Chapter 27 Fundamentals of Motor Systems

Sten Grillner

“All mankind can do is to move things … whether whispering a syllable or felling a forest.”—Sherrington

This quote is a reminder of the basic fact that all interactions with the surrounding world are through the actions of the motor system. When a human baby is born, it is a sweet but very immature survival machine, with a limited behavioral repertoire. It is able to breathe and has searching and sucking reflexes so that it can be fed from the mother’s breast. It can swallow, vomit, and process food, and it can cry for attention if something is wrong. A baby also has a variety of protective reflexes that mediate coughing, sneezing, and touch avoidance. These different patterns of motor behavior are thus available at birth and are due to innate motor programs (Fig. 27.1).

Figure 27.1 Motor development of the infant and young child. The pattern of maturation of the motor system follows a characteristic evolution. Two months after birth a child can lift its head, at 4 months it sits with support, and subsequently it is able to stand with support; later it crawls, stands without support, and finally walks independently. The approximate time at which a child is able to perform these different motor tasks is indicated above each figure. The variability in the maturation process is substantial. Modified from M. M. Shirley.

During roughly the first 15 years of life, the motor system continues to develop through maturation of neuronal circuitry and by learning through different motor activities. Playing represents an important element both in children and in young mammals such as kittens and pups. During the first year of life, the human infant matures progressively. It can balance its head at 2 to 3 months, is able to sit at around 6 to 7 months, and can stand with support at approximately 9 to 12 months. The coordination of different types of posture, such as standing, is a complex motor task to master, with hundreds of different muscles taking part in a coordinated fashion. Sensory information contributes importantly, in particular from the vestibular apparatus, eyes, muscle, and skin receptors located at the soles of the feet. This development represents to a large degree a maturation process following a given sequence but with individual variability among different children. When the postural system has evolved to a sufficient degree, a child is able to start walking, which requires that the body posture is maintained while the points of support are changing by the alternating movements of the two legs. In common language the child is said to “learn” to walk, but in reality a progressive maturation of the nervous system is taking place. Identical twins start to walk essentially at the same time, even if one has been subjected to training and the other has not. At this point the motor pattern is still very immature. Proper walking coordination followed by running appears later, and the basic motor pattern actually continues to develop until puberty. The fine details of the motor pattern are adapted to the surrounding world, but also to modification by will. The basic motor coordination underlying reaching and the fine control of hands and fingers undergo a similar characteristic maturation process over many years.

While the newborn human infant is comparatively immature, other mammals, such as horses and deer, represent another extreme. The gnu, an African buffalo-like antelope, needs to run away in order to survive attacks of predators such as lions. The young calf of the gnu can stand and run directly after birth and has been reported to be able to gallop ahead 10 minutes after delivery, tracking the running mother (Fig. 27.2). Clearly the neural networks underlying locomotion, equilibrium control, and steering must be sufficiently mature and available at birth, needing minimal calibration. This is astounding. A similar range of maturity is present in birds. To get out of the egg, a chick makes coordinated hatching movements to open up the eggshell to subsequently lift off the top of the egg and to stand up to walk away on two legs following the mother hen and start picking at food grains. Most birds are more immature when hatching, but after a few weeks, they leave the nest flying rather successfully, for the first time in their life, and thus without any previous experience.

Figure 27.2 Some animals are comparatively mature when they are born. Ten minutes after the calf of the gnu, a buffalo-like antelope, is born, it is able to track its mother in a gallop. This means that the postural and locomotor systems are sufficiently mature to allow the young calf to generate these complex patterns of motor coordination at birth. There is little time to calibrate the motor system after birth and obviously no time for learning. Courtesy of Erik Tallmark.

In addition to the basic motor skills such as standing, walking, and chewing, humans also develop skilled motor coordination, allowing delicate hand and finger movements to be used in handwriting or playing an instrument or utilizing the air flow and shape of the oral cavity to produce sound as in speech or singing. The neural substrates allowing learning and execution of these complex motor sequences are expressed genetically and are characteristic of our species. What is learned, however, such as which language one speaks or the type of letters one writes, is obviously a function of the cultural environment.

