Role of the primary motor and sensory cortex in precision grasping: a transcranial magnetic stimulation study.

Human precision grip requires precise scaling of the grip force to match the weight and frictional conditions of the object. The ability to produce an accurately scaled grip force prior to lifting an object is thought to be the result of an internal feedforward model. However, relatively little is known about the roles of various brain regions in the control of such precision grip-lift synergies. Here we investigate the role of the primary motor (M1) and sensory (S1) cortices during a grip-lift task using inhibitory transcranial magnetic theta-burst stimulation (TBS). Fifteen healthy individuals received 40 s of either (i) M1 TBS, (ii) S1 TBS or (iii) sham stimulation. Following a 5-min rest, subjects lifted a manipulandum five times using a precision grip or completed a simple reaction time task. Following S1 stimulation, the duration of the pre-load phase was significantly longer than following sham stimulation. Following M1 stimulation, the temporal relationship between changes in grip and load force was altered, with changes in grip force coming to lag behind changes in load force. This result contrasts with that seen in the sham condition where changes in grip force preceded changes in load force. No significant difference was observed in the simple reaction task following either M1 or S1 stimulation. These results further quantify the contribution of the M1 to anticipatory grip-force scaling. In addition, they provide the first evidence for the contribution of S1 to object manipulation, suggesting that sensory information is not necessary for optimal functioning of anticipatory control.

Motor cortex excitability after thalamic infarction.

Transcranial magnetic stimulation was used to map hand muscle representations in the motor cortex of a patient in whom infarction of the sensory thalamus deprived the sensorimotor cortex of sensory input. The threshold for activation of the motor cortex on the affected side was higher and the cortical representational maps of individual muscles were less well defined than those on the normal side. It is concluded that electrophysiological changes in cortical organisation can be demonstrated following withdrawal of, or imbalance in sensory afferent activity to the cerebral cortex in humans.

Stretch reflexes in the human masticatory muscles: a brief review and a new functional role.

Stretch reflexes play a vital role in fine-tuning movements and in automatically maintaining posture. This article briefly reviews the operation of the stretch reflex in the human masticatory system. The conventional approach of stretching muscles in an open-loop manner has yielded much valuable information on the operation of this reflex. In particular, it has revealed that stretching the jaw-closing muscles evokes a reflex response with two major components. The short-latency reflex is favoured when stretches are brisk, but slower stretches evoke an additional long-latency component. In the hand muscles, the long-latency response is transcortical: in the masticatory muscles, it is not. In addition to its role in servo-control of muscle length during chewing, the stretch reflex in the jaw-closing muscles maintains the vertical position of the mandible during vigorous head movements such as those that occur during running, jumping, hopping and other vigorous whole-body movements in which the head moves briskly up and down. This is an interesting model system in which to investigate stretch reflexes with natural stimuli under unrestrained, physiological conditions.

Does induction of plastic change in motor cortex improve leg function after stroke?

Combined peripheral nerve and brain stimulation ("dual stimulation") induces changes in the excitability of normal motor cortex. The authors sought to establish whether dual stimulation would also induce motor cortex plasticity and associated functional improvements in nine stroke patients with chronic stable hemiparesis. Following 4 weeks of daily dual stimulation, improvements were seen in some neurophysiological and functional measures. This technique may offer therapeutic opportunities in some stroke patients.

Cortical excitability is not depressed in movement-modulated stretch response of human thumb flexor.

There is strong evidence that the predominant pathway of the long-latency stretch reflex for flexor pollicis longus crosses the motor cortex. This reflex response is diminished during active thumb movements. We tested the hypothesis that this could be due to a decrease in the excitability of the transcortical component during movement. During isometric, concentric and eccentric thumb movements, transcranial magnetic stimulation (TMS) of the motor cortex was given at a time when the reflex signal was traversing the motor cortex. TMS was also given earlier in separate runs when the signal was traversing the spinal cord under each of the three contractile conditions. The electromyogram was analysed for non-linear summation between stretch responses and the potential evoked by the cortical stimulus. The response to TMS alone was uniform across the three types of contraction, and the lack of cortical involvement in the short-latency reflex was confirmed. The TMS-evoked response summed in a non-linear manner with the long-latency reflex response, confirming that the excitability of the motor cortex was increased as the reflex signal passed through it. The long-latency response was markedly depressed during isotonic compared with isometric contractions. However, the non-linear summation was not greater during the isometric contractions. Thus, the depressed reflex responses during isotonic movements do not stem from reduced motor cortical responsiveness or afferent input to the transcortical pathway, and may instead reflect modulation of cutaneous reflexes during isotonic contractions.

