Simple artificial neural network models can generate basic muscle activity patterns for human locomotion at different speeds

 A neural network model has been developed to represent the shaping function of a central pattern generator (CPG) for human locomotion. The model was based on cadence and electromyographic data obtained from a single human subject who walked on a treadmill. The only input to the model was the fundamental timing of the gait cycle (stride rate) in the form of sine and cosine waveforms whose period was equal to the stride duration. These simple signals were then shaped into the respective muscle activation patterns of eight muscles of the lower limb and trunk. A network with a relatively small number of hidden units trained with back-propagation was able to produce an excellent representation of both the amplitude and timing characteristics of the EMGs over a range of walking speeds. The results are further discussed with respect to the dependence of some muscles upon sensory feedback and other inputs not explicitly presented to the model. Received: 9 February 1998 / Accepted: 13 August 1998     

Modelling spinal circuitry involved in locomotor pattern generation: insights from deletions during fictive locomotion

The mammalian spinal cord contains a locomotor central pattern generator (CPG) that can produce alternating rhythmic activity of flexor and extensor motoneurones in the absence of rhythmic input and proprioceptive feedback. During such fictive locomotor activity in decerebrate cats, spontaneous omissions of activity occur simultaneously in multiple agonist motoneurone pools for a number of cycles. During these 'deletions', antagonist motoneurone pools usually become tonically active but may also continue to be rhythmic. The rhythmic activity that re-emerges following a deletion is often not phase shifted. This suggests that some neuronal mechanism can maintain the locomotor period when motoneurone activity fails. To account for these observations, a simplified computational model of the spinal circuitry has been developed in which the locomotor CPG consists of two levels: a half-centre rhythm generator (RG) and a pattern formation (PF) network, with reciprocal inhibitory interactions between antagonist neural populations at each level. The model represents a network of interacting neural populations with single interneurones and motoneurones described in the Hodgkin-Huxley style. The model reproduces the range of locomotor periods and phase durations observed during real locomotion in adult cats and permits independent control of the level of motoneurone activity and of step cycle timing. By altering the excitability of neural populations within the PF network, the model can reproduce deletions in which motoneurone activity fails but the phase of locomotor oscillations is maintained. The model also suggests criteria for the functional identification of spinal interneurones involved in the mammalian locomotor pattern generation.     

Modelling spinal circuitry involved in locomotor pattern generation: insights from the effects of afferent stimulation

A computational model of the mammalian spinal cord circuitry incorporating a two-level central pattern generator (CPG) with separate half-centre rhythm generator (RG) and pattern formation (PF) networks has been developed from observations obtained during fictive locomotion in decerebrate cats. Sensory afferents have been incorporated in the model to study the effects of afferent stimulation on locomotor phase switching and step cycle period and on the firing patterns of flexor and extensor motoneurones. Here we show that this CPG structure can be integrated with reflex circuits to reproduce the reorganization of group I reflex pathways occurring during locomotion. During the extensor phase of fictive locomotion, activation of extensor muscle group I afferents increases extensor motoneurone activity and prolongs the extensor phase. This extensor phase prolongation may occur with or without a resetting of the locomotor cycle, which (according to the model) depends on the degree to which sensory input affects the RG and PF circuits, respectively. The same stimulation delivered during flexion produces a temporary resetting to extension without changing the timing of following locomotor cycles. The model reproduces this behaviour by suggesting that this sensory input influences the PF network without affecting the RG. The model also suggests that the different effects of flexor muscle nerve afferent stimulation observed experimentally (phase prolongation versus resetting) result from opposing influences of flexor group I and II afferents on the PF and RG circuits controlling the activity of flexor and extensor motoneurones. The results of modelling provide insights into proprioceptive control of locomotion.     

