Ipsilateral Hindlimb Research Articles

Postinhibitory rebound facilitation of the extension reflex in the high spinal cat.

Neural mechanisms of the reflex rebound of the hindlimb were studied in the high spinalized cat. After the animal had recovered from spinal shock, a light touch-pressure on the foot pad elicited a rebound extension reflex of the ipsilateral hindlimb. Excitability changes in hindlimb motor nuclei during the rebound phenomenon were examined by recording test monosynaptic reflexes (MSR). During touch-pressure stimulation, flexors such as the biceps femoris and tibialis anterior were facilitated, while extensors such as the gastrocnemius and quadriceps femoris were inhibited. After cessation of a stimulus, extensor MSRs showed a rebound facilitation which lasted for a few seconds. The greater the preceding suppression of MSR, the greater the rebound facilitation. Membrane potential changes induced in extensor motoneurons by touch-pressure were investigated by intracellular recordings. In most of the motoneurons tested (21/24), a clear membrane hyperpolarization was observed during stimulation and was followed by a rebound firing on cesstion of stimulation. Possible mechanisms underlying the rebound extension reflex were discussed.

Somatotopic organization of climbing fiber projections from low threshold cutaneous afferents to pars intermedia of cerebellar cortex in the cat

The fine detail of climbing fiber projections to large areas of the pars intermedia of lobule V was demonstrated by means of low threshold natural cutaneous stimulation. These projections formed a complex mediolateral organization of patches that were elongated in the anteroposterior direction. In general, distal forelimb was represented medially and the face represented laterally, although there were also elongated patches of cells that did not respond to any cutaneous stimulus presentation. The entire ipsilateral anterior quadrant of the body was represented in lobules Vb, c, d with the addition of face projections which extended slightly across the midline. Ipsilateral hindlimb patches were observed in lobule Va. Although the boundary between patches was quite sharp, a slight overlap of the patches within the mosaic was often observed. In these areas of overlap, some cells possessed receptive fields that encompassed, either continuously or discontinuously, cutaneous areas of both the adjacent patches. It is proposed that the cerebellar cortex could correlate event timing in the climbing fiber patches with 'on-line' information relating the parameters of movement in the granule cell parallel fiber system, and that the areas described in the present study could mediate 'spatially organized and skilled movements' in the animal's repertoire, which involve paw-face-mouth interaction, such as feeding, cleaning and grooming.

Reaction of the mutant mouse nude to axonal injuries of the central nervous system

The effects of unilateral forebrain and spinal cord hemisections were examined in the mutant athymic mouse nude. The activities of tyrosine hydroxylase and glutamic acid decarboxylase within the striatum and olfactory tubercle and within the midbrain ventral tegmentum, terminal fields of dopaminergic and GABA-ergic neurons, respectively, were measured to characterize the anterograde reaction of these neurons to axonal injuries. After forebrain hemisections a rapid (within 2 days) and permanent decrease was observed for all enzymes, indicating a permanent degeneration of axons with no recovery. Unilateral spinal cord hemisections resulted in a flaccid paralysis of the ipsilateral hind limb which subsided within 2 to 3 days after the lesion, and thereafter a permanent spasticity ensued. When spinal cord hemisection was repeated the initial flaccid paralysis did not occur and spasticity of the ipsilateral hind limb was immediately evident. The results indicate that regeneration of cut axons or even an improved motor recovery does not occur in the central nervous system of the nude mouse.

Neurones in cat gracile nucleus with both local and widefield inputs.

1. Forty-three neurones were isolated in the cat gracile nucleus that could be driven by electrical stimulation of the ipsilateral forefoot or the contralateral hind food as well as having a normal low threshold localized receptive field on the ipsilateral hind limb. 2. Twenty-five (58%) of the cells were found to have axons projecting to the contralateral ventrobasal thalamus. 3. Most of the cells could only be driven from the 'widefield' receptive field on the forefoot or contralateral hindfoot by percutaneous electrical stimulation. 4. These results are discussed in the context of a model of the gracile nucleus in whick these occasional 'widefield' connexions are considered to be errors in connectivity which are not normally effective due to the patterns of excitation and inhibition in the normally functioning nucleus. 5. In this model, electrical stimulation is an abnormal type of stimulation that can drive cells through these erroneous connexions.

