Research reportSpatial zones for muscle coactivation and the control of postural stability
Introduction
The description of reciprocal and simultaneous patterns of agonist and antagonist muscle activation is considered a fundamental way of understanding motor function. Muscle activity patterns are commonly studied in terms of the temporal aspects and magnitudes of electromyographic (EMG) bursts during many different types of movements from fast single-joint movement to locomotion (e.g. 6, 66). EMG patterns have also been found to correlate with different spatial characteristics of movement such as direction and magnitude 22, 44.
A theoretical framework for the description of a mechanism underlying the specification of spatial characteristics of muscle activation has been proposed in the λ model of motor control 17, 18. We hypothesize that depending on the motor task, the angular range of a joint may be subdivided into zones in which agonist and antagonist muscles are coactive, only one group of muscles is active or neither group is active. According to this hypothesis, central commands may change the size and location of these spatial zones in the angular coordinates defined for each joint (single-joint movements) or group of joints (multi-joint movements) [18].
In the λ model for single-joint movement, at least two central commands have been defined which regulate the spatial characteristics of agonist and antagonist muscle activation (Fig. 1). These central commands may be associated with independent components of mono- or polysynaptic influences from descending systems onto flexor and extensor motoneurons. In other words, descending influences may, although not necessarily always, be specified independently of current events in the periphery. The “reciprocal” command (R) combines facilitation of agonist motoneurons with inhibition (or disfacilitation) of antagonist motoneurons. The “coactivation” command (C) simultaneously facilitates motoneurons of both muscle groups. In the λ model the two types of commands can be measured in terms of position-dimensional variables. In isolation, the R command specifies the threshold angle (R) at which the transition of agonist to antagonist activity or vice versa occurs (Fig. 1, upper panel). A second command (C) which may occur with R, specifies an angular range in which agonist and antagonist muscles may be simultaneously active (coactivation zone) if C>0 (Fig. 1, middle panel) or silent (silent zone) if C<0 (Fig. 1, lower panel). In other words, the C command separates agonist and antagonist thresholds such that angle R occurs between them. In the case when C>0, if no change in position is to occur, the C command should separate the thresholds so that the activity of agonist and antagonist muscles produces equal and opposite torques. In other words, at position R, the joint would remain motionless if the C command changes from zero to a positive value. Under constant central commands, if the joint is deflected from position R, active muscle torque elicited by external forces would tend to bring the joint back to R. The sign of the net joint torque produced by each muscle group to counteract external forces changes at angle R (Fig. 1, middle panel). Thus, although the R command when C≠0 no longer represents the threshold angle for the pure transition of activity from one group of muscles to the other, it still represents the referent angle which influences muscle recruitment and sets the location of the coactivation or silent zone in the angular range. Control inputs to motoneurons may shift the point (R) and/or change the width of zone (C).
A given combination of R and C commands is associated with a single-valued relationship between static joint torque and angle, called the invariant characteristic (IC) of the joint (Fig. 1, compare solid curves in different panels). The term “invariant” implies that for all combinations of muscle torques and angles for this IC the values of the R and C commands are the same. It does not imply that the shape of the IC is constant when central commands are changed. Similarly, it does not imply that tonic EMG levels are the same for different points on the IC [18].
Since R and C commands have the dimension of position (angle), the understanding of motor control processes in the λ model is different from the traditional one in which control processes are considered in terms of reciprocal and coactivation EMG patterns for agonist and antagonist muscles during different motor tasks. R and C commands are, in essence, independent of EMG patterns (but not vice versa [18]). For example, when C=0, the R command defines the angle at which the transition from agonist to antagonist activity occurs. Whether or not the EMG transition actually occurs depends on the spatial relationship between the actual joint position (Θ) and R. For example, for a given IC, passive extension from an angle less than R to one greater than R (Fig. 1, upper panel) will result in the transition of activity from the extensor to the flexor muscle group. On the other hand, passive extension in ranges less than R will result only in modulation of activity in one muscle group without any switching between muscle groups. With increasing speed of extension, phasic reflexes may modify this behavior (see [18]). At the same time, muscle activation also depends on the C command which specifies the width of the coactivation zone. An example of the latter can be seen in Fig. 1 (middle panel). If the final position of the joint is such that the load is balanced inside the coactivation zone (filled circles), both agonist and antagonist muscles will be tonically active. In contrast, although the coactivation zone may be present, the load may be balanced outside this zone (open circles) so that only one muscle group will be active. These examples illustrate a general notion of the λ model: EMG patterns are not programmed but emerge from the interaction of the central control signals, proprioceptive feedback, intrinsic muscle structures and external forces [1].
Both R and C commands influence the net joint stiffness defined as the slope of the IC at a given operating point. The wider the coactivation zone defined by the C command, the steeper the slope of the torque/angle characteristic in that zone (compare solid lines in upper and middle panels, Fig. 1). Since the shape of the IC is non-linear and the slope increases with increasing muscle torque depending on the difference between the actual angle and angle R (Fig. 1), the R command may also affect stiffness. Imagine that the threshold angle of the flexor muscles, λf, is shifted to the left by an R command when the joint is in an initial position, Θi (Fig. 1, middle panel). After the shift, a new operating point on the IC is attained in which the amount of torque and, as a consequence, stiffness associated with position Θi is greater. Thus the regulation of stiffness may result from the modification of the operating point on the IC by the R command. Stiffness influences the stability of posture and movement [57]. Stability also depends on parameters influencing velocity-dependent characteristics of sarcomeres and proprioceptive feedback.
