereas extensors are mainly
activated by proprioceptive feedback, the flexors are predominantly under central control.
b | Proposed situation in Parkinsons disease. In this condition, a load-related impairment of
proprioceptive feedback can be assumed (dotted lines). This leads to reduced leg extensor
activation during the stance phase, which is poorly adapted to actual requirements (for example,
ground conditions). Modified, with permission, from REF. 4 1981 Elsevier Science.
motor commands to ensure the accuracy of body
movements30. These internal models also have impli-
cations for the normal and impaired control of
human locomotion.
During locomotion, the threshold and amplitude of
the soleus H-reflex is modulated over the entire step
cycle26. The functional implications of this modulation
of group Ia afferent input during locomotion are
suggested to be threefold (FIG. 1). First, facilitation of
the gastrocnemius/soleus stretch reflex at the end of the
stance phase contributes to compensation for ground
irregularities and assists during the push-off phase26,31.
Second, the depression of type Ia inputs to leg extensor
motor neurons during the swing phase prevents the
occurrence of the extensor stretch reflex during ankle
dorsiflexion26. And third, type Ia afferents are proposed
to have an important role in the continuous online
control of joint movements32.
Nevertheless, the functional significance of mono-
synaptic stretch reflexes during gait remains unclear,
largely for reasons related to the properties of the
monosynaptic stretch reflex itself28. The monosynaptic
reflex system is highly sensitive to small inputs, and its
function during gait should, therefore, be restricted to
784 | OCTOBER 2002 | VOLUME 3
consistent with the observation that leg muscle vibra-
tion, which excites type Ia afferent input, affects human
locomotion only a little32,33.
Cutaneous reflexes. Muscle responses induced by elec-
trical stimulation of a sensory nerve of a limb are medi-
ated by cutaneous reflexes. These reflex responses
appear in different muscles of the limb with a latency
that is compatible with a spinal pathway. In human leg
muscles, the task-dependency of cutaneous reflexes
has been shown in standing versus running34,35 or walk-
ing36, and in cycling versus static contraction37.
Cutaneous reflexes in leg muscles are sensitive to the
specific motor task that is performed, and undergo pro-
found modulation depending on the context in which
they are evoked38,39. Furthermore, cutaneous-reflex
modulation is nerve specific (that is, it depends on the
nerve that is stimulated), and this seems to be impor-
tant functionally40,41. Certain features of cutaneous-
reflex modulation, such as task-dependency, have been
suggested to include central-pattern-generating ele-
ments42,43. Such influences of the CPG on rhythmical
and cyclical leg movements would be in parallel with
observations made in the cat44.
Polysynaptic reflexes. It has been proposed that poly-
synaptic reflexes are mediated mainly by muscle
proprioceptive input from group II afferent fibres4547
(for example, from static muscle spindles or skin) and
possibly from group III fibres (carrying information
mainly from joint capsules and ligaments48) (FIG. 2a).
Polysynaptic spinal reflexes produce functionally useful
compensatory responses during locomotion, which are
more complex than simple stretch-reflex responses49.
As this pathway is polysynaptic, it allows the integra-
tion of inputs from muscle, joint and cutaneous afferents,
and convergence with commands from supraspinal cen-
tres to common spinal interneurons19. In addition, this
reflex system has excitatory and inhibitory connections
to both extensors and flexors50,51.
The sensory input determines the direction, velocity
and amplitude of the adjustment that is needed to
restore the subjects centre of gravity over the feet and
to generate the required pattern of leg muscle acti-
vation8,14,52,53. Consequently, this reflex system leads to
functional activation of synergistic muscle groups of
both legs, and it can clearly be separated from seg-
mental stretch-reflex responses, which affect only
individual muscles.
A polysynaptic pathway also mediates the effects of
flexor-reflex afferent (FRA) fibres54. Although the
modulation of flexion reflexes has several similarities
to the polysynaptic spinal reflexes discussed above34,51,55,
there are also distinct differences. First, in people with
spinal cord lesions, there is a loss of polysynaptic spinal
reflexes56, but the flexor reflex can still be elicited57.
Second, the characteristic feature of the flexor reflex is
to serve as a withdrawal reflex to noxious stimuli,
which is released as a direct response by the CPG58,
rather than in modulating the locomotor pattern to
www.nature.com/reviews/neuro

DECEREBRATE adapt to irregularities of the ground. And third, the
Describes an animal in which activity of spinal polysynaptic reflexes depends on the
the spine and hindbrain have presence of contact forces, whereas flexor reflex
been isolated surgically from
responses can be elicited by tibial nerve stimulation
higher cortical inputs.
independently of loading57.
Two main sources of afferent input are probably
integrated in the polysynaptic reflex system: load-related
and joint-position-related information. Load receptors,
or graviceptors, are thought to signal the influence of
gravity on the body to the spinal cord. There is only
indirect evidence for such receptors in humans. For
many years, the question of how the position of the
bodys centre of mass relative to the feet is signalled to
the CNS has been neglected in most studies of human
locomotion. To achieve appropriate gain control of pos-
tural reflexes, information is needed that signals the
influence of gravity on the body. This information is
insufficiently provided by muscle stretch receptors and
the vestibular system.
Essential sources of proprioception
During locomotion, multisensory proprioceptive feed-
back is continuously weighted and selected. This process
depends on the requirements of a particular locomotor
task and the availability of afferent input. According to
observations made in healthy subjects59,60, small chil-
dren (BOX 1) and patients with paraplegia61,62, afferent
inputs from load receptors and hip joints make essential
contributions to the activation pattern of leg muscles
during human locomotion.
It has been proposed that proprioceptive input from
extensor muscles, and probably also from mechano-
Box 1 | Developmental aspects
The innate pattern of locomotion, so-called newborn stepping, is characterized in
humans, as in most of our mammalian ancestors, by digitigrade stepping (in which the
toes reach the ground first). This initial pattern of locomotion usually disappears after
a few weeks. At around nine months of age, a new phase begins, during which the child
adapts the innate locomotor programme to the external conditions. These changes in
leg muscle activity can be interpreted as a shift away from a low level of motor control
(with functionally ineffective monosynaptic stretch reflexes) towards patterns that are
modulated by polysynaptic spinal reflexes89. This shift is biomechanically reflected in a
stick-like usage of the legs in the early stage of stepping, whereas in older children, the
body rolls over the standing leg109,110. This maturation of the locomotor pattern can be
regarded as a process in which descending inhibition of monosynaptic (or early)
reflexes and facilitation of polysynaptic spinal reflexes are established.