Suppressive Effect of Morphine on Single-unit Activity of Cells in Rexed Lamina VII

Using an extracellular microelectrode recording technique, the effects of intravenously administered morphine sulfate upon the single-unit activities of cells in Rexed lamina VII of the lumbar spinal cord were studied in cats following decerebration and spinal cord transection at L1. These neurons responded principally to high-threshold mechanical and thermal stimuli applied to a receptive field in the ipsilateral hind limbs. Morphine sulfate, 0.25, 0.5, and 1.0 mg/kg, caused dose-related suppression of the spontaneous activities of these neurons. The firing rates at maximum suppression, observed 5-10 min after administration of morphine, were 63.9 +/- 9.2 (mean +/- 1 SE), 43.0 +/- 5.4, and 26.5 +/- 6.0 per cent of the control values, respectively. Since these cells have been shown by others to be associated with the spinothalamic and spinoreticular pathways, the results suggest that the analgesic state may result from the action of morphine on the cells of origin of these major ascending pathways in the spinal cord.

The ventral spino-olivocerebellar system in the cat. I. Identification of five paths and their termination in the cerebellar anterior lobe.

1. The spino-olivocerebellar paths ascending through the ventral funiculus (VF-SOCPs) and projecting to the cerebellar anteior lobe were investigated in cats with the spinal cord transected in the third cervical segment sparing only the ventral funiculus on one side. The climbing fibre responses evoked in Purkinje cells by limb nerve stimulation were studied by recording the mass activity at the cerebellar surface or in the molecular layer. 2. Five VF-SOCPs were distinguished on the basis of their receptive fields, response latencies, and projection areas. 3. The projection areas consist of narrow sagittal zones. Three zones (a, b1 and b2) lie in the vermis and extend throughout lobules II--V. Two zones (c1 and c3) lie in the pars intermedia and are restricted to the classical hindlimb area, lobules II--IV. The VF-SOCPs are labelled according to their termination zones: a-VF-SOCP, b1-VF-SOCP, etc. 4. The a-, c1- and c3-paths are activated from the ipsilateral hindlimb, whereas the b1- and b2-paths are activated bilaterally from the forelimbs and hindlimbs, respectively. The latencies of the responses evoked from the ipsilateral hindlimb are relatively short for the c1-path and successively longer for the c3-, b2- and a-paths. 5. The olivary transmission showed fluctuations in efficacy independent for the different VF-SOCPs. The effect of anaesthetics on this transmission also differed between the paths. 6. It is concluded that the five VF-SOCPs relay in different compartments of the inferior olive which are tentatively identified.

The ventral spino-olivocerebellar system in the cat III. Functional characteristics of the five paths

1. Five spino-olivocerebellar paths ascending through the ventral funiculus (VF-SOCPs) with different projection zones in the cerebellar cortex have been distinguished (Oscarsson and Sjöuld, 1977). The paths are denoted the a-, b1-, b2-, c1- and c3-VF-SOCPs according to the labelling of the projection zones. The functional organization of these paths has now been investigated. 2. All five paths are activated by the flexor reflex afferents from wide receptive fields (one or more limbs). Distal nerves are more effective in evoking responses than proximal nerves. 3. The a-, c1- and c3-paths are activated from the ipsilateral hindlimb only. The b1- and b2-paths are activated bilaterally from the forelimbs and hindlimbs, respectively. 4. The a-, b1- and b2-paths have long segmental delays indicating a polysynaptic linkage between the primary afferents and the tract neurones, whereas the c1- and c3-paths have short delays demonstrating a monosynaptic linkage. 5. The neurones giving origin to the paths are situated within a few segments of the dorsal root entrance. The axons cross the midline, ascend through the contralateral ventral funiculus, and activate the olivary neurones monosynaptically. 6. The spino-olivary tract of the c3-path has an upper conduction velocity of 30 m/s, whereas the other paths have upper conduction velocities of 40--45 m/s. 7. The segmental delay of the b1-VF-SOCP is much longer (mean 8.7 ms) than that of the b2-VF-SOCP (mean 3.9 ms). As a result simultaneous stimuli to the hindlimbs and forelimbs are signalled to reach the cerebellar cortex at the same time. 8. It is suggested that the information carried by the different paths is concerned with the segmental motor control.

Input from ipsilateral proprio- and exteroceptive hind limb afferents to nucleus Z of the cat medulla oblongata.