Patterns of muscle activation in human hopping.

In the present study, we examined the electromyogram (EMG) patterns of the soleus and medial gastrocnemius (MG) muscles during rhythmical, two-legged hopping to investigate the contributions of the monosynaptic short- and long-latency stretch reflexes during such a natural movement in human. During rhythmical hopping, soleus muscle is activated reflexly at near-monosynaptic latency by stretch resulting from passive ankle flexion upon landing. Soleus muscle also contracts voluntarily in order to launch the body into the next hop. This is part of the rhythmical bursts of activity producing the hops. Depending on the hopping interval, this phase of activation can follow the short-latency phase or precede landing at very short hopping intervals. In MG, there is an initial phase of activity that stiffens the muscle in preparation for landing, and continues through the contact phase. The monosynaptic reflex response to landing is usually superimposed on this activity. Depending on the hopping interval, both of these responses may be overlaid with activity that is time-locked to the take-off into the next hop, and serves to launch the body into the next hop. However, no evidence for a long-latency stretch reflex was found. In addition, the preferred hopping frequency for all subjects was about 2 Hz. This frequency is associated with a pattern of EMG activity the timing of which indicates that it balances the requirement for a comfortable landing from a hop with the optimal muscle activation required for launching the following hop.

Changes in corticomotor representations induced by prolonged peripheral nerve stimulation in humans.

OBJECTIVE: Manipulation of afferent input can induce reorganization within the sensorimotor cortex which may have important functional consequences. Here we investigate whether prolonged peripheral nerve stimulation can induce reorganization within the human motor cortex. METHODS: Using transcranial magnetic stimulation, we mapped the scalp representation of the corticospinal projection to hand muscles in 8 normal subjects before and after 2h of simultaneous repetitive electrical stimulation of the ulnar and radial nerves at the wrist. Control mapping experiments were conducted in 6 subjects. RESULTS: Following nerve stimulation, larger motor-evoked potentials were evoked from more scalp sites. The induced changes were most apparent in first dorsal interosseous, but were also seen in other hand muscles. The increases in area of the representational maps were accompanied by changes in the location of the optimal site for evoking responses in first dorsal interosseous, and changes in the centres of gravity of the maps. CONCLUSIONS: Prolonged afferent stimulation induces an increase in excitability of the corticospinal projection. This is accompanied by a significant shift in the centre of gravity of the stimulated muscles which we propose is evidence of a non-uniform expansion in their cortical representation.

Task-dependent control of human masseter muscles from ipsilateral and contralateral motor cortex.

Corticotrigeminal projections to human masseter motoneuron pools were investigated with focal transcranial magnetic stimulation (TMS). Responses in left and right masseter muscles were quantified from the surface electromyogram (EMG) during different biting tasks. During bilateral biting, TMS elicited motor evoked potentials (MEPs) in both masseter muscles. On average, the MEP area in the masseter contralateral to the stimulus was 39% larger than in the ipsilateral muscle, despite comparable pre-stimulus EMG in both muscles. MEPs elicited while subjects attempted unilateral activation of one masseter muscle were compared with those obtained in the same muscle during a bilateral bite at an equivalent EMG level. MEPs in the masseter contralateral to the stimulated hemisphere were significantly smaller during unilateral compared with bilateral biting. There was no significant difference in the size of ipsilateral MEPs during ipsilateral and bilateral biting. We conclude that the corticotrigeminal projections to masseter are bilateral, with a stronger contralateral projection. The command for unilateral biting is associated with a reduced excitability of corticotrigeminal neurons in the contralateral, but not the ipsilateral motor cortex. We suggest that this may be accomplished by reduced activity of a population of corticotrigeminal neurons which branch to innervate both masseter motoneuron pools.

Modulation of stretch-evoked reflexes in single motor units in human masseter muscle by experimental pain.