The framework of the model can be used for the understanding of the control of stability by comparing motor behavior of healthy subjects and patients with sensorimotor dysfunction. In particular, among other sensorimotor disturbances resulting from hemispheric stroke, the ability to produce smooth movement is impaired [49]. This occurs along with enhanced agonist/antagonist muscle co-contraction 36, 38and considerable slowing of movement [49]. Abnormal co-contraction during goal-directed movements such as reaching and locomotion may be associated with diminished agonist motor unit activation in these patients [10], impaired antagonist inhibition [38]or both [36]. In addition, weakness [5], altered mechanical properties of motor units [39], improper spatial and temporal muscle recruitment [20]and disruption in the organization of segmental reflex activity [9]may play a role in the appearance of abnormal coactivation during movement.
Spastic muscles in hemiparetic subjects may be characterized by significantly increased stretch reflex activity 3, 55which may be due to a decrease in the stretch reflex threshold and to limitations in its central regulation 40, 48, 51. These findings have led to the hypothesis that the regulation of reciprocal muscular activation and muscle coactivation may also be affected in these subjects.
Based on the suggestion of the λ model regarding the spatial zones for different patterns of muscle activation, we investigated how such zones were used in the postural control of the elbow joint in normal subjects and in those with postural control deficits due to unilateral stroke. Such data may improve our understanding of normal motor control and impaired control following lesions in the central nervous system. Some of the data have been presented in abstract form [52].
Section snippets
Experimental procedures
The forearm was placed on a horizontal manipulandum and the hand and forearm were stabilized in the neutral position between pronation and supination in a bi-valve splint adjusted by velcro straps. The flexion/extension axis of the elbow joint was aligned vertically with the axis of rotation of a torque motor (Mavilar Motors, MT 2000). Each initial position (approximately 130° and 100° flexion of the elbow; full extension being defined as 180°) was achieved by lining up a vertical cursor within
Kinematics, EMG patterns and torque/angle characteristics
In healthy subjects, unloading resulted in a silent period and an after-volley in the agonist muscles (BB and BR) and a stretch response in the antagonist muscles (TB and AN; Fig. 3). Following the dynamic phase of unloading, a new final combination of joint position and muscle torque was established (see torque and position traces) with correspondingly new levels of agonist and antagonist EMG activity. Mean tonic EMG levels for flexors and extensors before and after unloading were measured in
Basic results
We recorded EMG signals and torque/angle characteristics from two initial positions of the arm in healthy and hemiparetic subjects. In healthy subjects and in hemiparetic subjects with mild symptoms, agonist muscle activity systematically decreased whereas antagonist muscle activity increased with joint flexion implying length-dependent regulation of muscle activity associated with a tonic stretch reflex 19, 56, 59. Stable final positions were associated with coactivation or small silent zones
Acknowledgements
We would like to thank Dr. Anatol G. Feldman for his useful insights and comments related to this study. We also gratefully acknowledge the support of the National Science and Engineering Research Council of Canada and Medical Research Council of Canada. M.D. was a summer research fellow of the Fonds de la Recherche en Santé du Québec.
References (66)
- Adamovich, S.V., Levin, M.F. and Feldman, A.G., Central modifications of reflex parameters may underlie the fastest arm...
- Asatryan, D.G. and Feldman, A.G., Functional tuning of the nervous system with control of movement or maintenance of a...
- Ashby, P. and Verrier, M., Neurophysiologic changes in hemiplegia, Neurology, 26 (1976)...
- Ashworth, B., Preliminary trial of carisoprodal in multiple sclerosis, Practitioner, 192 (1964)...
- Bourbonnais, D. and Van den Noven, S., Weakness in patients with hemiparesis, Am. J. Occup. Ther., 43 (1989)...
- Brown, S.H. and Cooke, J.D., Initial agonist burst is modified by perturbations preceding movement, Brain Res., 377...
- Brunnstrom, S., Movement Therapy in Hemiplegia. A Neurophysiological Approach, Harper and Row, New York, 1970, 192...
- Burke, D., Stretch reflex activity in the spastic patient. In: P.A. Buser, W.A. Cobb and T. Okuma (Eds.), Kyoto...
- Burke, D., Spasticity as an adaptation to pyramidal tract injury. In: S.G. Waxman (Ed.), Advances in Neurology. Vol....
- Colebatch, J.G., Gandevia, S.C. and Spira, P.J., Voluntary muscle strength in hemiparesis: distribution of weakness at...
Cited by (90)
Synergies Stabilizing Vertical Posture in Spaces of Control Variables
2022, NeuroscienceEffects of hand muscle function and dominance on intra-muscle synergies
2022, Human Movement ScienceCitation Excerpt :Within this theory, the neural control of an effector acting along a certain coordinate can be described with two command, reciprocal (R-command) and co-activation (C-command). The two commands have been assumed to form a hierarchy (Levin & Dimov, 1997) with the hierarchically higher R-command defining a spatial goal for the effector and the C-command defining a spatial range where both agonist and antagonist muscles are activated simultaneously. By its definition, the R-command leads to better conditions for activation of one of the muscles within the agonist-antagonist pair and worse conditions for activation of the other muscle.
Central pattern generator and human locomotion in the context of referent control of motor actions
2021, Clinical NeurophysiologyQuantifying muscle coactivation in individuals with incomplete spinal cord injury using wavelets
2020, Clinical BiomechanicsAntagonist co-activation during short and medium latency responses in subjects with chronic ankle instability
2018, Journal of Electromyography and Kinesiology