During infant stepping, the significance of load-receptor input to the locomotor
pattern becomes clear62. Transient loading of infants during the stance phase of gait
produces an increase in ankle extensor electromyographic activity and a delay in
tibialis anterior activity. In adults, loading leads only to increased extensor activity, but
not to a delay in the flexor activity burst111.
Afferent input from the hip joints is also important for appropriate leg muscle
activation in humans112,113, as it is in the cat75. If the hip is prevented from obtaining an
extended position in small children, the generation of the flexor burst and the onset of
the swing phase are inhibited.
Human infants can generate coordinated bilateral muscle activation well before the
onset of independent walking. An initiation of the swing phase on one side is
contingent on the contralateral limb being in the stance phase62,112,113. This is
consistent with the interlimb coordination that is observed in the cat81,114,115 and in
adult humans52,53.
NATURE REVIEWS | NEUROSCIENCE
REVIEWS
receptors in the sole of the foot, provides load informa-
tion63. The afferents that signal hip-joint position come
mainly from muscles around the hip. The role of this
afferent activity in rhythmic locomotion is to shape the
pattern, to control phase transitions and to reinforce
ongoing activity. Simple stretch and cutaneous reflexes
might be involved in compensating for irregularities
and in adapting to ground conditions.
Load-related afferent input. A potentially excitatory
function of load receptors during locomotion was first
described for the extensor muscles of the cat6466.
Extensor load receptors are also suggested to provide
essential information about the bodys centre of gravity
during locomotion in humans59,67,68. Experiments in the
cat indicated that these receptor signals might arise
from Golgi tendon organs and be carried by type Ib
afferents to the spinal locomotor generator.
In humans, the influence of load receptors on the
regulation of stance and gait became evident from
studies of infant stepping67 and of weightlessness
induced in adult humans either during space flight68 or
by water immersion59. Furthermore, it became clear
that body load has an effect on the magnitude of the
polysynaptic response, but not on the short-latency
stretch reflex in the gastrocnemius60,69,70. An influence
of load-receptor input on vestibular-evoked postural
responses was also described recently for asymmetrical
standing71 (unequal distribution of body load on the
two legs). It was suggested that there is a central inter-
action between load-related afferent input from the
periphery and descending signals.
The effect of load-receptor input on leg extensor
activation during the stance phase of gait might be
reinforced by heteronymous reflexes from ankle dorsi-
flexors72. One can assume that, during the stance phase
of locomotion, type Ib afferents from extensors inhibit
the flexors64. This is functionally meaningful because
the load on the stance limb must decrease before
swing can be initiated. In addition, extensor activity is
reinforced during the stance phase by positive feed-
back, which contributes to load compensation without
leading to instability73,74.
Joint-position-related afferent input. In the cat, there
are two main sources of afferent input that lead to
rhythm entrainment and/or resetting of locomotor
activity. Such input can either block or induce a switch
between the alternating flexor and extensor locomotor
bursts. The first of the afferent sources that satisfy these
criteria is related to load, whereas the second is related
to hip position63,75,76. For example, a locomotor rhythm
can be entrained by using rhythmic hip movements in
spinalized77 and DECEREBRATE78 cats. The afferents that
signal hip-joint position come mainly from muscles
that act around the hip. It has been suggested that
receptors of the hairy skin can also provide high-
fidelity information about knee-joint movements
in humans79.
In humans, observations made in infant stepping
and in the isolated spinal cord of people with paraplegia
VOLUME 3 | OCTOBER 2002 | 7 8 5
REVIEWS
TRANSCRANIAL MAGNETIC
STIMULATION
(TMS). A technique that is
used to induce a transient
interruption of normal activity
in a relatively restricted area of
the brain. It is based on the
generation of a strong magnetic
field near the area of interest,
which, if changed rapidly
enough, will induce an electric
field that is sufficient to
stimulate neurons.
DRIVEN GAIT ORTHOSIS
A motorized exoskeleton that is
applied to the legs of a person
with locomotor disorder for
example, paraplegia and
imposes physiological stepping
movements.
786 | OCTOBER 2002 | VOLUME 3
(see below) highlight the significance to locomotor
activity of receptor input from the hip joint. By con-
trast, the contribution of ankle-joint input to human
locomotion seems to be restricted to the stretched
muscles around the ankle joint31.
Differential control of flexors and extensors
In recent years, several studies have indicated that there
is differential control of leg flexor and extensor muscles
under functional conditions (FIG. 2a). Furthermore,
some observations (see below) indicate that there is
centrally determined dominance in the control of leg
flexor activity, but that proprioceptive input determines
extensor activation.
In humans, there are powerful presynaptic inhibitory
effects of flexor group I afferents on extensor group I
afferents, but these effects are weak from extensors to
flexors80. Corticospinal projections to lower-limb motor
neurons in humans are stronger to the flexor tibialis ante-
rior muscle than to the extensor soleus muscle81. In line
with this, the effect of TRANSCRANIAL MAGNETIC STIMULATION
of the leg area of the brain is restricted mainly to the
flexor muscles at distinct phases of swing during loco-
motion17. During human locomotion, gastrocnemius
EMG activity is modulated continuously by peripheral
afferent input, whereas the EMG of the tibialis anterior
is predominantly centrally determined18,52. In cat loco-
motion, weak static fusimotor effects are present in
extensors, but there are strong effects in flexor muscles82.
Finally, in recent models of locomotor control83, the
neuronal circuits that control leg flexor activity on both
sides reciprocally inhibit each other during walking,
whereas the extensor half-centres (spinal neuronal cir-
cuits that are responsible for the activation of leg exten-
sor muscles) are not directly coupled to each other.