1. Extracellular recordings have been made from proprio- and exteroceptive neurones in nucleus Z of the cat medulla oblongata. 2. Seventeen non-relay cells were excited by ipsilateral hind limb group I muscle afferents. Their functional were similar to those of the corresponding relay cells (Johansson & Silfvenius, 1977), i.e. activation by axon-collaterals of dorsolateral fascicle (DLF) fibres was observed, indicative of an input from the dorsal spinocerebellar tract (DSCT) to these neurones. The non-relay cells differed, however, from the relay cells in that a larger proportion of them were disynaptically connected to the spinal fibres. It is suggested that the non-relay cells by inhibition modulate the group I relay transmission in nucleus Z. 3. Fifteen cells were excited by ipsilateral hind limb low threshold skin afferents with their spinal fibres at the first cervical segment located either in the DLF or in the dorsal column (DC). Evidence for a cutaneous hind limb bulbothalamic path independent of that of the DC and of the spinocervicolemniscal path is presented. 4. Six cells were activated by ipsilateral hind limb joint afferents at low threshold. At the C1 level the spinal fibres were located either in the DLF or in the DC. No joint activated relay-cells were encountered. Axon-collateral activation of DLF-linked joint cells was observed during cerebellar stimulation, indicating that segmental DSCT cells mediated the joint input to nucleus Z. 5. The topographical organization within nucleus Z and its input relation to the DSCT is discussed.

Connexions from large, ipsilateral hind limb muscle and skin afferents to the rostral main cuneate nucleus and to the nucleus X region in the cat.

1. Evidence is presented for an input from ipsilateral hind limb group I muscle afferents and low threshold cutaneous afferents, to cells in the rostral division of the main cuneate nucleus (rMCN) and in the region of the descending vestibular nucleus and the nucleus X of Brodal & Pompeiano (1957a), the (DV-X). 2. Thirteen group I-rMCN cells were recorded from. The functional properties of these cells were similar to those of nueleus Z (Landgren & Silfvenius, 1971; Johansson & Silfvenius, 1977a, b). The cells were monosynaptically linked to spinal dorsolateral fascicle (DLF) fibres. Nine cells projected to the contralateral thalamus, i.e. a second group I hind limb bulbothalamic tract is described. Ten cells were synaptically activated from the ipsilateral cerebellum from the anterior projection zone of the dorsal spinocerebellar tract (DSCT). Axon-collateral activation by DSCT fibres was established for two of these cells. They were both bulbothalamic relay cells. For the remaining eight cells, activated from the cerebellum, this was not proven. These cells could, however, either be linked to DSCT fibres or to short axon-collaterals of a cell body of unknown location. A projection from the rMCN to the cerebellum is described and agrees with recent anatomical findings. Two cells were not excited from the cerebellum. 3. Four rMCN cells were activated by cutaneous afferents with their secondary axons in the DLF. Suggestive evidence for a bulbothalamic cutaneous hind limb path via the rMCN is presented. Two cells were activated from the cerebellum, presumably via axon-collaterals of nonsegmental cells. 4. Eight group I-DV-X cells were recorded from. They were monosynaptically linked to spinal DLF fibres and resembled functionally the nucleus Z and rMCN cells when stimulated from the periphery. Two cells projected to the contralateral thalamus, and two others were synaptically excited. Seven cells were activated from the ipsilateral cerebellum. Two of them projected to the cerebellum, and three were synapitcally activated by axon-collaterals of an undefined non-segmental cell. 5. Two DV-X cells which were activated by cutaneous afferents possibly had their spinal fibres deep in the dorsal column. Both were activated from the cerebellum, one by collaterals of a spinal axon. The functional organization of the three juxtaposed medullary nuclei, Z, rMCN and DV-X is discussed.

Axon-collateral activation by dorsal spinocerebellar tract fibres of group I relay cells of nucleus Z in the cat medulla oblongata.