Basic Components of the Motor System

Motoneurons and Motor Units

The motoneurons that control different muscles are located in different motor nuclei along the spinal cord and in the brainstem. Each motoneuron sends its axon to one muscle and innervates a limited number of muscle fibers. A motoneuron with its muscle fibers is referred to as a motor unit. The muscle fibers of each motor unit have similar contractile properties and metabolic profile. The muscle fibers in different muscles are composed of three main types specialized for different demands, such as a continuous effort as in long-distance running (slow motor units) or fast explosive movements, such as lifting a heavy object [fast motor units (two subtypes)].

Motoneurons are activated by interneurons of different motor programs or reflex centers and by descending tracts from the forebrain and the brainstem. Thus motoneurons supplying different muscles can be activated with great precision by these different sources that together determine the degree of activation, as well as the exact timing of the motoneurons of a given muscle.

Sensory Receptors Are Important in Movement Control

Signals from sensory receptors are used by the motor system in a number of ways.

1. Sensory signals can trigger behaviorally meaningful motor acts such as withdrawal, coughing, or swallowing reflexes.

2. Sensory receptors may contribute to the control of an ongoing motor pattern and influence the switch from one phase of movement to another. For instance, in the case of breathing, sensory signals from lung volume receptors control when inspiration is terminated. In walking, sensory signals related to hip position and load on the limb help regulate the duration of the support and swing phases of the step cycle and correct for perturbations.

3. Sensory signals may also be more specific and influence only the level of activity of one muscle or a group of close synergists. Muscle receptors play a particularly important role in this context. They are of two types: Golgi tendon organs that sense the degree of contraction in a muscle (located in series with the muscle fibers at their insertion into the tendon), and muscle spindles that signal the length of a muscle (located in parallel to the muscle fibers), as well as the speed of changes in length. Moreover, the sensitivity of muscle spindles can be regulated actively through a separate set of small motoneurons, referred to as γ-motoneurons (in contrast to the larger α-motoneurons that control muscle contraction). These two muscle receptors have fast conducting afferent axons that provide rapid feedback to the spinal cord and take part in the autoregulation of the motor output to a given muscle. Fast muscle spindle afferents provide direct monosynaptic excitation to the α-motoneurons that control the muscle in which the muscle spindle is located. Thus, lengthening a muscle will lead to excitation of its motoneurons, counteracting the lengthening. This stretch reflex thus involves negative feedback. This reflex arc also provides the basis of the muscle contraction elicited by a brief tendon tap, which is used as a clinical test to investigate whether responsiveness of the motoneuronal pool is normal (the tendon reflex) (Fig. 27.3).

Figure 27.3 Feedback from muscle receptors to motoneurons: the stretch reflex. In a muscle connected between two bones, the Golgi tendon organ is located at the transition between muscle and tendon. It senses any active tension produced by the muscle fibers being in series between muscle and tendon. The muscle spindle is located in parallel with the muscle fibers. It signals the muscle length and dynamic changes in muscle length. Both the muscle spindle and the Golgi tendon organ have fast-conducting afferent nerve fibers in the range of ~100 m/s. The muscle spindle activates α-motoneurons directly, which causes the muscle fibers to contract. The sensitivity of the muscle spindle can be actively regulated by γ-motoneurons, which are more slowly conducting. The muscle spindle provides negative feedback. If the muscle with its muscle spindle is lengthened, the afferent activity from the muscle spindle increases, exciting the α-motoneurons and leading to an increased muscle contraction, which in turn counteracts the lengthening. This is called the stretch reflex. The Golgi tendon organ provides force feedback. The more the muscle contracts, the more the Golgi tendon organ and its afferent are activated. In the diagram the intercalated interneuron between afferent nerve fiber and motoneuron is inhibitory. Thus increased muscle force leads to inhibition of the α-motoneuron, which results in a decrease of muscle force. The efficacy of length and force feedback can be regulated independently in the spinal cord and via γ-motoneurons. Thus their respective contributions can vary considerably between different patterns of motor behavior.