Functional characterization of dI6 interneurons in the neonatal mouse spinal cord

Our understanding of the neural control of locomotion has been greatly enhanced by the ability to identify and manipulate genetically defined populations of interneurons that comprise the locomotor central pattern generator (CPG). To date, the dI6 interneurons are one of the few populations that settle in the ventral region of the postnatal spinal cord that have not been investigated. In the present study, we utilized a novel transgenic mouse line to electrophysiologically characterize dI6 interneurons located close to the central canal and study their function during fictive locomotion. The majority of dI6 cells investigated were found to be rhythmically active during fictive locomotion and could be divided into two electrophysiologically distinct populations of interneurons. The first population fired rhythmic trains of action potentials that were loosely coupled to ventral root output and contained several intrinsic membrane properties of rhythm-generating neurons, raising the possibility that these cells may be involved in the generation of rhythmic activity in the locomotor CPG. The second population fired rhythmic trains of action potentials that were tightly coupled to ventral root output and lacked intrinsic oscillatory mechanisms, indicating that these neurons may be driven by a rhythm-generating network. Together these results indicate that dI6 neurons comprise an important component of the locomotor CPG that participate in multiple facets of motor behavior.     

Afferent control of central pattern generators: experimental analysis of locomotion in the decerebrate cat

Changes in the motor activity of the spinal locomotor generator evoked by tonic and phasic peripheral afferent signals during fictitious locomotion of both slow and fast rhythms were analysed in the cat. The tonic afferent inflow was conditioned by the position of the hindlimb. The phasic afferent signals were imitated by electrical stimulation of hindlimb nerves. The correlation between the kinematics of hindlimb locomotor movement and sensory inflow was investigated during actual locomotion. Reliable correlations between motor activity parameters during fictitious locomotion were revealed in cases of both slow and fast "locomotor" rhythms. The main difference between these cases was that correlations "duration-intensity" were positive in the first and negative in the second case. The functional role of "locomotor" pattern dependence on tonic sensory inflow consisted of providing stability for planting the hindlimb on the ground. For any investigated afferent input the phase moments in the "locomotor" cycle were found, in which an afferent signal caused no rearrangement in locomotor generator activity. These moments corresponded to the transitions between "flexion" and "extension" phases and to the bursts of integral afferent activity observed during real locomotion. The data obtained are compared with the results previously described for the scratching generator. The character of changes in "locomotor" activity in response to tonic and phasic sensory signals was similar to that of such changes in "scratching" rhythm in the case of fast "locomotion". Intensification of the "flexion" phase caused by phasic high-intensity stimulation of cutaneous afferents during low "locomotor" rhythm was changed to inhibition (such as observed during "scratching") when this rhythm was fast. It is concluded that the main regularities of peripheral afferent control for both the locomotor and scratching generators are the same. Moreover, these central pattern generators are just working regimes of a general spinal motor optimal control system containing the intrinsic model of limb movement dynamics. The consequences of this concept and ways of further research are discussed.     

Identification of two types of synaptic activity in the earthworm nervous system during locomotion

In the ventral nervous system of the earthworm, a central pattern generator and motor neurons are activated during locomotion. We have previously reported that bath application of octopamine (OA) induces fictive locomotion in the earthworm, and the burst frequency of electrical activity from the first lateral nerves increases with OA concentration. However, there are no reports concerning locomotor neural networks in the earthworm. To identify neural networks involved in fictive locomotion, we optically monitored activity-dependent fluorescent staining in the earthworm ventral nerve cord (VNC) with a styryl dye, N-(3-triethylammoniumpropyl)-4-(4-(dibutylamino)styryl)pyridinium dibromide (FM1-43), and a confocal laser scanning microscope. OA induces FM1-43 fluorescence in a dose-dependent manner, with bright fluorescent spots of 3-10 microm in diameter observed to be localized around specified neurons in the segmental ganglion of the VNC. We compared OA dose-response curves for FM1-43 fluorescence with the bursting frequency for fictive locomotion, and found that two types of curves could be identified: one fluorescence response shows a similar dose-dependency to that of the burst frequency, while another response has a higher sensitivity to OA. From these results, we suggest that OA acts as one of the neuromodulators for the earthworm locomotion. This is the first attempt to record motor and inter-neuronal activities simultaneously in a locomotor network in the earthworm.     

