Fusimotor control of proprioceptive feedback during locomotion and balancing: can simple lessons be learned for artificial control of gait?

Pathophysiology of spasticity: implications for neurorehabilitation

Spasticity is the velocity-dependent increase in muscle tone due to the exaggeration of stretch reflex. It is only one of the several components of the upper motor neuron syndrome (UMNS). The central lesion causing the UMNS disrupts the balance of supraspinal inhibitory and excitatory inputs directed to the spinal cord, leading to a state of disinhibition of the stretch reflex. However, the delay between the acute neurological insult (trauma or stroke) and the appearance of spasticity argues against it simply being a release phenomenon and suggests some sort of plastic changes, occurring in the spinal cord and also in the brain. An important plastic change in the spinal cord could be the progressive reduction of postactivation depression due to limb immobilization. As well as hyperexcitable stretch reflexes, secondary soft tissue changes in the paretic limbs enhance muscle resistance to passive displacements. Therefore, in patients with UMNS, hypertonia can be divided into two components: hypertonia mediated by the stretch reflex, which corresponds to spasticity, and hypertonia due to soft tissue changes, which is often referred as nonreflex hypertonia or intrinsic hypertonia. Compelling evidences state that limb mobilisation in patients with UMNS is essential to prevent and treat both spasticity and intrinsic hypertonia.

Computational motor control: feedback and accuracy

Speed/accuracy trade-off is a ubiquitous phenomenon in motor behaviour, which has been ascribed to the presence of signal-dependent noise (SDN) in motor commands. Although this explanation can provide a quantitative account of many aspects of motor variability, including Fitts' law, the fact that this law is frequently violated, e.g. during the acquisition of new motor skills, remains unexplained. Here, we describe a principled approach to the influence of noise on motor behaviour, in which motor variability results from the interplay between sensory and motor execution noises in an optimal feedback-controlled system. In this framework, we first show that Fitts' law arises due to signal-dependent motor noise (SDN(m)) when sensory (proprioceptive) noise is low, e.g. under visual feedback. Then we show that the terminal variability of non-visually guided movement can be explained by the presence of signal-dependent proprioceptive noise. Finally, we show that movement accuracy can be controlled by opposite changes in signal-dependent sensory (SDN(s)) and SDN(m), a phenomenon that could be ascribed to muscular co-contraction. As the model also explains kinematics, kinetics, muscular and neural characteristics of reaching movements, it provides a unified framework to address motor variability.

Supraspinal locomotor control in quadrupeds and humans

Locomotion in humans and other vertebrates is based on spinal pattern generators, which are regulated by supraspinal control. Most of our knowledge about the hierarchical network of supraspinal locomotion centres derives from animal experiments, mainly in the cat. Here we summarize evidence that the supraspinal network of quadrupeds is conserved in humans despite their transition to bipedalism. By use of mental imagery of locomotion in fMRI we found (1), locomotion modulates sensory systems and is itself modulated by sensory signals. During automated locomotion in healthy subjects cortical sensory inhibition occurs in vestibular and somatosensory areas; this inhibition is cancelled in the congenitally blind; (2), we delineated separate and distinct areas in the brainstem and cerebellum which are remarkably similar to the feline locomotor network. The activations found here include homologues to the pacemakers for gait initiation and speed regulation in the interfastigial cerebellum and bilateral midbrain tegmentum (cerebellar and mesencephalic locomotor regions), their descending target regions in the pontine reticular formation, and the rhythm generators in the cerebellar vermis and paravermal cerebellar cortex. This conservation of the basic organization of supraspinal locomotor control during vertebrate phylogeny opens new perspectives for both, the diagnosis and treatment of common gait disorders. It is conceivable that electrical stimulation of locomotor brain stem centres may initiate and improve gait in selected patients suffering from Parkinson's disease or progressive supranuclear palsy.

Changes in Excitability of the Cortical Projections to the Human Tibialis Anterior After Paired Associative Stimulation

Paired associative stimulation (PAS) based on Hebb's law of association can induce plastic changes in the intact human. The optimal interstimulus interval (ISI) between the peripheral nerve and transcranial magnetic stimulus is not known for muscles of the lower leg. The aims of this study were to investigate the effect of PAS for a variety of ISIs and to explore the efficacy of PAS when applied during dynamic activation of the target muscle. PAS was applied at 0.2 Hz for 30 min with the tibialis anterior (TA) at rest. The ISI was varied randomly in seven sessions ( n = 5). Subsequently, PAS was applied ( n = 14, ISI = 55 ms) with the TA relaxed or dorsi-flexing. Finally, an optimized ISI based on the subject somatosensory evoked potential (SEP) latency plus a central processing delay (6 ms) was used ( n = 13). Motor-evoked potentials (MEPs) were elicited in the TA before and after the intervention, and the size of the TA MEP was extracted. ISIs of 45, 50, and 55 ms increased and 40 ms decreased TA MEP significantly ( P = 0.01). PAS during dorsi-flexion increased TA MEP size by 92% ( P = 0.001). PAS delivered at rest resulted in a nonsignificant increase; however, when the ISI was optimized from SEP latency recordings, all subjects showed significant increases ( P = 0.002). No changes in MEP size occurred in the antagonist. Results confirm that the excitability of the corticospinal projections to the TA but not the antagonist can be increased after PAS. This is strongly dependent on the individualized ISI and on the activation state of the muscle.