4. Sensory signals from a number of different receptor systems help detect and counteract any disturbance of body posture during standing or maintenance of another body position. Skin and muscle receptors and receptors signaling joint position, as well as vestibular receptors and vision, contribute to different aspects of the dynamic and static control of body position.

5. When an object is held by the fingers, skin receptors at the contact points at the fingertips play an important role. If the object tends to slip, the skin receptors become activated and signal rapidly to the nervous system that there is a need for extra muscle force.

6. A great variety of sensory signals provide information about the position of different parts of the body in relation to each other and to the external world. Such information is critical when initiating a voluntary movement. For example, to move a hand toward a given object, it is important to know the initial position of the arm and the hand in space. Depending on whether the hand is located to the left or the right of the object, different types of motor commands must be given to bring the hand to the target. Sensory information about the initial position of the hand (or any other body part) in relation to the rest of the body, as well as the target for the movement, is thus important for eliciting an effective neural command signal.

To summarize, the sensory contribution to motor control is very important in many different contexts. If sensory control is incapacitated, motor performance will in most cases be degraded. Without sensory information in different forms, movements can usually still be executed, but as a rule with much less perfection.

Feedback and Feed Forward Control

If a perturbation occurs when an object is held in the hand, or when standing in a moving bus, it will be detected by different sensory receptors. This sensory signal will be fed back to the nervous system and be used to counteract the perturbation rapidly. This represents feedback control that is a correction of the actual perturbation after it has occurred. A limiting factor for the efficacy of feedback control in biological systems is the delay involved. A sensory afferent signal must first be elicited in the receptors concerned. It then has to be conducted to the central nervous system and be processed there to determine the proper response. The correction signal must subsequently be sent back to the appropriate muscle(s) and make the muscle fibers build up the contractile force required. In large animals, including humans, the delays involved can be substantial, and during fast movement sequences there may be little or no time for feedback corrections. In less demanding, more static situations, such as standing, sensory feedback is of critical importance.

In many cases a perturbation is anticipated before it is initiated, and correction begins before it has actually occurred. This type of proactive control is often called feed forward control in contrast to feedback. Such anticipatory control mechanisms are often automatic and involuntary, and are part of an inherent control strategy. For example, when standing on two legs and planning to lift up one leg to stand on the other, the body position begins to shift over to the supporting leg before the other leg is lifted. The projection of the center of gravity will fall between the two feet initially and will have to shift over to a projection through the supporting limb. In most instances, movement commands are designed so that the corrections of body posture required for a stable movement occur before the particular movement has started, as when lifting a heavy object. If the converse situation occurred, and body position was corrected only after a perturbation had taken place, movements would become much less precise than with feed forward control.

Motor Programs Coordinate Basic Motor Patterns

Both vertebrates and invertebrates have preformed microcircuits—neuronal networks that contain the necessary information to coordinate a specific motor pattern such as swallowing, walking, or breathing. When a given neuronal network is activated, the particular behavior it controls will be expressed. A typical network consists of a group of interneurons that activate specific groups of motoneurons in a certain sequence and inhibit other motoneurons that may counteract the intended movement. Such a group of interneurons is often referred to as a central pattern generator (CPG) or motor program. This CPG can be activated by will, as at the start of walking, or be triggered by sensory stimuli, as in a protective reflex or swallowing. In most types of motor behavior, sensory feedback may also form an integral part of the motor control circuit and may determine the duration of the motor activity such as the inspiratory phase of breathing (Fig. 27.4).