5-HT prolongs ventral root bursting via presynaptic inhibition of synaptic activity during fictive locomotion in lamprey

Locomotor pattern generation is maintained by integration of the intrinsic properties of spinal central pattern generator (CPG) neurons in conjunction with synaptic activity of the neural network. In the lamprey, the spinal locomotor CPG is modulated by 5-HT. On a cellular level, 5-HT presynaptically inhibits synaptic transmission and postsynaptically inhibits a Ca2+-activated K+ current responsible for the slow afterhyperpolarization (sAHP) that follows action potentials in ventral horn neurons. To understand the contribution of these cellular mechanisms to the modulation of the spinal CPG, we have tested the effect of selective 5-HT analogues against fictive locomotion initiated by bath application of N-methyl-d-aspartate (NMDA). We found that the 5-HT1D agonist, L694-247, dramatically prolongs the frequency of ventral root bursting. Furthermore, we show that L694-247 presynaptically inhibits synaptic transmission without altering postsynaptic Ca2+-activated K+ currents. We also confirm that 5-HT inhibits synaptic transmission at concentrations that modulate locomotion. To examine the mechanism by which selective presynaptic inhibition modulates the frequency of fictive locomotion, we performed voltage- and current-clamp recordings of CPG neurons during locomotion. Our results show that 5-HT decreases glutamatergic synaptic drive within the locomotor CPG during fictive locomotion. Thus we conclude that presynaptic inhibition of neurotransmitter release contributes to 5-HT-mediated modulation of locomotor activity.     

Control of locomotor cycle durations

In intact animals and humans, increases in locomotor speed are usually associated with decreases in step cycle duration. Most data indicate that the locomotor central pattern generator (CPG) shortens cycle duration mainly by shortening the durations of extensor rather than flexor phases of the step cycle. Here we report that in fictive locomotion elicited by electrical stimulation of the midbrain locomotor region (MLR) in the cat, spontaneous variations in cycle duration were due more to changes in flexor rather than extensor phase durations in 22 of 31 experiments. The locomotor CPG is therefore not inherently extensor- or flexor-biased. We coined the term "dominant" to designate the phase (flexion or extension) showing the larger variation. In a simple half-center oscillator model, experimental phase duration plots were fitted well by adjusting two parameters that corresponded to background drive ("bias") and sensitivity ("gain") of the oscillator's timing elements. By analogy we argue that variations in background drive to the neural timing elements of the CPG could produce larger variations in phase duration in the half-center receiving the lower background drive, i.e., background drive may determine which half-center is dominant. The fact that data from normal cats were also fitted well by the model indicates that sensory input and central drive combine to determine locomotor phase durations. We conclude that there is a considerable flexibility in the control of phase durations in MLR-induced fictive locomotion. We posit that this may be explained by changes in background excitation of neural timing elements in the locomotor CPG.     

Stretch and H reflexes in triceps surae are similar during tonic and rhythmic contractions in high decerebrate cats