Accordingly, flexor activity is an important source of
modulatory signals from spinal interneurons to the
motor cortex in humans84 and in cat85.
Locomotor capacity of the isolated spinal cord
Evidence for a spinal pattern generator for locomotion
in humans has come from studies of spontaneously
occurring step-like movements86 and from locomotor
movements induced on a treadmill with body-weight
support in people with incomplete and complete para-
plegia12,13,87. Load-related input has an important role in
inducing and training the locomotor pattern in these
patients13,61. However, the strength of leg muscle activa-
tion remains low compared with that of healthy sub-
jects, and the effects of training are lost over time in
people with complete paraplegia88. These results indi-
cate that there is training-induced plasticity of neuronal
centres in the isolated spinal cord that is dependent on
specific afferent input. This might be of relevance for
future interventional therapies (by combining training
with regeneration-inducing agents).
In a recent study, locomotor movements were
assisted by a DRIVEN GAIT ORTHOSIS (DGO) that allowed
stepping movements to be induced even with 100%
body unloading62. Under these conditions, physiologi-
cal locomotor-like movements alone did not lead to leg
CNS motor lesion
Loss of
supraspinal drive
Effect on spinal Effect on
reflex function muscle function
Hyperexcitability Loss of Altered mechanical
of short-latency long-latency muscle properties
reflexes reflexes
Spastic movement disorder
Figure 3 | Schematic of the mechanisms that contribute
to spastic paresis and spastic movement disorder.
A central motor lesion leads to changes in the excitability of
spinal reflexes and a loss of supraspinal drive. As a
consequence, changes in muscle properties occur. The
combination of all sequels of the primary lesion leads to the
spastic movement disorder. CNS; central nervous system.
muscle activation; this occurred only in combination
with loading of the legs. This was the case for both
healthy subjects and people with complete paraplegia
or tetraplegia. In addition, the DGO allowed the effects
of locomotor movements restricted to hip joints to be
studied in people with complete paraplegia62. An
important observation was that the pattern of leg muscle
activation was almost unchanged after knee-joint move-
ments were blocked in these patients. Furthermore, iso-
lated joint movements of the foot evoked only local
responses. This indicates that afferent input related to
hip-joint position has an important influence on leg
muscle activation by the isolated spinal cord.
Movement disorders
Any damage to the central or peripheral nervous system
can be followed by an impairment of proprioception
that leads to a movement disorder. Here, only the fre-
quently occurring movement disorders associated with
spasticity and Parkinsons disease are dealt with.
Spasticity. After a central motor (cerebral or spinal)
lesion, a profound alteration of proprioception occurs
in the form of a disinhibition of short-latency stretch
reflexes and the loss of functionally important long-
latency reflexes (FIG. 3). These changes are associated with
two forms of adaptation that can lead to an improve-
ment in mobility: the development of spastic muscle
tone, which compensates for part of the loss of supra-
spinal drive; and plasticity of spinal locomotor centres,
which can be specifically trained.
www.nature.com/reviews/neuro

The impaired supraspinal control of spinal reflexes
results in a loss of inhibition of short-latency reflexes,
which leads to hyperexcitability of the simple stretch
reflexes. This is combined with reduced facilitation of
the functionally more important polysynaptic, or long-
latency, reflexes, which leads to a reduced proprioceptive
contribution to leg muscle activation during gait89.
Therefore, spastic gait is associated with reduced and
less-well-modulated leg muscle activity. Consequently,
the development of tension in leg extensor muscles dur-
ing gait differs from that in healthy subjects and seems
to be independent of exaggerated stretch reflexes (FIG. 3).
Furthermore, after a central motor lesion, the modula-
tion of both cutaneous90 and short-latency stretch91
reflexes during the step cycle is impaired. As a conse-
quence, the fast regulation of motor neuron discharge,
which characterizes normal muscle activation, is
absent92,93. This contributes to the impairment of
walking ability in people with spasticity.
Although muscle spasticity is neural in origin, there
is good evidence that spastic muscles are abnormal
(FIG. 3). For example, reflex stiffness of ankle extensors
in people with spastic hemiparesis seems to be normal,
but the intrinsic muscle stiffness is significantly higher
than in control muscles91. Even the clinically assessed
muscle hypertonia in hemiparetic patients is found to
be associated primarily with muscle contracture and
less with exaggerated reflexes94. In line with these
observations, recent studies indicate that spasticity
results in a major alteration of the normal musclejoint
anatomical relationship95.
Changes in the mechanical properties of muscle
fibres can be attributed, at least in part, to muscle shorten-
ing. This might be the result of a decrease in the num-
ber of sarcomeres along the myofibrils, accompanied
by increased resistance to stretch94. These observations
highlight the need to develop new rational treatment
procedures to replace the anti-spastic drugs that are
frequently prescribed for mobile people with
spasticity 89. For example, by appropriate training pro-
grammes, compensatory processes could be guided in
a preferable way.
In conclusion, after a central motor lesion, motor
units are transformed in such a way that regulation of
muscle tone is achieved at a lower level of neuronal
organization, which in turn enables the patient to walk89.
Therefore, the altered regulation of spastic gait can be
considered as compensating for the loss of central
motor-system function1.
Parkinsons disease. Several reports have indicated that
force control in Parkinsons disease is impaired during
both voluntary movements96,97 and locomotion98.
Impaired load sensitivity develops with age and
becomes exaggerated in Parkinsons disease99. The con-
sequence of such an impaired load-receptor mechanism
is a reduction in leg extensor activation that is proposed
to contribute to the gait disorder98,99 (FIG. 2b). A similar,
although opposite, change in threshold or bias has been
described in cerebellar patients, who show hypermetric
postural responses during stance100.
NATURE REVIEWS | NEUROSCIENCE
REVIEWS
It is unclear how dysfunction of the basal ganglia
could affect load perception and loading responses
during gait. In this context, it is interesting to note
that the administration of levodopa (3,4-dihydroxy-
phenylalanine, or L-DOPA) or dopamine agonists is
associated with a depression of proprioception in
Parkinsons disease101.
