Allan M. Smith

Relation of activity in precentral cortical neurons to force and rate of force change during isometric contractions of finger muscles

SummaryThe activity of single neurons within the hand area of the precentral motor cortex of primates was recorded during the performance of a maintained precision grip between the thumb and forefinger. The finger opposition forces were exerted against a strain gauge which allowed force changes to be studied under near isometric conditions. Task performance required the generation of a force ramp (the dynamic phase) and thereafter the maintenance of a stable force for one second (the static phase). Intracortical stimulation through the recording electrode was used to verify that the recordings were made from the appropriate somatotopographic area of the motor cortex.From a total of 221 recorded neurons, 76 were found to be either activated or deactivated during performance of the task. Among the 51 activated neurons, most discharged at higher frequencies during the dynamic phase, than during the static phase. The discharge of some of these neurons could be related to both force (F) and rate of force change (df/dt) whereas certain others could only be correlated with df/dt. The change in discharge frequency for these neurons generally occurred prior to the onset of EMG activity. Eight neurons were more active during maintained force than during the force ramp. The discharge frequency could not be correlated with df/dt and only one showed a significant positive relation to force. The change in discharge frequency for these neurons either coincided or occurred after the onset of EMG activity.

The cutaneous contribution to adaptive precision grip

Only after injury, or perhaps prolonged exposure to cold that is sufficient to numb the fingers, do we suddenly appreciate the complex neural mechanisms that underlie our effortless dexterity in manipulating objects. The nervous system is capable of adapting grip forces to a wide range of object shapes, weights and frictional properties, to provide optimal and secure handling in a variety of potentially perturbing environments. The dynamic interplay between sensory information and motor commands provides the basis for this flexibility, and recent studies supply somewhat unexpected evidence of the essential role played by cutaneous feedback in maintaining and acquiring predictive grip force control. These examples also offer new insights into the adaptive control of other voluntary movements.

Cerebellar cortical activity during antagonist cocontraction and reciprocal inhibition of forearm muscles

Monkeys were trained to perform a maintained isometric grip of the thumb and forefinger that elicited a simultaneous cocontraction of the antagonist muscles of the forearm. The same monkeys were also trained to flex and extend the wrist against a stop with the fingers extended and to maintain an isometric wrist position for 1.0-1.5 s. During wrist movement, some of the synergist forearm muscles contracted during both flexion and extension. However, during the maintained isometric wrist position, the prime mover and synergist muscles were reciprocally active or silent. In the culmen-simplex region of the cerebellar cortex bordering on the primary fissure, 62% of the Purkinje cells that were identified by the climbing fiber discharge and that changed firing frequency decreased activity during maintained prehension. Almost all of these same Purkinje cells were reciprocally active during isometric wrist flexion and extension, although three neurons had similar discharge patterns during movements in both directions. In contrast, 79% of the unidentified neurons recorded from the same region of the cerebellar cortex increased discharge frequency during prehension. In general, most of these same neurons had reciprocal patterns of discharge during wrist movement even though a few cells were active during the dynamic phase in both directions. Together, the Purkinje cells and the unidentified neurons with bidirectional response patterns were thought to be related to muscles active during both flexion and extension wrist movements. No cells were found that increased discharge with the static isometric wrist torque exerted in both directions. The discharge frequency of some Purkinje and some unidentified neurons could be shown to be related to prehensile force as well as wrist movement velocity and isometric wrist torque. These data suggest that the discharge of about two-thirds of the Purkinje cells related to forearm muscles located along the borders of the primary fissure may depend on whether antagonist muscles are activated reciprocally or coactively. As a consequence, these cells may play a role in the selection or alternation between either of these two modes of muscular contraction. The increased discharge of the remaining one-third of the Purkinje cells excited during antagonist coactivation may provide inhibition of nuclear cells to stabilize the posture at joints other than the wrist and fingers or, alternatively, they may act to reduce nuclear cell discharge in proportion to the intensity of cutaneous stimulation.