Figure 27.4 Motor coordination through interneuronal networks: central pattern generators. The brainstem and spinal cord contain a number of networks that are designed to control different basic patterns of the motor repertoire, such as breathing, walking, chewing, or swallowing. These networks are often referred to as central pattern generator networks (CPGs). CPGs contain the necessary information to activate different motoneurons and muscles in the appropriate sequence. Some CPGs are active under resting conditions, such as that for breathing, but most are actively turned on from the brainstem or the frontal lobes. For instance, the CPG coordinating locomotor movements is turned on from specific areas in the brainstem, referred to as locomotor centers. These descending control signals not only turn on the locomotor CPG, but also determine the level of activity in the CPG and whether slow or fast locomotor activity will occur. In order to have the motor pattern well adjusted to external conditions and different perturbations, sensory feedback acts on the CPG and can modify the duration of different phases of the activity cycle, providing feedback onto motoneurons. Although vertebrates as well as invertebrates have CPGs for a great variety of motor functions, the intrinsic operation of these networks of interneurons constituting the CPG is unclear in most cases. In invertebrates, a few CPG networks have been studied in great detail, such as the stomatogastric system of the lobster and networks coordinating the activity of the heart and locomotion in the leech. In vertebrates, detailed information concerning locomotor networks is only available in lower forms, such as the frog embryo and the lamprey.

There are several different types of motor programs that are located along the neuraxis from the spinal cord to the forebrain. Together they constitute a set of standard but adaptable motor programs that can be referred to as a motor infrastructure (Fig. 27.5A).

Box 27.1 Central Pattern Generator Networks

To establish a network as a central pattern generator network (CPG), one needs to show that the motor pattern can be generated in the absence of sensory input. This is the case for all networks discussed in this chapter, although in most cases sensory information interacts with the CPG for phase transitions.

To understand the intrinsic function of a given network, one needs as a minimal requirement to identify which nerve cells are part of the network, the synaptic interactions within the network, and the membrane properties of each cell type. A network like the one controlling locomotion (Fig. B27.1) can be turned on by signals from the brainstem commands region. If a population of excitatory interneurons (E) is turned on, and they interact synaptically with one another to some degree, burst activity can be generated. Such groups of excitatory interneurons can be said to form a burst generating kernel. It is easy to understand that when a group of interneurons are activated, they will generate a train of action potentials, but what is the mechanism responsible for terminating the activity and thus for generating a burst? In the spinal network generating locomotion in the lamprey, the cellular properties of the neurons are of critical importance. During each burst calcium and sodium ions accumulate intracellularly due to both synaptic and spike activity that engages different types of ion channels (Fig. B27.1). This in turn leads to an activation of calcium-dependent K⁺ channels and also potentially sodium-dependent K⁺-channels. They act in a hyperpolarizing direction in each neuron and help terminate the activity. Since the neurons interact, there will also be less excitatory drive as neurons progressively drop out. In Figure B27.1, the excitatory kernel has been drawn with only two E excitatory neurons, but in reality they represent a population of perhaps 40 or 50 interneurons. These interneurons are glutamatergic and activate both NMDA and AMPA receptors. There are a number of complementary processes that make the neurons cease being active at the end of a burst, and conversely to recruit neurons at the onset of a burst.

Figure B27.1 The segmental locomotor network of the lamprey. Schematic representation of the different components of the core neural circuitry that generates rhythmic locomotor activity. All neuron symbols denote populations rather than single cells. The reticulospinal (RS) glutamatergic neurons excite all types of neurons at the segmental level. The excitatory (E) neurons excite all three groups of neurons indicated, and the inhibitory neurons inhibit contralateral neurons and motoneurons (M). The reticulospinal neurons in turn receive excitatory drive from the mesencephalic (MLR) and diencephalic (DLR) locomotor regions, which at rest are under tonic inhibitory control from the output cells of the basal ganglia. When locomotor activity is to be initiated, the inhibitory control from the basal ganglia is removed (see above), and the network is free to operate. To the right, cellular mechanisms are indicated that aid in burst termination. During each burst, Ca²⁺ and Na⁺ ions accumulate in the cell that in turn activate calcium and also sodium-dependent K⁺ channels that progressively hyperpolarize the cells, thereby terminating ongoing spike activity.

In a network generating alternating activity—for instance, on two sides of a segment—there are two burst generating kernels, one on each side, connected with inhibitory neurons (Fig. B27.1). Thus when one kernel of excitatory neurons is active, the contralateral kernel is inhibited, and when its burst is terminated, the other side takes over due to the background drive from the brainstem. The crossed inhibitory neurons thus are responsible for the coordination beween the right and left sides, and they use glycine as a transmitter.