During locomotion in decerebrate and spinal cats the group Ia afferents from hind leg muscles are depolarized rhythmically. An earlier study concluded that this locomotor-related primary afferent depolarization (PAD) does not contribute to modulation of monosynaptic reflex pathways during locomotion. This finding indicated that the neural network generating the locomotor rhythm, the central pattern generator (CPG), does not presynaptically inhibit monosynaptic reflexes. In this investigation we tested this prediction in decerebrate cats by measuring the magnitude of reflexes evoked in ankle extensor muscles during periods of tonic contractions and during sequences of rhythmic contractions. The latter occurred when the animal was induced to walk on a treadmill. At the similar levels of activity in the soleus muscle there was no significant difference in the magnitude of the soleus H reflex in these two behavioral situations. Similar results were obtained for reflexes evoked by brief stretches of the soleus muscle. We also examined the reflexes evoked by ramp-and-hold stretches during periods of rhythmic and tonic activity of the isolated medial gastrocnemius (MG) muscle. At similar levels of background activity, the reflexes evoked in the MG muscle were the same during rhythmic and tonic contractions. Our failure to observe a reduction in the magnitude of H reflexes and stretch reflexes during rhythmic contractions, compared with reflexes evoked at the same level of background activity during tonic contractions, is consistent with the notion that the CPG for stepping does not presynaptically inhibit monosynaptic reflexes during the extension phase of locomotor activity. Our results indicate that presynaptic inhibition of the monosynaptic reflex associated with normal locomotion in cats or humans arises from sources other than the extensor burst generating system of the central pattern generator.     

Heat shock-mediated thermoprotection of larval locomotion compromised by ubiquitous overexpression of Hsp70 in Drosophila melanogaster

Maintaining the competence of locomotor circuitry under stressful conditions can benefit organisms by enabling locomotion to more tolerable microhabitats. We show that prior heat shock protects locomotion and the locomotor central pattern generator of larval Drosophila against subsequent hyperthermic stress. We combined molecular genetic, electrophysiological, and behavioral techniques to investigate heat shock-mediated thermoprotection. Prior heat shock increased the distance traveled by larvae during hyperthermia before failure. The frequency of the rhythm of peristaltic locomotor contractions and the velocity of locomotion were both less thermosensitive after heat shock and were less susceptible to failure at high temperatures. Rhythmic coordinated motor patterns, recorded intracellularly as excitatory junction potentials in body wall muscles of dissected preparations, were centrally generated because patterns could still be generated in the absence of sensory feedback (sensory function disrupted with shibire). Prior heat shock protected central circuit operation during hyperthermic stress by increasing the temperature at which it failed. Overexpression of Hsp70 after a heat shock using transgenic flies (traII) did not enhance thermoprotection, as expected, but had deleterious effects on parameters of behavior.     

Barosensory cells in the nucleus tractus solitarius receive convergent input from group III muscle afferents and central command

Some neural mechanism must prevent the full expression of the baroreceptor reflex during static exercise because arterial blood pressure increases even though the baroreceptors are functioning. Two likely candidates are central command and input from the thin fiber muscle afferents evoking the exercise pressor reflex. Recently, activation of the mesencephalic locomotor region, an anatomical locus for central command, was found to inhibit the discharge of nucleus tractus solitarius cells that were stimulated by arterial baroreceptors in decerebrated cats. In contrast, the effect of thin fiber muscle afferent input on the discharge of nucleus tractus solitarius cells stimulated by baroreceptors is not known. Consequently in decerebrated unanesthetized cats, we examined the responses of barosensory nucleus tractus solitarius cells to stimulation of thin fiber muscle afferents and to stimulation of the mesencephalic locomotor region, a maneuver which evoked fictive locomotion. We found that electrical stimulation of either the mesencephalic locomotor region or the gastrocnemius nerve at current intensities that recruited group III afferents inhibited the discharge of nucleus tractus solitarius cells receiving baroreceptor input. We also found that the inhibitory effects of both gastrocnemius nerve stimulation and mesencephalic locomotor region stimulation converged onto the same barosensory nucleus tractus solitarius cells. We conclude that the nucleus tractus solitarius is probably the site whereby input from both central command and thin fiber muscle afferents function to reset the baroreceptor reflex during exercise.     