Locomotor deficits in the mutant mouse, Lurcher.

SummaryThe effect of total Purkinje cell degeneration on treadmill locomotion was studied in the cerebellar mutant mouse Lurcher. Other movements such as swimming and scratchting were also studied in order to evaluate the cerebellar control of rhythmic actions. Cinematographic and electromyographic recordings were taken from normal and Lurcher mice that were subsequently perfused to obtain a Purkinje cell count. Walking deteriorated progressively and was clearly abnormal in 30 day old Lurchers with 90% Purkinje cell degeneration. In adult Lurcher mice in which Purkinje cells were totally absent, walking was characterized by short steps with exaggerated hindlimb flexion in the swing phase. Also, both the interlimb step ratio, defined as the step length of the reference limb divided by the step length of the opposite limb, and the interlimb coupling, defined as the temporal relation of one footfall with respect to the footfall of another limb, varied more than in normal mice. Furthermore, the locomotion of Lurcher mice displayed increased vertical displacement of the hip and an inability to produce continuous step cycles without stumbling. Both the EMG onset relative to foot contact and the EMG burst duration were highly variable, and a greater overlap in the activities of antagonist muscles at the transition from ankle extension to flexion was evident. Although both walking and swimming involve cyclical limb movements, the disorganization of the cycle and the irregular EMG pattern seen in the Lurcher during walking were not observed during swimming. Furthermore, scratching was well executed in the Lurcher mice. However, a consistently higher tonic extensor activity at the ankle appeared during walking, swimming and scratching. These results suggest that, in contrast to swimming and scratching, the requirements of walking depend to a greater degree on a functional cerebellar cortex for successful performance.

The effects of muscimol inactivation of small regions of motor and somatosensory cortex on independent finger movements and force control in the precision grip

Abstract This study investigated the effects of inactivating small regions of the primary somatosensory (SI) and motor (MI) cortex on the control of finger forces in a precision grip. A monkey was trained to grasp and lift a computer-controlled object between the thumb and index finger and to hold it stationary within a narrow position window for 2 s. The grip force applied perpendicular to the object surface, the lifting or load force applied tangentially in the vertical direction, and the vertical displacement were sampled at 100 Hz. Also, the ability of the monkey to extract small pieces of food from narrow wells of a Klüver board was analyzed from video-tape. Preliminary single-unit recordings and microstimulation studies were used to map the extent of the thumb and index-finger representation within SI and MI. Two local injections of 1 µl each (5 µg/µl) of the GABAA-agonist muscimol were used to inactivate the thumb and index region of either the pre- or post-central gyrus. The precision grip was differently affected by muscimol injection into either SI or MI. MI injections produced a deficit in the monkey’s ability to perform independent finger movements and a general weakness in the finger muscles. Whole-hand grasping movements were inappropriately performed in an attempt to grasp either the instrumented object or morsels of food. Although the effect seemed strongest on intrinsic hand muscles, a clear deficit in digit extension was also noted. As a result, the monkey was unable to lift and maintain the object within the position window for the required 2 s, and, over time, the grip force decreased progressively until the animal stopped working. Following SI injections, the most obvious effect was a loss of finger coordination. In grasping, the placement of the fingers on the object was often abnormal and the monkey seemed unable to control the application of prehensile and lifting forces. However, the detailed analysis of forces revealed that a substantial increase in the grip force occurred well before any deficit in the coordination of finger movements was noted. This observation suggests that cutaneous feedback to SI is essential for the fine control of grip forces.

Internal models of the motor system that explain predictive grip force control

Only after injury, or perhaps prolonged exposure to cold that is sufficient to numb the fingers, do we suddenly appreciate the complex neural mechanisms that underlie our effortless dexterity in manipulating objects. The nervous system is capable of adapting grip forces to a wide range of object shapes, weights and frictional properties, to provide optimal and secure handling in a variety of potentially perturbing environments. The dynamic interplay between sensory information and motor commands provides the basis for this flexibility, and recent studies supply somewhat unexpected evidence of the essential role played by cutaneous feedback in maintaining and acquiring predictive grip force control. These examples also offer new insights into the adaptive control of other voluntary movements.