1. The problem whether the group I hind limb cerebral tract and the dorsal spinocerebellar tract (DSCT) have common spinal axons has been investigated in the present study. 2. Thirty-two cells located in the nucleus Z of the cat medulla oblongata and activated by spinal fibres in the dorsolateral fascicle were selected for the study. 3. Extracellular recording from these neurones demonstrated that most of them were monosynaptically linked to spinal fibres excited by ipsilateral hind limb group I muscle afferents. The cells exhibited a restricted spatial convergence and had a limited excitatory convergence from group II muscle and from skin afferents. 4. Antidromic activation from the contralateral thalamus showed that they were bulbothalamic relay cells. 5. Cerebellar surface or depth stimulation activated 88% of the twenty-six cells tested at a short latency. With a collision technique it was demonstrated that twelve out of twenty-three (52%) of these group I relay neurones were activated by axon-collaterals of the DSCT. 6. 43% of the cells activated from the cerebellum, but not proven to be linked to the DSCT, could, nevertheless, have been excited by DSCT axon-collaterals, if it is assumed that different fibres converging with excitation on the group I relay cells were activated in the collision test.

Modulation of hindlimb reflexes by tonic neck positions in cats.

In cats with the head fixed in a stereotactic frame a slight facilitation of the gastrocnemiussoleus (GS) monosynaptic reflexes in the ipsilateral hindlimb, accompanied by an inhibition of the monosynaptic reflex of the antagnostic deep peroneal nerve (DP), occurs following lateral flexion and rotation around the body axis at an angle of 25 degrees. Following dorsiflexion of the body a moderate inhibition of the extensor reflexes (GS) up to 10% and a reciprocal slight excitation in the same range of the monosynaptic reflexes of flexor muscles (DP) is recorded. On the other hand after ventriflexion of the body there is a marked inhibition in the range of 50% of all the monosynaptic extensor and flexor reflexes for all the stimulus intensities used. This inhibition can already be demonstrated in ventriflexion of 10 degrees-15 degrees. All these patterns of reflex modulation are similar in cortically intact, ischemically decorticated and precollicularly decerebrated cats. Furthermore these patterns are not essentially influenced by cerebellectomy, although all reflex amplitudes are reduced. However, the spinal reflexes can not be modulated by these body movements after acute spinalization.

Reflex changes in discharge activities of gamma efferents to varying skin temperatures in cats.

Changes in spontaneous discharges of single γ-efferents dissected from the lumbrosacral ventral root of spinal cats were observed during cooling of warming of a small area of the foot pad skin of the ipsilateral hind limb. In a certain range of skin temperature, varying from 34°C to 38°C from preparation to preparation, temperature changes resulted in either an increase or decrease in discharge frequency, whereas above this temperature range, heating produced an increase in discharge frequency. By contrast, all γ-efferents examined responded with a negative temperature coefficient to a change in skin temperature below the respective temperature. The thermal activation of these efferents elicited by a small temperature change of the skin was affected considerably by the velocity rather than by the degree of temperature changes but the effect of both variables differed according to the skin temperature at which the temperature change occurred. Ratios of increment or decrement in discharge frequency (±Δ imp./s) to degree (±Δ °C) and velocity of temperature change (±Δ °C/s) respectively were plotted as a function of skin temperature, and indicated a directional relationship between stimulus and response by attaching to each ratio either a positive or negative temperature coefficient. The maximum activity response in the range of 20–30°C was −19 imp./s per °C at 22°C and the average was −8.33±4.77 imp./s per °C. In the warmer range, the coefficients were positive and the average was 23.79±12.21 imp./s per °C in the range of 40–44°C. The steady state activity at a constant temperature showed no sign of dependency on skin temperature. On the basis of these results, the possible mechanisms for thermal inflow contributing to a thermally-induced reflex of the γ-motoneurons and the functional implications of these reflexes for the thermoregulatory system are discussed.

Modifications and development of spinal reflexes in the alert baboon ( Papio papio) following an unilateral vestibular neurotomy

Unilateral vestibular neurotomy performed on 6 baboons was followed by the classical postural and locomotor postoperative disorders. Postoperative spinal reflex excitability was studied in these awake baboons by measuring monosynaptic recruitment ratios in soleus motoneuronal pool and the recovery cycle of the monosynaptic response. Vestibular neurotomy induces: (1) asymmetrical spinal reflexes with a decrease of the recruitment on the side of the section and an increase on the opposite side. (2) a modification of phases IV and V of the ipsilateral recovery cycle. The recovery during postoperative days of normal recruitment curves and recovery cycles is described. This recovery is relatively fast and relates closely to the compensation of postural and locomotor disorders. We note the importance of an early active motor exploration on the time-course of recovery: postoperative disorders in posture and locomotion are reduced much later when baboons are submitted to a motor restriction. These findings suggest a combined alpha and gamma-hypoexcitability of the ipsilateral hindlimb. The study of the recovery process suggests that the cerebellum is involved in the compensatory mechanisms. The regression of postural disorders, observed both in restrained and unrestrained baboons, demonstrates the importance of central regulations using proprioceptive afferents for recovery.

