movement, control of

movement, control of While walking to the station with a companion, we walk, navigate obstacles, maintain balance, carry objects, talk, and gesticulate. The performance of these multiple motor tasks is seemingly effortless, although they involve the co-ordinated activity of thousands of motor units in dozens of different muscles. By contrast, after damage to the brain's motor control system, the performance of even simple movements can be very exhausting and difficult; lesions of different parts of the system leave us with characteristic and often devastating deficits. We are unaware of much of the finer aspects of our own motor co-ordination and in fact have very little insight into how we perform a complex skilled act, such as riding a bicycle. We can conclude that for effortless performance of habitual tasks, our motor control system is much more sophisticated than one requiring constant supervision by the conscious brain. The distributed nature of the motor control system reflects the complexity of human movement, and this system comprises many different structures within the brain and spinal cord. The control system must ultimately act through the ‘final common path’: the discharge of motor neurons in the brain stem and spinal cord, and the mechanical response of the muscles they innervate. The interface between the neural ‘command’, ultimately expressed in terms of the pattern of motoneuronal discharge, and the mechanical response of the skeleto-muscular apparatus is a complex one. In engineering terms, understanding the motor system requires how the ‘controller’ delivers these neural commands to the ‘plant’; namely, the muscles, tendons, ligaments, bones, and joints. The enormous variety and type of human movement, including our unique capacity to speak, must all be performed with the same ‘plant’; although the number of muscles is large, the enormous motor repertoire must be a reflection of the flexibility of the motor control system. This repertoire can be greatly expanded and refined by training and practice — as demonstrated, for example, by musicians, athletes, artists, and surgeons.

The process of motor control

Consider the act of reaching and grasping a glass of water, raising it to the lips, and drinking. Models of motor control envisage the process as three different stages: the idea or plan of action to achieve a motor goal; the programme required to bring this plan about; and, finally, the execution of the movement. In the example given, the idea describes the objective or goal of the movement: acquiring the glass of water. This could be achieved in a variety of ways: with either arm, or by gripping the glass between the teeth. Thus the idea of the movement is quite general and does not need to be mapped out in terms of specific muscles or joints. It is known that some parts of the brain involved in motor control, such as the posterior parietal lobe of the cerebral cortex, are more concerned with the idea or plan, than with the selection of which limb or set of muscles to use. The Russian cyberneticist Bernstein, one of the pioneers of the study of human movement, recognized that when an action is carried out, then some feature of the plan will be present irrespective of the particular effector used to perform it. This common plan is assumed to underlie the phenomenon of equivalence of movement, by which a movement pattern, such as your own signature, is preserved whether you write it with the pen held in the dominant or in the non-dominant hand, by the foot or even the mouth. The programme of the movement must control the entire motor act: once the plan has been adopted then the whole sequence of movements is expressed. The reaching movement might employ the right hand, passing along a straight line path towards the glass. To programme this movement, it is proposed that the brain has to solve the inverse kinematics problem, calculating the timing and scale of changes in angle of appropriate joints (shoulder, elbow, and wrist) that will be needed to make the hand follow the selected path of movement. Once the desired angular changes are specified, the programme must define how the muscles are to produce them. This is not a trivial problem, since the torque that contraction of a muscle exerts at a joint will, in general, also have mechanical influences on remote joints (look at what happens to your wrist during rapid elbow flexion, for example). Thus the solution of this inverse dynamics problem is exceedingly complex. The forces and movements resulting from a neural command to contract a given muscle will vary greatly depending, for instance, on the length of the muscle, the speed at which it is shortening, and the contractile state of other muscles acting with it (its synergists) or against it (its antagonists). The execution of the movement requires the activation of selected sets of muscles in a manner that will achieve the objective of the programme. This may require co-ordination of groups of muscle synergists within one limb, or between several limbs, trunk, head, and eyes. The posture at the onset of the movement (for example, your arm resting on the table before reaching for a drink) must be taken into account, as must the desired velocity, acceleration, and deceleration of the movement, such that your hand arrives accurately at the glass, rather than under- or over-reaching it, or knocking it over.