The interaction between muscle pain and motor function of the jaw has been examined in recent years, but the nature of the modulation of the short-latency stretch reflex by pain is not fully understood. In this study, the reflex responses to stretch were measured in single low-threshold motor units that were kept discharging at a constant frequency, before, during and after the induction of experimental pain in one masseter muscle by controlled infusion of hypertonic saline. The probability of evoking a reflex response in individual motor units in the painful muscle at near-monosynaptic latency was reduced by a mean of about 20%. However, the overall reflex response in the surface electromyogram of both the ipsi- and contralateral masseter muscles was greater during pain. This was apparently a secondary response to the pain-induced increase in pre-stimulus activity in the motoneurone pools of both muscles, because increased motoneurone excitability may facilitate stretch reflexes. It is concluded that the most likely explanation for the reduced reflex response of low-threshold masseter motor units during experimental pain is a tonic reduction in the fusimotor drive to the masseter spindles.

Changes in muscle responses to stimulation of the motor cortex induced by peripheral nerve stimulation in human subjects.

The aim of this study was to determine whether prolonged, repetitive mixed nerve stimulation (duty cycle 1 s, 500 ms on-500 ms off, 10 Hz) of the ulnar nerve leads to a change in excitability of primary motor cortex in normal human subjects. Motor-evoked potentials (MEPs) generated in three intrinsic hand muscles [abductor digiti minimi (ADM), first dorsal interosseous (FDI) and abductor pollicis brevis (APB)] by focal transcranial magnetic stimulation were recorded during complete relaxation before and after a period of prolonged repetitive ulnar nerve stimulation at the wrist. Transcranial magnetic stimuli were applied at seven scalp sites separated by 1 cm: the optimal scalp site for eliciting MEPs in the target muscle (FDI), three sites medial to the optimal site and three sites lateral to the optimal stimulation site. The area of the MEPs evoked in the ulnar-(FDI, ADM) but not the median-innervated (APB) muscles was increased after prolonged ulnar nerve stimulation. Centre of gravity measures demonstrated that there was no significant difference in the distribution of cortical excitability after the peripheral stimulation. F-wave responses in the intrinsic hand muscles were not altered after prolonged ulnar nerve stimulation, suggesting that the changes in MEP areas were not the result of stimulus-induced increases in the excitability of spinal motoneurones. Control experiments employing transcranial electric stimulation provided no evidence for a spinal origin for the excitability changes. These results demonstrate that in normal human subjects the excitability of the cortical projection to hand muscles can be altered in a manner determined by the peripheral stimulus applied.

Trial-to-trial fluctuations in H-reflexes and motor evoked potentials in human wrist flexor.

The H-reflexes and the motor potentials (MEPs) evoked by electromagnetic brain stimulation in the human wrist flexor were recorded over many trials. The responses from each stimulus at two steady levels of muscle activation were sorted into three groups, based on their amplitudes. The electromyogram (EMG) in each of these groups was rectified and averaged. The level of pre-response muscle activity was found to correlate with the amplitude of both the averaged H-reflexes and the averaged MEPs. This suggests that much of the amplitude fluctuations of both H-reflexes and MEPs can be attributed to moment-to-moment changes in the level of activity of the motoneurone pool. Overall, however, the amplitude of MEPs increased more rapidly than the amplitude of H-reflexes as the pre-stimulus EMG activity increased. This is probably because, while the amplitude of H-reflexes depends primarily on the level of motoneurone pool excitability, the amplitude of an MEP depends not only on this, but also on the excitability of the motor cortex, and the former is to some extent also dependent on the latter.

Studies of stimulus-evoked responses in single motoneurones in humans.

Surface electromyography (EMG) has been a powerful technique for studying reflex and other stimulus-evoked responses in the human nervous system. However, important additional insights can be gained into the operation of neural circuits by studying the responses of single motor units to various stimuli. In this paper, some of the advantages of single motor unit recording will be canvassed, and some examples of the application to this method to the study of reflex responses to sensory stimuli and brain stimulation will be presented.

Observations on the variability of the H reflex in human soleus.

H reflexes were evoked in human soleus by stimulating the tibial nerve at a constant intensity. Each trial was then assigned to one of three groups on the basis of the amplitude of its H reflex; all trials in each group were then full-wave rectified and reaveraged. There was a strong positive relationship between the amplitude of the H reflex and the level of electromyographic activity in the muscle at the time of onset of the H reflex, which reflects the activity of the motoneuronal pool when the afferent volley arrived. Thus, much of the variability of the H reflex is due to small changes in the level of activation of the motoneuronal pool during repeated trials. The steady torque preceding the H reflex was a poor predictor of the H-reflex amplitude, presumably because of the delay between the changes in the electrical activity of motoneurons and the mechanical outcome thereof.