Exteroceptive influences on the lumbar back muscle tone and reflexes in the cat.

The exteroceptive influences on tonic activity and on stretch reflexes of the longissimus dorsi and multifidus spinae muscles were investiagted in decerebrate, spinal and chloralose-anesthetized cats. Adequate skin stimulation was used to map out facilitatory and inhibitory skin areas on the trunk and the extremities. On the trunk facilitatory areas are relatively large and located at the dorsal side while inhibitory areas are confined to the ventrolateral part of the contralateral body half. The facilitatory skin fields are of approximately the same size in decerebrate and spinal cats. On leg skin stimulation facilitation can be evoked from the ipsilateral hind limb while inhibition results from stimulation of the other limbs. Spinal cord transection increased excitatory effects of ipsilateral hind limb stimulation. Reflex responses in the back muscles to applied stretch are described. These reflexes were used as test reflexes in experiments with conditioning stimulation of the peripheral nerves supplying skin areas from which effects on back muscle activity were evoked by adequate stimulation. The conditioning-test experiments and those using adequate stimulation show that the longissimus dorsi and multifidus spinae are activated or facilitated by an ipsilateral stimulus to skin afferents. The extent of the effects induced by stimulation of skin differs in the types of preparation used. These differences may be accounted for by assuming a supraspinal control of the reflex pathways studied.

The lateral reticular nucleus in the cat. V. Does collateral activation from the dorsal spinocerebellar tract occur?

1. The assertion made by Burton, Bloedel and Gregory (1971) that neurones in the mLRN (major portion of lateral reticular nucleus comprising its parvi- and magnocellular parts) receive collateral excitation from the dorsal spinocerebellar tract (DSCT) has been tested. The activation of mLRN neurones on stimulation of limb nerves was studied in preparations with the spinal cord interrupted at C3 sparing only the dorsal part of the ipsilateral lateral funiculus (iDLF preparations), only the ipsilateral ventral quadrant (iVQ preparations), or only the ipsilateral lateral and ventral funiculi (iLFVF preparations). 2. In iDLF preparations (with the DSCT intact) stimulation of limb nerves evoked responses in only occasional mLRN neurones. These responses could be attributed to activation from some dorsally located fibres belonging to the bilateral ventral flexor reflex tract (bVFRT) and to activation from a previously unknown path activated after a long latency from ipsilateral forelimb and hindlimb afferents. 3. In iLFVF preparations (with the DSCT intact) and iVQ preparations (with the DSCT interrupted) most mLRN neurones responsed to limb nerve stimulation. The responses in two preparations were similar and could be attributed largely or exclusively to activation from the bVFRT. 4 It is concluded that the DSCT does not activate any major group of mLRN neurones.

Analysis of moving supernumerary limbs of Xenopus laevis

Behavioral, electrophysiological, and histological observations were made of adult Xenopus laevis with grafted supernumerary hindlimbs which were innervated by lumbar spinal cord segments. The supernumerary hindlimb moved in a coordinated manner in apparent perfect synchrony with the normal ipsilateral hindlimb. Frogs which had received a grafted limb at a younger tadpole stage (49 or 50) exhibited better homologous movements than did frogs which received a limb at an older stage (52 or 53). We investigated a pair of area-specific reflex responses evoked by natural mechanical stimulation of the limbs while recording activity in nerves supplying knee flexor and knee extensor muscles of the two limbs simultaneously. Appropriate, synchronous reflex responses were recorded in these homologous muscle nerves of the two limbs whether or not the knee moved. An important condition was that the frog had received its grafted limb before a critical stage (54). Some electrophysiological evidence was obtained suggesting that motor neurons do not branch to supply homologous muscles of the two limbs. Our data are consistent with the assumption of an underlying neuronal basis for the behaviorally observed homologous movements, but we can only speculate about the level of neural organization responsible for the coordination of the limbs. Histological studies confirmed an increase in the number of lateral motor column neurons on the experimental side in all cases (ranging from +3.7% to +28.1%). A significant increase in cell size (nuclear cross-sectional area) was shown in three cases. The age of the host tadpole when it receives its grafted supernumerary limb may determine the degree of the two types of cellular responses observed (increase in cell number vs. increase in cell size).