A generalizing conclusion is that in a number of burst generating circuits for, for example, different forms of locomotion, respiration, and saccadic eye movements, burst generating kernels of excitatory interneurons play a key role. In all cases the synaptic interaction is critical, and the cellular properties contribute significantly.

Figure 27.5 Location of different networks (CPGs) that coordinate different motor patterns in vertebrates. (A) Represents an overview of the location of the different CPGs in vertebrates. The spinal cord contains CPGs for locomotion (B) and protective reflexes, whereas the brainstem contains CPGs for breathing (C), chewing, swallowing, and a motor map for saccadic eye movements (D). The hypothalamus in the forebrain contains centers that regulate eating and drinking. These areas can coordinate the sequence of activation of different CPGs. For instance, if the fluid intake area is activated, the animal starts looking around for water, walks toward the water, positions itself to be able to drink, and finally starts drinking. The animal will continue to drink as long as the stimulation of the hypothalamic area is sustained. This is an example of recruitment of different CPGs in a behaviorally relevant order. The expression of emotion is represented by seven motor programs as illustrated for anger (E) by a child sculptured by Vigeland. The cerebral cortex is important, particularly for fine motor coordination involving hands and fingers and for speech.

The intrinsic function of the neuronal networks that comprise the CPGs have in most cases been difficult to deduce due to the complexity of the neuronal interactions. Box 27.1 provides an example from the locomotor CPG of the lamprey, in which the brainstem-spinal cord networks are understood.

Other types of eye movement control are used when tracking an object that is moving, as when watching a game of tennis. These tracking movements are represented in the cerebral cortex and the brainstem.

Posture

Animals, including humans, have the ability to position the body in a variety of postures. The most stable position is lying horizontally in bed, which requires little activity from the nervous system. The situation is very different when standing and even more so when supporting oneself on one leg. This requires a very structured command to handle the activation of different muscles in the legs, trunk, shoulders, and neck. The nervous system issues this complex motor command to achieve a certain position. The ability to maintain a stable position standing on one leg is then critically dependent on feedback from a variety of sensory receptors that detect falling to one side or the other. These receptors activate reflex circuits that try to counteract the impending loss of balance. Of great importance are the vestibular receptors located in the head, which sense head movements in the different directions. Vestibular reflexes act on the neck muscles to correct the position of the head, and vestibulospinal reflexes act on the trunk and the extensor muscles of the limbs. In addition, vestibular effects are also mediated indirectly by the reticulospinal system. Muscle receptors in neck and limb muscles are of equal importance. They also play a critical role in detecting and compensating for postural perturbations, in particular around the ankle, where skin receptors on the bottoms of the feet are activated. Finally, visual feedback may contribute as when one is standing on a moving surface such as the deck of a boat.

One can decide at will what posture to assume, whether standing on one leg or sitting in an armchair. In that sense, assuming different postures represents a set of voluntary motor acts organized at both the brainstem and forebrain level. At the same time, any given posture is stabilized in an “automatic” way by the sensory mechanisms described above.

Roles of Different Parts of the Nervous System in the Control of Movement

The different basic motor patterns are present in different forms in most vertebrates and invertebrates. In the former, the underlying neural networks are located either in the brainstem or in the spinal cord, and in the latter, in the chain of ganglia. In vertebrates, from fish to mammals the basic organization of the nervous system is similar with regard to the spinal cord, brainstem (medulla oblongata and mesencephalon), and diencephalon. Each species and larger group of vertebrates may have specializations related to particular types of functions, and complexity increases during evolution. The cerebral cortex, with its distinct lamination, is present in mammals and most highly developed in primates, including humans (Fig. 27.5).