The enhanced activation of the tibialis anterior dur-
ing the swing phase of gait in people with Parkinsons
disease might reflect defective extensor suppression of
the flexor burst-generating circuitry74. The control
of gait in these patients relies more on visual infor-
mation and is exerted through modulation of leg
flexor EMG activity79,102. This is in keeping with the
more general observation that visual information can
substitute to some extent for the reduced proprio-
ception24. A strong dependency of people with
Parkinsons disease on visual cues during walking
becomes evident when an optical flow pattern is
imposed during stepping on a treadmill 103. Whereas
the walking velocity of healthy subjects is affected for
only a short time by the pattern, people with
Parkinsons disease continuously change their speed
with the movements of the optical flow.
Reliability of clinical tests
Neurorehabilitation is one of the most rapidly expand-
ing fields in medicine, and this is leading to increasing
costs of various forms of treatment. Only recently have
studies been conducted to address the effects of re-
habilitation treatments on locomotion and functional-
level outcomes, especially in people with hemiparesis
after stroke and with Parkinsons disease. Nevertheless,
the assessment of isolated physical signs still prevails.
Therefore, an important aim for the future is to establish
standardized functional tests.
Physical signs and function. Movement disorders are
usually the first and most pronounced symptoms of an
impairment in motor-centre function. The physical
signs obtained during clinical examination, including
reflex excitability and muscle tone, can lead to a diagno-
sis, but give little information about the pathophysiol-
ogy, course and appropriate treatment of a movement
disorder. For example, stretch-reflex excitability and
muscle tone differ fundamentally between the passive
(clinical examination) and active (movement) condi-
tions. An appropriate treatment should not be cosmetic
(that is, correcting an isolated clinical parameter that
does not affect function). For adequate therapy of a
movement disorder, it is essential to understand and
analyse the function of the reflexes and motor centres
that are involved, and their interaction with the bio-
mechanical effector system in the motor task that is
impaired. Adapted behaviour emerges as an integration
of the biomechanical effector system and control
properties of the nervous system104.
This requires evaluation of the behaviour and func-
tion of neuronal (EMG) and biomechanical (joint
movement) measurements, as any changes in these sys-
tems might lead to a movement disorder. Nevertheless,
VOLUME 3 | OCTOBER 2002 | 7 8 7
REVIEWS

it is important to be aware that any change in a mea- a

50 20
surement might be secondary to or compensate for the
primary dysfunction of the motor system that is 40
15
involved in the impaired movement. The effect of any

Motor score
30

WISCI
treatment, either drugs or physical therapy, has to be 10
assessed on the basis of function. 20
Motor score
There are three requirements for appropriate eval- Locomotor function 5
10
uation of a therapeutic effect on function: that the (WISCI)
spontaneous recovery of function be separated from 0 0
25 36 39 71 83 179
the effect of any therapy; that the intensity and dura- Days after SCI
tion of a particular physical therapy, which strongly b
influences its effect105, be determined; and that 50 20
patients functional impairments be made comparable
40
by the use of internationally standardized scores for 15

Motor score
classification. 30

WISCI
10
20
Appropriate assessment of function. Owing to the
5
exquisite task-dependent regulation of nervous- 10
system function (discussed earlier), clinical tests must be
0 0
functional and specific. At present, it is a common and 17 29 41 55 69 83 98 104
accepted approach to score isolated clinical measures, Days after SCI
such as reflex excitability, muscle tone or voluntary force c
50 20
of single muscles. For example, muscle tone and spasm
frequency can be assessed by the Ashworth scale and the 40
15
Penn spasm-frequency scale, respectively106. For people

Motor score

with spinal cord injury, the American Spinal Injury
Association (ASIA) has developed a standardized
neurological assessment the ASIA classification of
motor and sensory deficits107. The question is, first,
whether such scoring systems can serve as a sensitive
outcome measure for new interventional therapies, and
second, whether they can reflect the functional impair-
ment, which is the most important aspect in terms of
the patients quality of life.
Only recently has a score been developed that
relates to function. Locomotor ability has been classi-
fied into 19 items108. An ongoing study indicates that a
close relationship between motor scores and locomo-
tor ability exists only in patients with moderately
impaired motor function. Patients with a low motor
score who undergo intensive locomotor training can
achieve improved locomotor function without a
change in motor score (FIG. 4 and V.D. et al., unpub-
lished observations). In these cases, relatively little vol-
untary force in the leg muscles (reflected in the ASIA
score) is required to acquire the ability to walk when
functional training is given.
For the future, the effectiveness of any new interven-
tional therapy should be assessed by internationally
accepted functional scores for upper- and lower-limb
movements in combination with motor scores of
selected limb muscles. Motor and sensory scores are
most likely to reflect a spontaneous recovery of func-
tion, as they depend on the integrity of corticospinal
connections. By contrast, improvement of locomotor
function after spinal cord injury also reflects the plastic-
ity of neuronal spinal centres below the level of the
lesion. With the combined assessment of voluntary
force and automatic function, the superiority of any
new interventional therapy on functional movements
might reliably be assessed.
WISCI
30
10
20
5
10
0 0
156 196 226 230 240 268 307
Days after SCI
Figure 4 | Data from three patients with incomplete
paraplegia after spinal cord injury. In three individual
patients (ac) undergoing locomotor training, there is a
differential course of motor score, which reflects the ability to
voluntarily contract selected muscles, and locomotor function
with respect to time. The motor score is assigned according to
American Spinal Injury Association standards, with a maximum
score of 50. The WISCI (walking index for spinal cord injury)
score ranges from 0 (no walking ability) to 19 (full walking
ability). SCI, spinal cord injury.
Conclusions
Effective rehabilitation after a central motor lesion
depends on the following points. First, knowledge
about the neuronal mechanisms that are involved in
the normal movement, and about the interactions
between the central programme and afferent inputs. It
is also necessary to take into account the possibility
that a movement disorder is the consequence not only
of the primary motor lesion, but also of secondary
processes that can be compensatory and should be
supported during rehabilitation. The aim of rehabilita-
tion should focus on an improvement of function by
taking advantage of the plasticity of neuronal centres,
rather than being directed towards the correction of
isolated clinical signs, such as reflex excitability. Finally,
to monitor the outcome and assess the effectiveness of
any interventional therapy, standardized functional
tests should be established.