Motor cortical control of human masticatory muscles.

The corticotrigeminal projections to masseter and anterior digastric motoneuron pools that are activated by TMS are bilateral, but not symmetrical. This conclusion is supported by whole-muscle data showing larger MEPs in the contralateral muscle with unilateral focal TMS, as well as evidence that TMS stimulation of one hemisphere may produce excitation in a masseter or digastric single motor unit while stimulation of the opposite hemisphere produced inhibition of the same motor unit. The asymmetry is particularly marked for masseter, in which the low-threshold motor units were most commonly excited with contralateral TMS and inhibited with ipsilateral TMS. Spike-triggered averaging of digastric motor unit activity revealed cross-talk in surface EMG recordings from digastric muscles, and no evidence that muscle fibres in both digastric muscles were innervated by a common motor axon. Narrow excitatory peaks in the PSTH of motor unit discharge elicited by TMS in masseter (either hemisphere) and digastric motor units (ipsilateral hemisphere) suggest a direct corticomotoneuronal projection. The contralateral projection to digastric motoneurons may include additional oligosynaptic connections, as judged by the broader peaks in the PSTH with contralateral TMS. The organisation of bilateral corticotrigeminal inputs revealed with TMS suggests that: (a) the contralateral hemisphere provides relatively more of the excitatory input delivered via the fast corticotrigeminal pathway for both masseter and digastric motoneuron pools, and (b) corticotrigeminal projections from either hemisphere are capable of contributing to the voluntary command mediating activation of masseter, and (to a lesser extent) anterior digastric muscles on one side, that is independent of the homologous muscles on the other side.

Bilateral cortical control of the human anterior digastric muscles.

Transcranial magnetic stimulation (TCMS) was used to determine the organization of cortical motor projections to the anterior digastric muscles in 12 normal human subjects. Two distinct types of potentials were evoked in anterior digastric with a figure-of-eight coil. A short-latency (3 ms) response appeared bilaterally on the surface electromyogram (EMG), but only ipsilaterally on intramuscular recordings: this was the result of direct stimulation of the ipsilateral trigeminal motor root. Motor evoked potentials (MEPs) were elicited in the anterior digastric muscles at variable onset latencies of around 10 ms by stimulation of scalp areas antero-lateral to the area for the first dorsal interosseous muscle of the hand. These were evoked bilaterally in relaxed anterior digastric muscles in six of the seven subjects. In the other subject, the responses in the relaxed muscle were exclusively ipsilateral. However, when the anterior digastric muscles were contracted, the responses were bilateral in all subjects. TCMS and spike-triggered averaging revealed that the bilateral responses were not due to the branching of axons from individual digastric motoneurones to muscles on each side. Because the digastric motor nucleus may contain separate populations of ipsi- and contralateral projecting motoneurones, it was necessary to study single motor-unit responses to TCMS to demonstrate a bilateral corticobulbar projection. The responses of 17 single motor units in the anterior digastric muscle to TCMS were recorded. All were activated by contralateral stimulation. Approximately 80% were also activated by ipsilateral TCMS, although one well-characterised motor unit was inhibited by ipsilateral TCMS. When bilateral activation was present, the ipsilateral responses were more secure than the contralateral responses, which may indicate an additional interneurone in the pathway to the contralateral motoneurone. The major conclusions from this study are that (1) the cortical representation of the anterior digastric muscle is antero-lateral to hand muscles; (2) the cortical projection to the anterior digastric muscles is bilateral; (3) the corticobulbar projection is stronger contralaterally than ipsilaterally but may involve at least one additional synapse; and (4) anterior digastric motoneurones do not branch to innervate the muscles bilaterally.

Movements modulate the reflex responses of human flexor pollicis longus to stretch.

The reflex responses to brisk, ramp stretch perturbations of the human flexor pollicis longus muscle (FPL) were recorded during isometric and slow concentric or eccentric contractions at similar levels of muscle excitation. The subjects flexed their thumb to push down against a thumb-rest, whose position was controlled by a servo-controlled motor. In different runs, the stretch perturbations were imposed when the thumb-rest was stationary (isometric) or was flexing or extending the interphalangeal joint of the thumb at a constant velocity, i.e. during concentric or eccentric contractions of FPL. The latency of the most prominent component of the electromyographic reflex in the isometrically contracting muscle was about 60 ms, measured from the command signal. The amplitude of this response was sharply reduced during the non-isometric contractions. While not dependent on the direction, this modulation of the reflex response increased with the speed of active movement of the interphalangeal joint (flexion or extension). The response was greatly reduced during concentric or eccentric movements as slow as 1.6 mm x s[-1] (approximately 5 degrees x s (-1) at the joint). When the force rather than the position of the thumb-rest was servo-controlled, the stretch response to perturbation again diminished with speed in a self-paced flexion task, compared with an isometric "hold" condition.