Ascending long spinal actions on forelimb motoneurons in the acute spinal cat.

Ascending long spinal reflex system were investigated by means of monosynaptic reflex testing in the acutely spinalized unanesthetized cat. 1. Hindlimb nerve stimulation gave bilateral facilitatory effects on the motoneuron pools of pectoralis major and physiological flexors of the forelimb such as biceps brachii, extensor carpi radialis, extensor digitorum communis, but elicited depressive effects on the physiological extensors such as triceps brachii, flexor carpi radialis, flexor digitorum profundus. 2. In the latissimus dorsi, which is the antagonist of pectoralis major, a depressive effect was elicited by the stimulation of ipsilateral hindlimb nerves, and a facilitatory effect by contralateral stimulation. 3. These effects were evoked mainly from group II afferent fibers in muscle as well as cutaneous nerves. 4. Intracellular recordings from motoneurons of extensor carpi radialis revealed EPSPs following stimulation of hindlimb nerves with time courses corresponding to those of the facilitatory effects. We failed to detect any potential changes in the motoneurons of flexor carpi radialis following stimulation of hindlimb nerves.

Reinforcement influences upon topography of treadmill locomotion by cats

Cats were trained to walk and trot on a treadmill for different reinforcers: food alone, food plus avoidance of an air jet, or food plus shock avoidance. Cinematographic measurements of ankle angles differed significantly across treatments at liftoff and at touchdown. Angles were widest for the food group and most closely approximated those seen previously in normal overground stepping. An interlimb timing between ipsilateral hindlimb and forelimb touchdowns did not differ significantly in trotting, but during walking the footfalls in the food-shock group were most normal and widely spaced. Swing-stance durations (the limb is off and then on the belt) were relatively invariant across treatments during walking, but not trotting. The observed mutability of stepping topography was concluded to be due in part to reinforcement type and in part to segmental afferentation, as these various inputs are integrated with central stepping commands.

Effect of local depression of inhibition on intercentral relations in the rat spinal cord

Changes in spinal reflexes of rats were studied after the formation of local depression of inhibition (a "determining dispatch station" of paroxysmal activity), generating an increased excitation wave, by means of tetanus toxin. Tonic and rhythmic paroxysmal activity generated in the poisoned half of the lumbar segments was shown to evoke discharges of all spinal and bulbar motoneurons with certain temporal characteristics. Depression of unit activity in the focus with glycine abolished this phenomenon. Excitation of the focus created on one side of the cervical segments evoked a pathologically increased scratch reflex of the ipsilateral hind limb, unconnected with a disturbance of inhibition of lumbar motoneurons. The focus had enhanced excitatory and inhibitory effects on monosynaptic reflexes of lumbar flexor motoneurons. The role of local depression of inhibition in the function of the nervous system is discussed.

The role of cutaneous afferents from the distal hindlimb in the regulation of the step cycle of thalamic cats.

The pad and the plantar surface of the foot were stimulated electrically in thalamic cats. Weak stimulation evoked an extensor reflex in the animal at rest. The same stimuli in a spontaneously walking animal applied during the stance phase produced an increase both in amplitude and duration of the ongoing extensor activity. When given during the swing phase, the stimuli either prolonged the ongoing flexor activity and/or shortened the following extensor burst. These changes in flexor and extensor burst duration were reflected in changes in the step cycle duration. Similar results were seen with direct stimulation of the sural nerve. For the latter experiments the ipsilateral hindlimb was fixed and denervated except for the ankle extensors and flexors, which showed rhythmic contractions correlated normally with the walking movements of the three remaining limbs. At rest, threshold stimulation of the sural nerve evoked a reflex contraction in the triceps surae of the fixed leg. The same stimuli applied during the contraction phase of the fixed triceps surae during walking resulted in a larger and longer extensor contraction and a delay of the following flexion. Stimulation during the relaxation phase of the fixed triceps surae reduced the duration of the following contraction phase. The findings are discussed in relation to the possible role of cutaneous input during locomotion.