788 | OCTOBER 2002 | VOLUME 3 www.nature.com/reviews/neuro


1. Latash, M. L. & Anson, J. G. What are normal movements
in atypical populations? Behav. Brain Sci. 19, 55106 (1996).
This paper addresses an important point concerning
the self-compensation of a motor deficit and the
associated problem of how far it is useful
cosmetically to correct secondary changes, such as
muscle tone.
2. Grillner, S. in Wenner-Gren International Symposium Series
Vol. 45. Neurobiology of Vertebrate Locomotion (eds
Grillner, S., Stein, P. S. G., Stuart, D. G., Forssberg, F. &
Herman, R. M.) 505512 (Macmillan, London, 1986).
3. Sanes, J. N., Mauritz, K.-H., Dalakas, M. C. & Evarts, E. V.
Motor control in humans with large-fiber sensory
neuropathy. Hum. Neurobiol. 4, 101114 (1985).
This paper describes the changes in motor
behaviour that are observed in deafferented people
(who lack proprioceptive feedback information
during movement).
4. Jankowska, E. & Lundberg, A. Interneurones in the spinal
cord. Trends Neurosci. 4, 230233 (1981).
5. Capaday, C., Lavoie, B. A., Barbeau, H., Schneider, C. &
Bonnard, M. Studies on the corticospinal control of human
walking. I. Responses to focal transcranial magnetic
stimulation of the motor cortex. J. Neurophysiol. 81,
129139 (1999).
6. Schubert, M., Curt, A., Colombo, G. & Berger, W. Voluntary
control of human gait: conditioning of magnetically evoked
motor responses in a precision stepping task. Exp. Brain
Res. 126, 583588 (1999).
7. Schubert, M., Curt, A., Jensen, L. & Dietz, V. Corticospinal
input in human gait: modulation of magnetically evoked
motor responses. Exp. Brain Res. 115, 234246 (1997).
The first evidence of corticospinal control of human
locomotor movement, restricted to tibialis anterior
activity at distinct phases of the swing.
8. Drew, T. Role of the motor cortex in the control of visually
triggered gait modifications. Can. J. Physiol. Pharmacol.
74, 426442 (1996).
9. Leblond, H., Menard, A. & Gossard, J. P. Corticospinal
control of locomotor pathways generating extensor activities
in the cat. Exp. Brain Res. 138, 173184 (2001).
10. Bosco, G. & Poppele, R. E. Proprioception from a
spinocerebellar perspective. Physiol. Rev. 81, 539568
(2001).
11. Dietz, V. Neurophysiology of gait disorders: present and
future applications. Electroencephalogr. Clin. Neurophysiol.
103, 333355 (1997).
12. Taub, E., Gitendra, U. & Elbert, T. New treatments in
neurorehabilitation founded on basic research. Nature Rev.
Neurosci. 3, 228236 (2002).
13. Barbeau, H. & Fung, J. The role of rehabilitation in the
recovery of walking in the neurological population. Curr.
Opin. Neurol. 14, 735740 (2001).
14. Dietz, V., Colombo, G., Jensen, L. & Baumgartner, L.
Locomotor capacity of spinal cord in paraplegic patients.
Ann. Neurol. 37, 574582 (1995).
15. Macpherson, J. M., Horak, F. B., Dunbar, C. D. C. & Dow,
R. S. Stance dependence on automatic postural
adjustments in humans. Exp. Brain Res. 78, 557566 (1989).
16. Nilsson, J. A., Thorstensson, A. & Halbertsma, J. Changes
in leg movements and muscle activity with speed of
locomotion and mode of progression in humans. Acta
Physiol. Scand. 123, 457475 (1985).
17. Vilensky, J. A. Locomotor behaviour and control in human
and non-human primates: comparison with cats and dogs.
Neurosci. Biobehav. Rev. 11, 263274 (1987).
18. Dietz, V., Quintern, J. & Sillem, M. Stumbling reactions in
man: significance of proprioceptive and pre-programmed
mechanisms. J. Physiol. (Lond.) 386, 149163 (1987).
19. Schomburg, E. D. Spinal sensory systems and their
supraspinal control. Neurosci. Res. 7, 265340 (1990).
20. Jones, K. E., Wessberg, J. & Vallbo, A. Proprioceptive
feedback is reduced during adaptation to a visuomotor
transformation: preliminary findings. Neuroreport 12,
40294033 (2001).
21. Brandt, T., Strupp, M. & Benson, J. You are better off
running than walking with acute vestibulopathy. Lancet 35,
746 (1999).
This study provided the first evidence that the
influence of vestibulospinal drive depends on the
speed of locomotion.
22. Dietz, V., Baaken, B. & Colombo, G. Proprioceptive input
overrides vestibulo-spinal drive during human locomotion.
Neuroreport 12, 27432746 (2001).
23. Horak, F. B. & Hlavacka, F. Somatosensory loss increases
vestibulospinal sensitivity. J. Neurophysiol. 86, 575585
(2001).
This study provided evidence of automatic
compensation of impaired proprioception by other
balance-regulating systems.
NATURE REVIEWS | NEUROSCIENCE
24. Pearson, K. G. Motor system. Curr. Opin. Neurobiol. 10,
649654 (2000).
25. Hoffmann, P. Die Eigenreflexe Menschlicher Muskeln
(Springer, Berlin, 1922).
26. Capaday, C. & Stein, R. B. Difference in the amplitude of
the human soleus H-reflex during walking and running.
J. Physiol. (Lond.) 392, 513522 (1987).
27. Morin, C., Katz, R., Mazires, L. & Pierrot-Deseilligny, E.
Comparison of soleus H-reflex facilitation at the onset of
soleus contractions produced voluntarily and during the
stance phase of human gait. Neurosci. Lett. 33, 4753
(1982).
28. Matthews, P. B. C. Mammalian Muscle Receptors and Their
Central Actions (Arnold, London, 1972).
29. Loeb, G. E. & Hoffer, J. A. Activity of spindle afferents from
cat anterior thigh muscles. II. Effects of fusimotor blockade.