Control of motor units in human flexor digitorum profundus under different proprioceptive conditions.

1. Changing the posture of the human fingers can functionally 'disengage' the deep finger flexor muscle from its normal action on the terminal phalanx of the fourth (or third) finger. This enables the activity of the muscle to be studied both with and without its normal proprioceptive inputs. 2. Spike trains of long duration from pairs of concurrently active motor units in this muscle were recorded in both the engaged and disengaged hand postures. Subjects voluntarily kept one of the motor units (the 'controlled' unit) discharging at the same target frequency in both postures. The strength of short-term synchrony, the strength of common drive, and the variability of discharge of these pairs of motor units were determined in both postures. 3. All subjects reported that the effort required to activate the motor units in the disengaged hand posture was substantially greater than in the normal engaged posture. 4. Short-term synchrony, which is a function of common corticospinal inputs to pairs of motor units, was similar in both hand postures. However, the strength of common drive was significantly decreased when the muscle was disengaged. Although the neural substrate for common drive is not known, this observation suggests that proprioceptive feedback is involved either directly or indirectly. 5. Although the discharge rate of the 'uncontrolled' motor units increased when the muscle was disengaged, the variability of discharge of these and the 'controlled' motor units increased significantly. This supports the idea that the precision with which fine motor tasks can be performed is improved when proprioceptive feedback is intact.

Electrophysiological observations on an unusual, task specific jaw tremor.

A patient with no other neurological signs or symptoms presented with a prominent tremor restricted to the mandible. This 5-6 Hz tremor was interesting in that it was normally confined to the digastric muscles and was highly task specific. In the course of her normal daily activities, it began only when the patient drank from a cup or glass. The localisation of this tremor to a muscle that has no muscle spindles and no reciprocal inhibitory reflexes suggests that such tremors must be capable of being generated centrally.

Influence of muscle blood flow on fatigue during intermittent human hand-grip exercise and recovery.

1. The influence of muscle blood flow on fatigue and recovery was studied in the forearm muscles of eight male subject performing a powerful isometric hand-grip exercise. The exercise was performed with the exercising forearm normally perfused and, on a separate occasion, with its blood flow occluded with a sphygmomanometer cuff. 2. In the no cuff condition, peak force declined to an initial plateau at 40-50% of the maximal voluntary grip force (MVC). When perfusion was occluded, the force decline was similar during the first minute of exercise, then force fell rapidly to exhaustion. 3. In a separate experiment to investigate the mechanisms underlying the plateau in force loss, occlusion of blood flow during the force plateau phase resulted in a rapid decline in force to exhaustion. 4. Recovery of peak force after the cuff exercise was significantly greater during the initial 3.5 min of recovery than after no-cuff exercise. After this time, recovery was similar for both conditions. 5. Muscle blood flow occlusion during intermittent exercise profoundly reduces endurance without prolonging recovery. Recovery time may depend on the duration and energy cost of the exercise rather than on the degree of force loss. 6. The present study suggests that the fall in muscle force induced by a continuous MVC is a combination of profound short-term fatigue in anaerobic muscle fibres due to the consumption of their short-term energy supplies, plus a decline in force production by aerobic muscle fibres that is the consequence of hypoxia. Thus, MVC may not be a good model of fatigue occurring under submaximal conditions, as hypoxia of type I fibres is unlikely to occur under physiological conditions in which muscle contractions are usually intermittent.

Estimating post-synaptic potentials in tonically discharging human motoneurons.

The activity of single motor units in human muscles can be recorded with relative ease, and the spike train of a single motor unit precisely reflects the spike train of the parent motoneurone. This has led to the proposal of a number of methods to estimate stimulus-evoked post-synaptic potentials in human motoneurones. All of these methods rely on manipulating the spike trains of motor units over a number of trials. All are based on a number of assumptions, all have limitations, and none so far have passed the test of a direct comparison of the estimate of the shape of the post-synaptic potential with a direct intracellular measurement of it. These techniques are summarised in this review.