Importance of sensory input

An important aspect of motor control is the incorporation of sensory input into the different parts of the idea–programme–execute sequence. Sensory inputs from the eyes, from the ears, and from skin, muscle, and joint receptors can characterize the location, size, shape, weight, and texture of the object forming the goal of the movement (in our example, the glass of water). Our perception of the expected attributes of the task can exert a powerful influence over performance (as, for example, in the size–weight illusion). There is increasing evidence that while some movements may rely on sensory feedback for their control, most rely on a ‘feedforward’ mechanism — a neural process based entirely on an ‘internal model’ within the brain of the required movement: that is, the entire motor action has an internal representation. Feedforward mechanisms are evidently essential for rapid movements (in which there is no time for feedback to play any part) and predominate in the execution of highly predictable and well-practised movements. They are also critical for movements of the eyes, such as saccades (very small ‘scanning’ movements), where no external loads or disturbances are likely to impede the production of the desired movement.

However, it is important to realize that sensory feedback must be used to create the brain's internal model of the movement in the first place. Feedback is important for updating and refining the model and for controlling unpredictable or novel movements (e.g. reaching out and lifting an object while underwater), and this process contributes to the acquisition of new motor skills. Through disease, some individuals completely lack any somatic sensations — they have no incoming information from the body — but retain a normal motor innervation. They can perform complex movements under continuous visual control, but have great difficulty in doing so. In normal subjects, any movement generates a large amount of sensory feedback and an important function of the internal model may be to predict the type and time course of this movement-induced sensory input or ‘reafference’. For example, the model could cancel the predicted cutaneous inputs experienced when the hand first contacts the glass of water. If the glass is unexpectedly slippery, warm, or heavy, an error signal will arise between the predicted and actual sensory input, and this will generate a corrective action. Some movements, such as those that occur during tactile exploration of an object, are carried out in order to generate such reafference and provide further information about the physical properties of the explored object.

The ‘degree of freedom’ problem

There are very many different muscles in the body and this means that the control system could adopt a very large number of possible solutions to achieve its goal. Bernstein recognized that the adoption of functional muscle synergies helps to solve this ‘degrees of freedom’ problem. These synergies are a well-recognized and important part of our motor control system, and are used for the co-ordination of activity within a limb (such as stabilization of the elbow joint while moving the wrist and fingers), between limbs (such as weight-bearing by one leg during walking), and for the co-ordination of eye, head, and body movements during orientation to a visual or auditory stimulus of interest (for example, when someone calls your name). Perhaps the most impressive example is the co-ordination of muscles in the face, mouth, tongue, larynx, chest, and abdomen during speech. Interestingly, many central motor structures, including the primary motor cortex and cerebellum, appear to have a motor representation that involves control of muscle groups, rather than single muscles. All purposeful motor acts involve the contraction of multiple muscles.