J. Neurophysiol. 54, 565577 (1985).
30. Lackner, J. R. & DiZio, P. A. Aspects of body self-calibration.
Trends Cogn. Sci. 4, 279288 (2000).
31. Sinkjaer, T., Anderson, J. B. & Larsen, B. Soleus stretch
reflex modulation during gait in humans. J. Neurophysiol.
75, 11121120 (1996).
32. Verschueren, S. M. P., Swinnen, S. P., Desloovere, K. &
Duysens, J. Effects of tendon vibration on the
spatiotemporal characteristics of human locomotion. Exp.
Brain Res. 143, 231239 (2002).
33. Ivanenko, Y. P., Grasso, R. & Lacquantini, F. Influence of leg
muscle vibration on human walking. J. Neurophysiol. 84,
17371747 (2000).
34. Duysens, J., Tax, A. A. M., Trippel, M. & Dietz, V.
Increased amplitude of cutaneous reflexes during human
running as compared to standing. Brain Res. 613, 230238
(1993).
35. Tax, A. A., van Wezel, B. M. H. & Dietz, V. Bipedal reflex
coordination to tactile stimulation of the sural nerve during
human running. J. Neurophysiol. 73, 19471964 (1995).
36. Komiyama, T., Zehr, E. P. & Stein, R. B. Absence of nerve
specificity in human cutaneous reflexes during standing.
Exp. Brain Res. 133, 267272 (2000).
37. Zehr, E. P., Hesketh, K. L. & Chua, R. Differential regulation
of cutaneous and H-reflexes during leg cycling in humans.
J. Neurophysiol. 85, 11781184 (2001).
38. Bastiaanse, C. M., Duysens, J. & Dietz, V. Modulation of
cutaneous reflexes by load receptor input during human
walking. Exp. Brain Res. 135, 189198 (2000).
39. Zehr, E. P. & Stein, R. B. What functions do reflexes serve
during human locomotion? Prog. Neurobiol. 5, 185205
(1999).
40. van Wezel, B. M., Ottenhoff, F. A. & Duysens, J. Dynamic
control of location-specific information in tactile cutaneous
reflexes from the foot during human walking. J. Neurosci.
17, 38043814 (1997).
41. Zehr, E. P., Stein, R. B. & Komiyama, T. Function of sural
nerve reflexes during human walking. J. Physiol. (Lond.)
507, 305314 (1998).
42. Duysens, J. & van de Crommert, H. W. A. A. Neural control
of locomotion. Part 1: the central pattern generator from
cats to humans. Gait Posture 7, 131141 (1998).
43. Brooke, J. D. et al. Sensori-sensory afferent conditioning
with leg movement: gain control in spinal reflex and
ascending paths. Prog. Neurobiol. 51, 393421 (1997).
44. Drew, T. & Rossignol, S. A kinematic and electromyographic
study of cutaneous reflexes evoked from the forelimb of
unrestrained walking cats. J. Neurophysiol. 57, 11601184
(1987).
45. Berger, W., Dietz, V. & Quintern, J. Corrective reactions to
stumbling in man: neuronal co-ordination of bilateral leg
muscle activity during gait. J. Physiol. (Lond.) 357, 109125
(1984).
46. Lundberg, A., Malmgren, K. & Schomburg, E. D. Reflex
pathway from group II muscle afferents. 3. Secondary
spindle afferents and the FRA: a new hypothesis. Exp. Brain
Res. 65, 294306 (1987).
47. Nardone, A., Grasso, M., Giordano, A. & Schiepatti, M.
Different effect of height on latency of leg and foot short- and
medium-latency EMG responses to perturbation of stance
in humans. Neurosci. Lett. 206, 8992 (1996).
48. Hasan, Z. & Stuart, D. G. Animal solutions to problems of
movement control: the role of proprioceptors. Annu. Rev.
Neurosci. 1, 199223 (1988).
49. Hansen, P. D., Woollacott, M. H. & Debu, B. Postural
responses to changing task conditions. Exp. Brain Res.
73, 627636 (1988).
50. Duysens, J., Trippel, M., Horstmann, G. A. & Dietz, V. Gating
and reversal of reflexes in ankle muscles during human
walking. Exp. Brain Res. 82, 351358 (1990).
51. Duysens, J., Tax, A. A. M., Murrer, L. & Dietz, V. Backward
and forward walking use different patterns of phase-
dependent modulation of cutaneous reflexes in humans.
J. Neurophysiol. 76, 301310 (1996).
REVIEWS
52. Dietz, V., Horstmann, G. A. & Berger, W. Interlimb
coordination of leg muscle activation during perturbation of
stance in humans. J. Neurophysiol. 62, 680693 (1989).
53. Dietz, V. Human neuronal control of automatic functional
movements: interaction between central programs and
afferent input. Physiol. Rev. 72, 3369 (1992).
54. Lundberg, A. Multisensory control of spinal reflex pathways.
Prog. Brain Res. 50, 1228 (1979).
55. Yang, J. F. & Stein, R. B. Phase-dependent reflex reversal in
human leg muscles during walking. J. Neurophysiol. 63,
11091117 (1990).
56. Berger, W., Horstmann, G. & Dietz, V. Tension development
and muscle activation in the leg during gait in spastic
hemiparesis: the independence of muscle hypertonia and
exaggerated stretch reflexes. J. Neurol. Neurosurg.
Psychiatry 47, 10291033 (1984).
57. Hiersemenzel, L., Curt, A. & Dietz, V. From spinal shock to
spasticity. Neuronal adaptations to a spinal cord injury.
Neurology 54, 15741582 (2000).
58. Bussel, B. et al. Myoclonus in a patient with spinal
transection. Possible involvement of the spinal stepping
generator. Brain 11, 12351245 (1988).
One of the first indications that the isolated human
spinal cord contains neuronal networks that can
generate locomotor-like EMG activity.
59. Dietz, V., Horstmann, G. A., Trippel, M. & Gollhofer, A.
Human postural reflexes and gravity an underwater
simulation. Neurosci. Lett. 106, 350355 (1989).