Basic Motor Programs Are Located at the Brainstem–Spinal Cord Level

The brainstem–spinal cord is, to a large extent, responsible for the coordination of different basic motor patterns. The spinal cord contains the motor programs (CPGs) for protective reflexes and locomotion, whereas those for swallowing, chewing, breathing, and fast saccadic eye movements are located in the brainstem (mesencephalon and medulla oblongata). In most cases, however, both the brainstem and the spinal cord are involved to some degree. In mammals as well as lower vertebrates, the brainstem–spinal cord, isolated from the forebrain (di- and telencephalon), is able to produce breathing movements and swallowing, as well as walking and standing. These brainstem–spinal cord animals (referred to as decerebrate models) can thus be made to walk, trot, and gallop with respiration adapted to the intensity of the movements and to swallow when food is put in their mouth. However, they perform these maneuvers in a stereotyped fashion, such as a robot or a reflex machine. The movements are thus not goal directed nor adapted to the surrounding environment, but are nevertheless coordinated in an appropriate way.

Diencephalon and Subcortical Areas of Telencephalon Are Involved in Goal-Directed Behavior

Mammals that are devoid of the cerebral cortex but have the remaining parts of the nervous system intact have been used to investigate the importance of these areas. Such animals display a more advanced behavioral repertoire when compared to the brainstem animals described earlier. At first glance, these decorticate animals look surprisingly normal. They move around spontaneously in a big room and can avoid obstacles to some extent. They eat and drink spontaneously and can learn where to obtain food and search for food when supposedly hungry. They may also display emotions such as rage and attack other animals; however, they appear unable to interact in a normal way with other individuals of the same species. Even without the cerebral cortex, a surprisingly large part of the normal motor repertoire can be performed, including some aspects of goal-directed behavior. Thus the cerebral cortex is not required to achieve this level of complex motor behavior (however, see later discussion). The diencephalon and subcortical areas of the telencephalon of the forebrain contain two major structures that are important in this context: the hypothalamus and the basal ganglia.

Hypothalamus

The hypothalamus is composed of a number of nuclei that control different autonomic functions, including temperature regulation and intake of fluid and food. The latter nuclei become activated when the osmolality is increased (fluid is needed) or the glucose levels become low (food is required). Stimulation of the paraventricular nucleus involved in the control of fluid intake results in a sequence of motor acts involving a number of different motor programs. Continuous activation of this nucleus by electrical stimulation or local ejection of a hyperosmolalic physiological solution leads to an alerting reaction. The animal first starts looking for water (perhaps experiencing a feeling of thirst). It then starts walking toward the water, positions itself at the water basin, bends forward, and starts drinking. The animal will continue drinking as long as the nucleus is stimulated. These hypothalamic structures are thus able to recruit a sequence of motor acts that appear in a logical order. Other parts of this region can elicit rage and attack behavior, as was first described by the Nobel laureate Walter Hess. Humans have seven distinct motor programs to express different types of emotions from crying (already present directly after birth) to smiling, laughing and the display of anger. These motor programs are inborn but what you laugh at is of course culturally bound.

Basal Ganglia

Basal ganglia constitute a second major set of nuclei that is conserved from lamprey to man. They are of critical importance for the normal initiation of motor behavior. They are subdivided into an input region (the dorsal and ventral striatum) activated by the cerebral cortex, thalamus, and an output region referred to as pallidum (globus pallidus interna and substantia nigra reticulata). The output neurons are inhibitory and have a very high level of activity at rest. The purpose is to keep the different brainstem motor centers under tonic inhibition so that they are not activated under resting conditions and also thalamic neurons projecting back to the motor areas of the cerebral cortex. The input stage of the striatum determines the level of activity in the different pallidal output neurons. When a motor pattern, such as a saccadic eye movement, is going to be initiated, the pallidal output neurons that are involved in eye motor control become inhibited by the striatum. This means that the tonic inhibition produced by these neurons at rest is removed, and the saccadic motor centers in mesencephalon (superior colliculus) are relieved from tonic inhibition and become free to operate and induce an eye saccade to a new visual target.