Neural oscillators triggered by loading and hip orientation can generate activation patterns at the ankle during walking in humans

Spinal pattern generators (SPGs), which are neural networks without a central input from the brain may be responsible for controlling locomotion. In this study, we used neural oscillators to examine the rhythmic patterns generated at the ankle during walking. Seven healthy male subjects were requested to walk at their normal self-selected speed on a treadmill. Force measurements acquired from pressure insoles, electromyography and kinematic data were captured simultaneously. The SPG model consisted of a simple oscillator made up of two neurons; one neuron will activate an ankle extensor and the other will activate an ankle flexor. The outputs of the oscillator represented the muscle activation of each muscle. A nonlinear least squares algorithm was used to determine a set of parameters that would optimise the differences between model output and experimental data. Insole forces and hip angles of six consecutive strides were used as inputs to the model, which generated outputs that closely fitted experimental data. Our results showed that it is possible to reproduce muscle activations using neural oscillators. A close correlation between simulated and measured muscle activations indicated that spinal control should not be underestimated in models of human locomotion.     

Muscle spindle and fusimotor activity in locomotion

Mammals may exhibit different forms of locomotion even within a species. A particular form of locomotion (e.g. walk, run, bound) appears to be selected by supraspinal commands, but the precise pattern, i.e. phasing of limbs and muscles, is generated within the spinal cord by so-called central pattern generators. Peripheral sense organs, particularly the muscle spindle, play a crucial role in modulating the central pattern generator output. In turn, the feedback from muscle spindles is itself modulated by static and dynamic fusimotor (gamma) neurons. The activity of muscle spindle afferents and fusimotor neurons during locomotion in the cat is reviewed here. There is evidence for some alpha–gamma co-activation during locomotion involving static gamma motoneurons. However, both static and dynamic gamma motoneurons show patterns of modulation that are distinct from alpha motoneuron activity. It has been proposed that static gamma activity may drive muscle spindle secondary endings to signal the intended movement to the central nervous system. Dynamic gamma motoneuron drive appears to prime muscle spindle primary endings to signal transitions in phase of the locomotor cycle. These findings come largely from reduced animal preparations (decerebrate) and require confirmation in freely moving intact animals.     

大脑皮层信号作用下人体步态节律运动的探讨

节律运动是由中枢模式发生器自激产生的。目前的中枢模式发生器(CPG)建模研究仅表现了CPG的自激行为,对于人脑信号的调节性并没有涉及。本研究提出了基于Matsuoka神经振荡器的CPG模型的修正模型,目的在于以这一修正模型体现大脑皮层信号对于CPG网络的调控性。通过对原有模型中输入刺激与网络内部参数的复杂关联的构建,使得模型本身各参数随输入信号的变化而变化,增强了输入信号对于网络自身的影响,令网络不局限于自激状态,还能够产生自我调节的运动形式,从而体现出大脑信号的调控作用。数值模拟计算结果表明,修正后的模型随着输入刺激的变化可以产生不同模式及不同频率的运动形式,且各不同形式之间可以相互转换,从而在理论上很好地反映出大脑信号在步态节律运动过程中,对步态的模式和频率所起到的一定的调节作用,实现了各种步态运动之间的行为转换及恢复的功能,从理论上实现了自发节律与大脑调节性节律运动的共存性,做到大脑信号与CPG模型的统一。     

Glial-derived adenosine modulates spinal motor networks in mice

The activation of purinergic receptors modulates central pattern generators controlling rhythmic motor behaviors, including respiration in rodents and swimming in frog tadpoles. The present study aimed to determine whether purinergic signaling also modulates the mammalian locomotor central pattern generator. This was investigated by using isolated spinal cord preparations obtained from neonatal mice in which locomotor-related activity can be induced pharmacologically. The application of either ATP or adenosine led to a reduction in the frequency of locomotor activity recorded from ventral roots. ATP had no effect when applied in the presence of both the adenosine receptor antagonist theophylline and the ectonucleotidase inhibitor ARL67156, demonstrating that the effects of ATP application result from the breakdown of ATP to adenosine and subsequent activation of adenosine receptors. The application of theophylline or the A(1)-specific antagonist cyclopentyl dipropylxanthine, but not the A(2A)-receptor antagonist SCH58261, caused an increase in locomotor burst frequency, demonstrating that endogenously derived adenosine activates A(1) receptors during locomotor network activity. Furthermore, theophylline had no effect in the presence of the ectonucleotidase inhibitor ARL67156 or the glial toxins methionine sulfoximine or ethyl fluoracetate, suggesting that endogenous adenosine is derived from ATP, which is released from glia. Finally, adenosine had no effect on slow rhythmic activity recorded upon blockade of all inhibitory transmission, suggesting that adenosine may act via the modulation of inhibitory transmission. Together, these data highlight endogenous purinergic gliotransmission, involving activation of A(1) receptors, as an important intrinsic modulatory system controlling the frequency of activity generated by spinal locomotor circuitry in mammals.     