The motor network of the CNS

The different parts of the central nervous system concerned with movement control are organized in a distributed fashion and are generally considered to act in a parallel rather than in a hierarchical fashion: damage to individual structures, such as the lateral cerebellum or motor cortex, does not abolish the capacity to move. Mechanisms within the spinal cord and brain stem are concerned with reflex activity and integration of spinal and supraspinal control. Complex patterns of movement, such as locomotion and swallowing, are encoded within specific groups of connected interneurons referred to as ‘central pattern generators’ (CPGs). The CPG contains the complete spatio-temporal pattern of the complex act, and although it can act independently of peripheral sensory inputs and of descending motor pathways, these can influence and modulate the activity of the CPG. Descending motor pathways from the brain stem, midbrain, and cerebral cortex exert specific influences over groups of interneurons and motor neurons concerned with trunk, shoulder, pelvic girdle, and distal limb (hand or foot) movements. All of the higher order motor centres, including the cerebellum, basal ganglia, and cerebral cortex, must exert their respective influence through these pathways. The human corticospinal tract is derived from extensive regions of the frontal and parietal lobes of the brain (including the primary motor and premotor cortex, and the supplementary motor area); it also includes input from part of the limbic system, associated with innate and emotional behaviour, and from the sensory cortical areas. Measurement of regional cerebral blood flow (by the imaging techniques of positron emission tomography (PET) or functional magnetic resonance imaging (fMRI)) show that some or all of these areas mentioned are active in human volunteers performing a variety of complex movements. The corticospinal tract is the largest single descending nerve pathway from the brain down to the spinal cord. Its nerve endings at synapses within the spinal grey matter enable the control of sensory input and the modulation of reflex activity as well as the generation of movement. The human corticospinal tract, like that in other primates, is characterized by direct, single nerve fibre-to-nerve cell (monosynaptic) connections with the spinal motoneurones, which go out to all the voluntary muscles, and this is thought to be of particular importance for the capacity to perform skilled hand and finger movements.

Interaction among cerebellum, basal ganglia and cerebral cortex

The motor areas of the cerebral cortex also influence sub-cortical structures from which other descending motor pathways arise (e.g. from the vestibular apparatus and from the brain stem to the spinal cord). The organization of the motor system is characterized by ‘re-entrant loops’. This means that the output of a particular structure is sent to a number of other regions which process this information and then feed it back, directly and indirectly, to the structure from which it originated. The two most important loops are those linking the areas of cerebral cortex concerned with movement with the cerebellum and basal ganglia; wide regions of the cerebral cortex send nerve fibres to these structures and the motor areas of the cerebral cortex receive a massive returning input from them via relays in the thalamus. There is detectable activity in the brain as much as a whole second before the onset of a voluntary movement; this is a long period with respect to nerve conduction velocity, so there is time to process information through these loops several times before the final command goes out. Transmission of information is altered by training and practice; pathways in the cerebellum have been shown to exhibit considerable plasticity — forging new connections — and the cerebellum has an essential role in the learning and co-ordination of movement.

Damage to the different levels of the motor system hierarchy causes characteristic changes in movement performance: for instance, patients with damage to the basal ganglia may exhibit paucity or slowness of movements, also associated with rigidity and tremor (Parkinson's disease), while in other conditions there may be uncontrolled and involuntary movements, called dyskinesias. Patients with cerebellar damage show deficits of timing and co-ordination. Damage to the primary motor cortex and its descending motor pathways often results in the complete loss of skilled hand movements, whilst disturbances of head and body posture result from lesions of other descending pathways, such as the vestibulospinal tract. These results can best be interpreted in terms of the site of termination of the different motor pathways within the spinal cord.

Neurophysiological recordings in experimental animals show that neurons in different parts of the motor system are active before and during movement and that their rate of discharge can encode the different parameters, such as the force or direction of the intended movement. These same neurons often respond to afferent feedback from peripheral receptors. The encoding of movement parameters at the single cell level is not sufficiently precise to explain the accuracy of movements such as pointing to a target, and it is probable that whole assemblies of neurons co-operate for this purpose. There is also evidence that neurons show task-specific activity, and, in some of the ‘higher’ cortical areas, they can encode complex sequences of movement.

R. N. Lemon

Bibliography

Porter, R. and and Lemon, R. N. (1993). Corticospinal function and voluntary movement. Oxford University Press.
Rosenbaum, D. A. (1991). Human motor control. Academic Press, San Diego.
Rothwell, J. C. (1994). Control of human voluntary movement. Chapman and Hall, London.

See nervous system.See also basal ganglia; brain; brain stem; cerebellum; cerebral cortex; musicianship and other finger skills; skeletal muscle; proprioception; spinal cord.