60. Dietz, V., Gollhofer, A., Kleiber, M. & Trippel, M. Regulation of
bipedal stance: dependence on load receptors. Exp. Brain
Res. 89, 229231 (1992).
61. Harkema, S. J. et al. Human lumbosacral spinal cord
interprets loading during stepping. J. Neurophysiol. 77,
797811 (1997).
62. Dietz, V., Mller, R. & Colombo, G. Locomotor activity in
spinal man: significance of afferent input from joint and load
receptors. Brain (in the press).
63. Dietz, V. & Duysens, J. Significance of load receptor input
during locomotion: a review. Gait Posture 11, 102110
(2000).
64. Duysens, J. & Pearson, K. G. Inhibition of flexor burst
generation by loading extensor muscles in walking cats.
Brain Res. 187, 321332 (1980).
The first demonstration that loading strongly
influences locomotor activity in the walking cat.
65. Conway, B. A., Hultborn, H. & Kiehn, O. Proprioceptive input
resets central locomotor rhythm in the spinal cat. Exp. Brain
Res. 68, 643656 (1987).
66. Pearson, K. G. & Collins, D. F. Reversal of the influence
of group Ib-afferents from plantaris on activity in medial
gastrocnemius muscle during locomotor activity.
J. Neurophysiol. 70, 10091017 (1993)
67. Yang, J. F., Stephens, M. J. & Vishram, R. Transient
disturbances to one limb produce coordinated, bilateral
responses during infant stepping. J. Neurophysiol. 79,
23292337 (1998).
68. Clement, G., Gurfinkel, V. S., Lestienne, F., Lipshits, M. J. &
Popov, K. E. Changes of posture during transient
perturbations in microgravity. Aviat. Space Environ. Med. 56,
666671 (1985).
69. Fouad, K., Bastiaanse, C. M. & Dietz, V. Reflex adaptations
during treadmill walking with increased body load. Exp.
Brain Res. 137, 133140 (2001).
70. Sinkjaer, T., Andersen, J. B., Ladoucoer, M., Christensen,
L. O. D. & Nielsen, J. B. Major role for sensory feedback in
soleus EMG activity in the stance phase of walking in man.
J. Physiol. (Lond.) 523, 817827 (2000).
71. Marsden, J. F., Castellote, J. & Day, B. L. Bipedal distribution
of human vestibular-evoked postural responses during
asymmetrical standing. J. Physiol. (Lond.) 542, 323331
(2002).
72. Marchand-Pauvert, V. & Nielsen, J. B. Modulation of
heteronymous reflexes from ankle dorsiflexors to hamstring
muscles during human walking. Exp. Brain Res. 142,
402408 (2002).
73. Prochazka, A., Gillard, D. & Bennet, D. J. Positive force
feedback control of muscles. J. Neurophysiol. 77,
32263236 (1997).
74. Prochazka, A., Gillard, D. & Bennett, D. J. Indications
of positive feedback in the control of movement.
J. Neurophysiol. 77, 3751 (1997).
References 73 and 74 provide convincing evidence
for movement control by positive force feedback
in the cat.
75. Pearson, K. G. Proprioceptive regulation of locomotion.
Curr. Opin. Neurobiol. 5, 786791 (1995).
76. Whelan, P. Control of locomotion in the decerebrate cat.
Prog. Neurobiol. 49, 481515 (1996).
77. Andersson, O. & Grillner, S. Peripheral control of the cats
step cycle. II. Entrainment of the central pattern generators
VOLUME 3 | OCTOBER 2002 | 7 8 9
REVIEWS
for locomotion by sinusoidal hip movements during fictive
locomotion. Acta Physiol. Scand. 118, 229239 (1983).
78. Kriellaars, D. J., Brownstone, R. M., Noga, B. R. &
Jordan, L. M. Mechanical entrainment of fictive locomotion
in the decerebrate cat. J. Neurophysiol. 71, 20742086
(1994).
79. Edin, B. Cutaneous afferents provide information about knee
joint movements in humans. J. Physiol. (Lond.) 531,
289297 (2001).
80. Iles, J. F. & Roberts, R. C. Inhibition of monosynaptic reflexes
in the human lower limb. J. Physiol. (Lond.) 385, 6987
(1987).
81. Brower, B. & Ashby, P. Corticospinal projections to lower
limb motoneurons in man. Exp. Brain Res. 89, 649654
(1992).
This paper provided a physiological basis for the
differential neuronal control of lower-limb muscles.
82. Loeb, G. E. The control and responses of mammalian
muscle spindles during normally executed tasks. Exerc.
Sport Sci. Rev. 12, 157204 (1984).
83. Hiebert, G. W. et al. Contribution of hind limb flexor muscle
afferents to the timing of phase transitions in the cat step
cycle. J. Neurophysiol. 75, 11261137 (1996).
84. Dietz, V., Horstmann, G. A. & Berger, W. Perturbations of
human posture: influence of impulse modality on EMG
responses and cerebral evoked potential. J. Motor Behav.
21, 357372 (1989).
85. Beloozerova, I. N. & Sirota, M. G. in Stance and Motor,
Facts and Concepts (eds Gurfinkel, V. S., Joffe, M. E.,
Massion, J. & Roll, J. P.) 163176 (Plenum, New York,
1988).
86. Calancie, B. et al. Voluntary stepping after chronic spinal
cord injury. Evidence for a central rhythm generator for
locomotion in man. Brain 117, 11431159 (1994).
87. Dobkin, B. H., Harkema, S., Requejo, P. & Edgerton, V. R.
Modulation of motor-like EMG activity in subjects with
complete and incomplete spinal cord injury. J. Neurol.
Rehabil. 9, 183190 (1995).
88. Wirz, M., Colombo, G. & Dietz, V. Long term effects of
locomotor training in spinal man. J. Neurol. Neurosurg.
Psychiatry 71, 9396 (2001).
89. Dietz, V. Spinal cord lesion: effects and perspectives for
treatment. Neural Plast. 8, 8390 (2001).
90. Jones, C. A. & Yang, J. F. Reflex behavior during walking in
incomplete spinal-cord-injured subjects. Exp. Neurol. 128,
239248 (1994).