In contrast to the tonically active output neurons of the basal ganglia (pallidum), the striatal neurons are silent at rest and have membrane properties that make it difficult to activate them from cortex or thalamus, unless the dopamine system is active. The level of activity in dopaminergic neurons in the substantia nigra and associated nuclei determines the responsiveness of striatal neurons. If the level of dopamine activity is reduced, as in Parkinson’s disease, it becomes very difficult to activate the striatal neurons and thereby to initiate and carry out most types of movements. This results in a severe handicap characterized by paucity of movement (hypokinesia). The converse condition with enhanced levels of dopamine in the basal ganglia (which can occur as a side effect of medication) results instead in a richness of movements, and even in unintended movements (hyperkinesias). These movements may be well coordinated, but without a purpose, and occur without the conscious involvement of the patient. Dopamine neurons are thus of critical importance for the operation of the striatum and they regulate the responsiveness of the striatal circuitry to input from the cerebral cortex and thalamus. Striatum in itself clearly has a key role for the operation of the entire motor system.

The Cerebral Cortex and Descending Motor Control

In the frontal lobe there are several major regions that are involved directly or indirectly in the execution of different complex motor tasks like Broca’s area for speech, or the skilled movements used to control hands and fingers when writing, drawing, or playing an instrument. Some neurons in the frontal lobe are activated not only when we perform a given movement, but also when we see somebody perform the same type of movement. They are therefore called mirror-neurons and may underlie the ability to imitate. Many of these regions are organized in a somatotopic fashion. In the largest area, referred to as the primary motor cortex or M1 in the precentral gyrus, the legs and feet are represented most medially and the trunk, arm, neck, and head are represented progressively more laterally. Areas taken up by the hands and the oral cavity are very large in humans, and are much larger than that for the trunk. This is explained by the fact that speech and hand motor control require a greater precision and thus a larger cortical processing area than the trunk. The latter is important for postural control but is less involved in the type of skilled movement controlled by the motor cortex (Fig. 27.6).

Figure 27.6 Organization of the primary motor cortex (M1). The different parts of the body are represented in a somatotopic fashion in M1, with the legs represented most medially, arms and hand more laterally, and the oral cavity and face even more so. Note that the two areas that are represented with disproportionally large regions are the hand with fingers and the oral cavity. They are represented in this way due to the fine control required in speech and fine manipulation of objects with the fingers. Adapted from Penfield and Rasmussen (1950).

The large pyramidal cells in the motor cortex send their axons to the contralateral side of the spinal cord and are able to activate their target motoneurons directly, but a number of interneurons in the brainstem or spinal cord are also influenced. These long projection neurons are called corticospinal neurons. Corticospinal neurons of the arm region in M1 thus project to arm motoneurons in the spinal cord, as well as interneurons involved in the control of arm, hand, and fingers. In addition to M1, other areas in the frontal lobe, such as the supplementary motor area and prefrontal areas, are involved in other aspects of motor coordination (Fig. 27.7).

Figure 27.7 Descending pathways projecting from the brain that mediate motor actions to the spinal cord. From the cerebral cortex, including M1, a number of neurons project directly to the spinal cord to both motoneurons and interneurons. In addition, they also project to different motor centers in the brainstem, which in turn sends direct projections to the spinal cord. One of these centers is the red (rubrospinal) nucleus, which influences the spinal cord via the rubrospinal pathway. Corticospinal and rubrospinal pathways act on the contralateral side of the spinal cord and have somewhat overlapping functions. They are sometimes referred to as the lateral system, as most of the fibers descend in the lateral funiculus of the spinal cord. Vestibulospinal and reticulospinal pathways are important both for regulating posture and for correcting perturbations, and they mediate commands initiating locomotion. The different reticulo- and vestibulospinal pathways are sometimes lumped together as the medial system, as most of the axons project in the medial funiculus of the spinal cord.