Distribution and terminal arborizations of cutaneous mechanoreceptors in the glabrous finger pads of the monkey.

Recent electrophysiological studies demonstrated that neurons in the somatosensory cortex of monkeys respond to tangential forces applied to glabrous skin. To unravel the peripheral basis for this cortical response, we determined the distribution of presumptive low‐threshold mechanoreceptors innervating the distal finger pads of monkeys. Endings were reconstructed in immunolabeled serial sections imaged by epifluorescence and confocal microscopy. Although classically implicated as cutaneous stretch receptors, no Ruffini corpuscles were found in the glabrous skin. Ruffini‐like endings were only detected at the base of the finger nails. Pacinian corpuscles were sparsely distributed in the deep dermis. Meissner corpuscles (MCs) in dermal papillary ridges had a comparably high density in the thumb, index, and fifth fingers. Each MC was innervated by several large‐caliber axons. Within the limits of our reconstructions, some of these axons terminated in only one MC, whereas others innervated several MCs. Merkel endings covered about 80% of the base of the intermediate epidermal ridges that form the pattern of fingerprints. In some cases, the distal tip of a Merkel‐related axon gave rise to a several terminal branches that supplied endings to tightly circumscribed (30–70 μm) clusters of Merkel cells. In other cases, the nodes of axons gave rise to en passant branches that formed extended chains of endings among Merkel cells spread over territories up to 300 μm long. Based on their relatively diffuse distributions, the axons that innervate multiple MCs or the axons with en passant Merkel terminations seem most suited to transduce tangential forces. J. Comp. Neurol. 445:347–359, 2002.

The activity of supplementary motor area neurons during a maintained precision grip

Two monkeys were trained to exert a precision grip of the thumb and forefinger and to maintain constant near-isometric force for a one-second duration. Both animals were trained to perform the task with about equal proficiency with either hand. A total of 134 neurons were recorded from the supplementary motor area (SMA) of the hemisphere contralateral to the performing hand. SMA neurons were identified by either the presence of peripheral fields on the contralateral arm or by consistent changes in discharge frequency during contralateral arm movement. Sixty-one cells demonstrated reliable changes in firing frequency during performance of the maintained precision grip. SMA neurons showed little tendency to discharge at higher frequency during force change rather than during maintained force. Only two neurons significantly increased firing frequency with increased finger force and no modulation of discharge related to rate of force change could be shown. The changes in spike frequency among SMA cells related to the arm were, on the average, about 100 msec after the onset of contraction in the forearm flexors and extensors of the wrist and fingers, although a contingent of cells discharging consistently before the onset of muscular activity was found.

Does the mutant mouse lurcher have deficits in spatially oriented behaviours

Lurcher mutant mice, in comparison to normal mice, had directional deficits in the Morris milk tank test and in a water-maze spatial alternation task. The lurcher mutants also showed an initial lack of spontaneous alternation and did not alternate in one behavioral condition when the inter-trial interval was lengthened. Lurcher mice were slower to learn a simple left/right position response to escape a T-maze by swimming although their motor coordination was good. Paradoxically, no deficit was observed in learning to select a left or right position for food in the same T-maze although the goal-directed locomotion was very ataxic. Overall, the lurcher mutants have difficulty in guiding themselves in the water toward a visible goal.

Spontaneous alternation and habituation in lurcher mutant mice

There is a spontaneous degeneration of Purkinje cells, granule cells and inferior olivary neurons in lurcher mutant mice. It was found that, in comparison to littermate controls, the lurcher mutants alternated less often in a discrete two-trial procedure of spontaneous alternation and did not habituate to maze stimuli in a T-maze. Results are discussed in terms of a possible role for the cerebellum in spatial learning.