Time to task failure varies with the gain of the feedback signal for women, but not for men

Varying the gain of the feedback signal during a target-matching task alters the synaptic input onto the motor neuron pool. The purpose was to determine the influence of the gain of the feedback signal on the time to failure for men and women when maintaining arm position while supporting a submaximal inertial load with the elbow flexor muscles. While seated with the upper arm vertical, 15 women and 14 men maintained a constant elbow angle (1.57 rad) and supported a load equal to 15% of maximal voluntary contraction (MVC) force until failure. The task was performed on separate days with either a low gain or a high gain for the joint-angle signal. The percent decline in MVC force after the fatiguing contraction was similar for the low- and high-gain conditions (P = 0.24), and did not differ for men and women (P = 0.11). The discharge of motor units in biceps brachii declined at a greater rate during the high-gain condition for men and women, but only the women experienced a briefer time to failure for the high-gain session (8.7 +/- 2.3 min) compared with the low-gain session (11.9 +/- 4.8; P = 0.003). The men had similar times to failure for the low- (6.0 +/- 2.2 min) and high-gain conditions (5.9 +/- 2.1 min; P = 0.35). Linear and stepwise, multiple-regression analyses revealed that the time to failure for the men was associated with the absolute target force, the standard deviation (SD) for the resultant wrist acceleration, and the brachialis aEMG (P <or= 0.02), whereas the time to failure for the women was associated with the rate of decline in motor unit discharge, the SD for the resultant wrist acceleration, and the changes in mean arterial pressure and heart rate (P <or= 0.001). Despite each subject exerting the same net muscle torque during the two gain conditions and a similar effect of feedback gain on the discharge rate of motor units for all subjects, the time to failure for the fatiguing contractions was limited by different mechanisms for the men and women.

Prolonged Vibration of the Biceps Brachii Tendon Reduces Time to Failure When Maintaining Arm Position With a Submaximal Load

Vibration reduces the amplitudes of the tendon jerk response and the Hoffmann and stretch reflexes in the muscle exposed to the vibration, yet does not alter the time to task failure when the task involves exerting a submaximal force against a rigid restraint. Because the amplitude of the stretch reflex is greater when a limb acts against a compliant load than a rigid restraint, the purpose was to determine the influence of prolonged tendon vibration on the time to failure when maintaining limb position with the elbow flexor muscles. Twenty-five healthy men performed the fatiguing contraction by maintaining elbow angle at 1.57 rad until failure while supporting a load equal to 20% of maximal voluntary contraction (MVC) force. The fatiguing contraction was performed on 3 separate days with different levels of vibration applied to the biceps brachii tendon: no vibration, subthreshold for a tonic vibration reflex (TVR), and suprathreshold for a TVR. MVC force before the fatiguing contraction was similar across the three sessions (mean of 3 sessions: 313 ± 54 N, P = 0.83). Despite a similar decline in MVC force after the fatiguing contraction across conditions (–18.0 ± 8.0%, P > 0.05), the time to task failure was 3.7 ± 1.4 min for the suprathreshold TVR condition, 4.3 ± 2.1 min for the subthreshold TVR condition, and 5.0 ± 2.2 min for the no-vibration condition ( P < 0 0.001). The average EMG of the elbow flexor muscles was similar ( P = 0.22) during the fatiguing contractions. However, the fluctuations in limb acceleration at task onset were greater for the suprathreshold TVR condition ( P < 0.01), but were not different between the subthreshold TVR and no-vibration conditions ( P ≥ 0.22). Furthermore, the difference in the SD of limb acceleration between the no-vibration and vibration conditions was correlated with the difference in time to failure for the no-vibration and subthreshold TVR conditions ( P = 0.03; r2 = 0.22), but not for the no-vibration and suprathreshold TVR conditions ( P = 0.90; r2 = 0.001). These findings indicate that prolonged vibration reduced the time to failure of a sustained contraction when subjects maintained limb position, suggesting that peripheral inputs to the motor neuron pool play a significant role in sustaining a contraction during tasks that require active control of limb position.