Serotonin modulates the properties of ascending commissural interneurons in the neonatal mouse spinal cord

The interneuron populations that constitute the central pattern generator (CPG) for locomotion in the mammalian spinal cord are not well understood. We studied the properties of a set of commissural interneurons whose axons cross and ascend in the contralateral cord (aCINs) in the neonatal mouse. During N-methyl-D-aspartate (NMDA) and 5-HT-induced fictive locomotion, a majority of lumbar (L2) aCINs examined were rhythmically active; most of them fired in phase with the ipsilateral motoneuron pool, but some fired in phase with contralateral motoneurons. 5-HT plays a critical role in enabling the locomotor CPG to function. We found that 5-HT increased the excitability of aCINs by depolarizing the membrane potential, reducing the postspike afterhyperpolarization amplitude, broadening the action potential, and decreasing the action potential threshold. Serotonin had no significant effect on the input resistance and sag amplitude of aCINs. These results support the hypothesis that aCINs play important roles in coordinating left-right movements during fictive locomotion and thus may be component neurons in the locomotor CPG in neonatal mice.     

Polarization of primary afferent terminals of lumbosacral cord elicited by the activity of spinal locomotor generator

The changes of electrical polarization of primary afferent terminals in the lumbosacral spinal cord have been investigated on immobilized, decorticated and spinal cats during fictive locomotion. Fictive locomotion was spontaneous or provoked by stimulation of either dorsal root or dorsal funiculi in the lumbar segments. The activation of the locomotor generator and appearance of fictive locomotion were always associated with a sustained dorsal root hyperpolarization. On the background of this positivity periodic negative dorsal root potential oscillations appeared synchronously with efferent discharges in motor hind-limb nerves. These periodic waves of primary afferent depolarization occurred in phase in different ipsilateral lumbosacral segments. On the contralateral side the periodic changes in dorsal root potential were out of phase during fictive stepping and in phase during fictive galloping. The use of Wall's technique has shown that tonic and periodic changes in dorsal root potential reflect the changes occurring in polarization of central terminals of cutaneous and muscle (Ia and Ib) groups of afferent fibres. It is concluded that the level of electrical polarization of primary afferent terminals is determined directly by the activity of the spinal locomotor generator; activation of the generator is followed by hyperpolarization of primary afferent terminals. By so modulating the polarization of afferent terminals, the locomotor generator can perform tonic and phase-dependent selection of afferent information.     

A neural network approach for determining gait modifications to reduce the contact force in knee joint implant