91. Sinkjaer, T. & Magnussen, I. Passive intrinsic and reflex-
mediated stiffness in the ankle extensors of hemiplegic
patients. Brain 117, 355363 (1994).
92. Dietz, V., Ketelsen, U. P., Berger, W. & Quintern, J. Motor unit
involvement in spastic paresis: relationship between leg
muscle activation and histochemistry. J. Neurol. Sci. 75,
89103 (1986).
790 | OCTOBER 2002 | VOLUME 3
93. Rosenfalck, A. & Andreassen, S. Impaired regulation of force
and firing pattern of single motor units in patients with
spasticity. J. Neurol. Neurosurg. Psychiatry 43, 907916
(1980).
94. ODwyer, N. J., Ada, L. & Neilson, P. D. Spasticity and
muscle contracture following stroke. Brain 119, 17371749
(1996).
This paper provided clear evidence that increased
muscle tone in spastic hemiparesis is due, in large
part, to muscle contracture.
95. Lieber, R. L. & Frieden, J. Spasticity causes a fundamental
rearrangement of musclejoint interaction. Muscle Nerve 25,
265270 (2002).
96. Delwaide, P. J., Pepin, J. L. & Maertens de Noordhout, A.
Short-latency autogenic inhibition in patients with
parkinsonian rigidity. Ann. Neurol. 30, 8389 (1991).
97. Stelmach, G. E. in Tutorials in Motor Neuroscience (eds
Requin, J. & Stelmach, G. E.) 137148 (Kluwer, Dordrecht,
The Netherlands, 1991).
98. Dietz, V., Zijlstra, W., Prokop, T. & Berger, W. Leg muscle
activation during gait in Parkinsons disease: adaptation and
interlimb coordination. Electroencephalogr. Clin.
Neurophysiol. 97, 408415 (1995).
99. Dietz, V. & Colombo, G. Influence of body load on the gait
pattern in Parkinsons disease. Mov. Disord. 13, 255261
(1998).
100. Horak, F. B. & Diener, H. C. Cerebellar control of postural
scaling and central set in stance. J. Neurophysiol. 72,
479486 (1994).
101. OSuilleabhain, P., Bullard, J. & Dewey, R. B. Proprioception
in Parkinsons disease is acutely depressed by
dopaminergic medications. J. Neurol. Neurosurg. Psychiatry
71, 607610 (2001).
102. Dietz, V., Schubert, M. & Trippel, M. Visually-induced
destabilization of human stance: neuronal control of leg
muscles. Neuroreport 3, 449452 (1992).
103. Prokop, T. & Berger, W. Influence of optic flow on locomotion
in normal subjects and patients with Parkinsons disease.
Electroencephalogr. Clin. Neurophysiol. 99, 402 (1996).
104. Taga, G. A model of the neuro-musculo-skeletal system for
human locomotion. II. Real-time adaptability under various
constraints. Biol. Cybern. 73, 113121 (1995).
105. Kwakkel, G., Wagenaar, R. C., Twisk, J. W. R., Lankhorst, G.
J. & Koetsier, J. C. Intensity of leg and arm training after
primary middle-cerebral-artery stroke: a randomised trial.
Lancet 354, 191196 (1999).
106. Priebe, M. M., Sherwoord, A. M., Thornby, J. I., Kharas, N. F.
& Markowski, J. Clinical assessment of spasticity in spinal
cord injury: a multidimensional problem. Arch. Phys. Med.
Rehabil. 77, 713716 (1996).
107. Maynard, F. M. et al. International standards for neurological
and functional classification of spinal cord injury. Spinal Cord
35, 266274 (1997).
108. Ditunno, J. F. et al. Walking index for spinal cord injury
(WISCI): an international multicenter validity and reliability
study. Spinal Cord 38, 234243 (2000).
109. Berger, W., Altenmller, E. & Dietz, V. Normal and impaired
development of childrens gait. Hum. Neurobiol. 3, 163170
(1984).
110. Sutherland, D. H., Olshen, R., Cooper, L. & Woo, S. L. Y.
The development of mature gait. J. Bone Joint Surg. 62,
336353 (1980).
111. Stephens, M. J. & Yang, J. F. Loading during the stance
phase of walking in humans increases the extensor EMG
amplitude, but does not change the duration of step cycle.
Exp. Brain Res. 12, 363370 (1999).
112. Pang, M. Y. & Yang, J. F. The initiation of the swing phase in
human infant stepping: importance of hip position and leg
loading. J. Physiol. (Lond.) 528, 389404 (2000).
This study provided the first evidence in children of
the significance of hip position and load-receptor
input to the locomotor pattern.
113. Pang, M. Y. & Yang, J. F. Interlimb co-ordination in human
infant stepping. J. Physiol. (Lond.) 53, 617625 (2001).
114. Gorassini, M. A., Prochazka, A., Hiebert, G. W. & Gauthier,
M. J. A. Corrective responses to loss of ground support
during walking. I. Intact cats. J. Neurophysiol. 71, 603610
(1994).
115. Schomburg, E. D., Petersen, N., Barajon, I. & Hultborn, H.
Flexor reflex afferents reset the step cycle during fictive
locomotion in the cat. Exp. Brain Res. 122, 339350 (1998).
Acknowledgements
I thank B. Ghwiler for helpful comments and R. Jurd for correcting
the English. The Swiss National Science Foundation and the
International Research Institute of Paraplegia supported this work.
Online links
DATABASES
The following terms in this article are linked online to:
OMIM: http://www.ncbi.nlm.nih.gov/Omim/
Parkinsons disease
FURTHER INFORMATION
American Spinal Injury Association: http://www.asia-
spinalinjury.org/
Encyclopedia of Life Sciences: http://www.els.net/
bipedalism | central pattern generators | locomotion |
motor system organization | nervous control of movement |
proprioceptive sensory feedback | spinal reflexes |
tetrapod walking and running
Volker Dietzs lab:
http://www.neuroscience.unizh.ch/e/groups/dietz00.htm
Access to this interactive links box is free online.
www.nature.com/reviews/neuro