Frequency Modulation of Motor Unit Discharge Has Task-Dependent Effects on Fluctuations in Motor Output

The rate of change in the fluctuations in motor output differs during the performance of fatiguing contractions that involve different types of loads. The purpose of this study was to examine the contribution of frequency modulation of motor unit discharge to the fluctuations in the motor output during sustained contractions with the force and position tasks. In separate tests with the upper arm vertical and the elbow flexed to 1.57 rad, the seated subjects maintained either a constant upward force at the wrist (force task) or a constant elbow angle (position task). The force and position tasks were performed in random order at a target force equal to 3.6 ± 2.1% (mean ± SD) of the maximal voluntary contraction (MVC) force above the recruitment threshold of an isolated motor unit from the biceps brachii. Each subject maintained the two tasks for an identical duration (161 ± 93 s) at a mean target force of 22.4 ± 13.6% MVC. As expected, the rate of increase in the fluctuations in motor output (force task: SD for detrended force; position task: SD for vertical acceleration) was greater for the position task than the force task ( P < 0.001). The amplitude of the coefficient of variation (CV) and the power spectra for motor unit discharge were similar between tasks ( P > 0.1) and did not change with time ( P > 0.1), and could not explain the different rates of increase in motor output fluctuations for the two tasks. Nonetheless, frequency modulation of motor unit discharge differed during the two tasks and predicted ( P < 0.001) both the CV for discharge rate (force task: 1–3, 12–13, and 14–15 Hz; position task: 0–1, and 1–2 Hz) and the fluctuations in motor output (force task: 5–6, 9–10, 12–13, and 14–15 Hz; position task: 6–7, 14–15, 17–19, 20–21, and 23–24 Hz). Frequency modulation of motor unit discharge rate differed for the force and position tasks and influenced the ability to sustain steady contractions.

Task failure during fatiguing contractions performed by humans

By comparing the physiological adjustments that occur when two similar fatiguing contractions are performed to failure, it is possible to identify mechanisms that limit the duration of the more difficult task. This approach has been used to study two fatiguing contractions, referred to as the force and position tasks, which differed in the type of feedback given to the subject and the amount of support provided by the surroundings. Even though the two tasks required a similar net muscle torque during submaximal isometric contractions, the duration that the position task could be sustained was consistently much briefer than that for the force task. The position task involved a greater rate of increase in EMG activity and more marked changes in motor unit recruitment and rate coding compared with the force task. These observations are consistent with the hypothesis that the motor unit pool was recruited more rapidly during the position task. The difference in motor unit behavior appeared to be caused by variation in synaptic input, likely involving heightened sensitivity of the stretch reflex during the position task. Upon repeat performances of the two fatiguing contractions, some subjects were able to increase the time to failure for the force task but not the position task. Furthermore, the time to failure for the position task could be influenced by the postural demands associated with maintaining the position of the limb, and the difference in the two durations was enhanced when the postural activity evoked a pressor response. These observations indicate that the difference in the duration of the two fatiguing contractions was attributable to differences in the control strategy used to sustain the tasks and the magnitude of the associated postural activity.

Review of motor control mechanisms underlying impact absorption from falls

The absorption of impacts resulting from contact with a landing surface during gait, running and drop landings has received considerable attention in the literature. This research has important clinical relevance as failure to appropriately plan and control impact absorption may lead to injuries to the musculo-skeletal system. This review attempts to summarize evidence gathered by studies on the motor control aspects of impact absorption during landing movements. Although this review focuses primarily on the control of landings from self-initiated falls or 'drop landings', an understanding of the motor control mechanisms underlying impact absorption is essential to understand common anticipatory and reflex mechanisms involved in a broader variety of movements such as running and jumping. The review is structured in three parts: the first two parts examine the preparatory muscle activity occurring during the fall (Part I) and after touch down (Part II). Part III explores the proposed sensorimotor mechanisms underlying the control of landing. The review concludes with as yet unresolved questions and directions for future research.