There is a growing interest in non-surgical gait rehabilitation treatments to reduce the loading in the knee joint. In particular, synergetic kinematic changes required for joint offloading should be determined individually for each subject. Previous studies for gait rehabilitation designs are typically relied on a “trial-and-error” approach, using multi-body dynamic (MBD) analysis. However MBD is fairly time demanding which prevents it to be used iteratively for each subject.This study employed an artificial neural network to develop a cost-effective computational framework for designing gait rehabilitation patterns. A feed forward artificial neural network (FFANN) was trained based on a number of experimental gait trials obtained from literature. The trained network was then hired to calculate the appropriate kinematic waveforms (output) needed to achieve desired knee joint loading patterns (input). An auxiliary neural network was also developed to update the ground reaction force and moment profiles with respect to the predicted kinematic waveforms. The feasibility and efficiency of the predicted kinematic patterns were then evaluated through MBD analysis.Resuls showed that FFANN-based predicted kinematics could effectively decrease the total knee joint reaction forces. Peak values of the resultant knee joint forces, with respect to the bodyweight (BW), were reduced by 20% BW and 25% BW in the midstance and the terminal stance phases. Impulse values of the knee joint loading patterns were also decreased by 17% BW*s and 24%BW*s in the corresponding phases. The FFANN-based framework suggested a cost-effective forward solution which directly calculated the kinematic variations needed to implement a given desired knee joint loading pattern. It is therefore expected that this approach provides potential advantages and further insights into knee rehabilitation designs.     

Activation of group I metabotropic glutamate receptors modulates locomotor-related motoneuron output in mice

Fast glutamatergic transmission via ionotropic receptors is critical for the generation of locomotion by spinal motor networks. In addition, glutamate can act via metabotropic glutamate receptors (mGluRs) to modulate the timing of ongoing locomotor activity. In the present study, we investigated whether mGluRs also modulate the intensity of motor output generated by spinal motor networks. Application of the group I mGluR agonist (S)-3,5-dihydroxyphenylglycine (DHPG) reduced the amplitude and increased the frequency of locomotor-related motoneuron output recorded from the lumbar ventral roots of isolated mouse spinal cord preparations. Whole cell patch-clamp recordings of spinal motoneurons revealed multiple mechanisms by which group I mGluRs modulate motoneuron output. Although DHPG depolarized the resting membrane potential and reduced the voltage threshold for action potential generation, the activation of group I mGluRs had a net inhibitory effect on motoneuron output that appeared to reflect the modulation of fast, inactivating Na(+) currents and action potential parameters. In addition, group I mGluR activation decreased the amplitude of locomotor-related excitatory input to motoneurons. Analyses of miniature excitatory postsynaptic currents indicated that mGluRs modulate synaptic drive to motoneurons via both pre- and postsynaptic mechanisms. These data highlight group I mGluRs as a potentially important source of neuromodulation within the spinal cord that, in addition to modulating components of the central pattern generator for locomotion, can modulate the intensity of motoneuron output during motor behavior. Given that group I mGluR activation reduces motoneuron excitability, mGluRs may provide negative feedback control of motoneuron output, particularly during high levels of glutamatergic stimulation.     

Patterns of locomotor drive to motoneurons and last-order interneurons: clues to the structure of the CPG.

We have examined the linkage between patterns of activity in several hindlimb motor pools and the modulation of oligosynaptic cutaneous reflex pathways during fictive locomotion in decerebrate unanesthetized cats to assess the notion that such linkages can shed light on the structure of the central pattern generator (CPG) for locomotion. We have concentrated attention on the cutaneous reflex pathways that project to the flexor digitorum longus (FDL) motor pool because of that muscle's unique variable behavior during normal and fictive locomotion in the cat. Differential locomotor control of last-order excitatory interneurons in pathways from low-threshold cutaneous afferents in the superficial peroneal and medial plantar afferents to FDL motoneurons is fully documented for the first time. The qualitative patterns of differential control are shown to remain the same whether the FDL muscle is active in early flexion, as usually found, or during the extension phase of fictive locomotion, which is less common during fictive stepping. The patterns of motor pool activity and of reflex pathway modulation indicate that the flexion phase of fictive locomotion has distinct early versus late components. Observations during "normal" and unusual patterns of fictive stepping suggest that some aspects of locomotor pattern formation can be separated from rhythm generation, implying that these two CPG functions may be embodied, at least in part, in distinct neural organizations. The results are discussed in relation to a provisional circuit diagram that could explain the experimental findings.