Motor-unit activity differs with load type during a fatiguing contraction

Despite a similar rate of change in average electromyographic (EMG) activity, previous studies have observed different rates of change in mean arterial pressure, heart rate, perceived exertion, and fluctuations in motor output during the performance of fatiguing contractions that involved different types of loads. To obtain a more direct measure of the motor output from the spinal cord, the purpose of this study was to compare the discharge characteristics of the same motor unit in biceps brachii during the performance of two types of fatiguing contractions. In separate tests with the upper arm vertical and the elbow flexed to 1.57 rad, the seated subjects maintained either a constant upward force at the wrist (force task) or a constant elbow angle (position task) for a prescribed duration. The force and position tasks were performed in random order at a target force equal to 3.5 +/- 2.1% (mean +/- SD) of the maximal voluntary contraction (MVC) force above the recruitment threshold of the isolated motor unit. Each subject maintained the two tasks for an identical duration (161 +/- 96 s) at a mean target force of 22.2 +/- 13.4% MVC (range: 3-49% MVC). The dependent variables included the discharge characteristics of the same motor unit in biceps brachii, fluctuations in motor output (force or acceleration), mean arterial pressure, heart rate, and rating of perceived exertion. Despite similar increases in the amplitude of the averaged EMG (% MVC) for the elbow flexor muscles during both tasks (P = 0.60), the rates of increase in mean arterial pressure (P < 0.001), rating of perceived exertion (P = 0.023), and fluctuations in motor output (P = 0.003) were greater during the position task compared with the force task. Consistent with these differences, mean discharge rate declined at a greater rate during the position task (P = 0.03), and the coefficient of variation for discharge rate increased only during the position task (P = 0.02). Furthermore, more motor units were recruited during the position task compared with the force task (P = 0.01). These findings indicate that despite a comparable net muscle torque, the rate of increase in the motor output from the spinal cord was greater during the position task.

Brain activation patterns during imagined stance and locomotion in functional magnetic resonance imaging

Posture and gait are sensorimotor actions that involve peripheral, spinal, and supraspinal structures. To investigate brain activity during stance and locomotion, 13 healthy subjects were asked to stand, walk, run, and lie down; subsequently, they were trained to imagine standing, walking, running, and lying [imagined lying as rest condition in functional magnetic resonance imaging (fMRI)]. Separate and distinct activation/deactivation patterns were found for the three imagined conditions: (1) standing imagery was associated with activation in the thalamus, basal ganglia, and cerebellar vermis; (2) walking imagery was associated with activation in the parahippocampal and fusiform gyri (areas involved in visuospatial navigation), occipital visual areas, and in the cerebellum; (3) running imagery caused a predominantly cerebellar activation in the vermis and adjacent hemispheres (six times larger than during imagination of walking or standing), but activations in the parahippocampal and fusiform gyri were smaller than during walking. Deactivations were found for walking and running, but not for standing imagery. They were located in the vestibular (posterior insula, superior temporal gyrus, supramarginal gyrus) and somatosensory (postcentral gyrus) cortex with right-hemispheric dominance. These findings support the concept of a hierarchical organization of posture and locomotion. Automated locomotion, for example, running, is based on spinal generators whose pace is driven by the cerebellar locomotor region. Deactivation in the vestibular and somatosensory cortex prevents adverse interactions with the optimized spinal pattern and sensory signals; this confirms earlier findings of a multisensory inhibition during unhindered locomotion. During slow walking, spatial navigation, mediated by the parahippocampal cortex, becomes more important. Postural control during standing involves a low intensity cerebellar activity and sensorimotor control via the thalamus and basal ganglia.

Modulation of responses of feline gamma-motoneurones by noradrenaline, tizanidine and clonidine

1. Effects of noradrenaline (NA) and the alpha2 agonists tizanidine and clonidine were tested on extracellularly recorded responses of gamma-motoneurones in deeply anaesthetized cats. Two types of responses were used; firstly, short latency phasic responses evoked by electrical stimulation of group II afferents in a muscle nerve and, secondly, tonic background discharges. 2. Responses evoked by group II muscle afferents were depressed when NA and tizanidine were applied ionophoretically close to a gamma-motoneurone and when clonidine was applied systemically. The number of spike potentials evoked by stimulation of these afferents decreased and their latencies increased. Responses evoked by flexor or extensor afferents in gamma-motoneurones innervating flexors or extensors were similarly depressed. 3. Tonic discharges were inconsistently and/or insignificantly affected by locally applied NA and tizanidine but were depressed by systemically applied clonidine. 4. Control tests indicate specific effects of NA and tizanidine application since similarly ionophoresed H+ ions did not change responses of gamma-motoneurones to stimulation of group II afferents, or only weakly enhanced their background discharges. Furthermore, serotonin ejected from a solution with a similar pH facilitated rather than depressed responses of gamma-motoneurones. 5. The results indicate that some antispastic effects of clonidine and tizanidine may be due to the depression of group II-evoked responses of gamma-motoneurones, resulting in weaker responses of muscle spindles to muscle stretches.

Supraspinal and segmental interactions

For the most part descending systems evoke movements through spinal interneurons interposed in reflex pathways. The advantage of this arrangement is that it ensures an integration of descending commands and proprioceptive and other exteroceptive feedback during the production of purposeful movement. It has also become clear that spinal reflex pathways can be reorganized during movement and that this could profoundly modify the effects of supraspinal commands on motor output. Recent experiments illustrate the existence and regulation of intrinsic motoneuron membrane currents that can dramatically change how motoneurons respond to descending commands. To understand how movements are controlled we must, therefore, understand both the activity of supraspinal motor systems and the spinal substrate upon which these commands are exerted.