The cortical influence on movements is exerted through several different channels, via direct projections to the input stage of the basal ganglia (striatum), through effects exerted via brainstem circuits and as mentioned above also directly to the spinal cord via corticospinal projections. The brainstem contains a number of descending pathways (rubrospinal, vestibulospinal, and reticulospinal pathways) that can initiate movements and correct motor performance, while others provide more subtle modulation of the spinal circuitry. The former act rapidly in the millisecond time frame that is required in motor control. Examples of the latter are the slow-conducting and slow-acting noradrenergic and serotonergic pathways. These pathways set the responsiveness of different types of neurons, synapses, and spinal circuits. The cortical control of motor coordination is, in addition to the effects via the basal ganglia, achieved through both direct action on spinal and brainstem motor centers but also to a significant degree by parallel action on a variety of brainstem nuclei. The execution of movements that are initiated from the cortex is, to a large degree, a collaborative effort of many parts of the nervous system. The reticulospinal and vestibulospinal pathways mediate the control of posture, and the former take part in the initiation of locomotion, via the CPG located in the spinal cord (Fig. 27.8).

Figure 27.8 Summarizing scheme of the interaction between different motor centers. The different major compartments of the motor system and their main pathways for interaction are indicated. The basic functions of the different compartments and descending tracks are summarized on and below the scheme. CS, corticospinal; RbS, rubrospinal; VS, Vestibulospinal; RS, reticulospinal.

Judging from experiments on mammals and primates and on patients that have suffered focal lesions of the frontal lobe, the cortical control of movement is of particular importance for dexterous and flexible motor coordination, such as the fine manipulatory skills of fingers and hands and also for speech. It is very difficult for a casual observer to see the difference between a normal monkey moving around in a natural habitat and a monkey that is lacking the corticospinal direct projections to the spinal cord but has all other cortical circuits intact. They are able to move around, climb, and locomote as normal monkeys. It is only when special tests are performed that one can see that delicate independent movements of the individual fingers are incapacitated. This is a type of motor coordination that is needed in monkeys and other primates for skilled manipulations of the environment, such as picking fine food objects from small holes.

The most flexible types of motor coordination, such as the skilled movements of the fingers and hands or in speech, involve the frontal lobes. These types of movements are often referred to as voluntary because they are performed at will. This terminology is commonly used, although inappropriate in that more basic types of motor coordination, such as the ones used when walking, chewing, or positioning ourselves in a relaxed posture, are also controlled by will, although motor coordination is handled to a larger degree by motor programs located at the brainstem–spinal cord level. Thus it appears important not to draw a clear-cut dividing line between the different types of movements, at least with respect to the voluntary aspect of their control.

It is also useful to realize that, to some degree, simple reflexes can also be modulated by will. For instance, the simple skin avoidance reflex (flexion reflex) that causes rapid withdrawal of a finger after touching something painful is subject to such control. Knowledge that the object will be sufficiently hot to be somewhat painful enables this modulation if it is important. If, however, the object is touched without knowing that it is hot, the hand will be withdrawn with the shortest possible latency. Thus, in the entire motor apparatus from the simplest reflex to a skilled movement, there is the possibility for modulation and flexibility. The human nervous system allows a remarkable combination of movements, adapted to different situations. This flexibility is perhaps greater for primates and humans than for any other species. However, other species may perform much better in particular types of specialized movements. A cheetah runs faster, a hawk catches prey at higher speeds, and a fly walks upside down on the ceiling.

Visuomotor Coordination—Reaction Time

Visuomotor coordination is amazing. Consider an approaching object like a ball that you would like to catch. This requires a rapid calculation of the path of the ball and being able to extend the arms and position the hands in the appropriate spot at just the right time. The motor command needs to consider not only the timing but also the visuomotor details about joint angles from the eyes to the hand via the neck and head. How is this achieved? The parietal lobe processes visual information that will contribute to the initiation of the motor commands.

In order to arrive at an accurate and precise motor command, a wealth of information needs to be processed about the dynamic and static conditions of the different parts of the body in relation to each other and to the surrounding world. The processing time has been investigated experimentally by having a subject respond to a signal by performing a movement, such as pressing a button when a light goes on. The delay is around 0.1–0.2 s and is called the simple reaction time. It represents a minimal delay for a given test situation that cannot be shortened by training (Fig. 27.9).

Figure 27.9 Reaction time tasks of different complexity and in different age groups. The choice reaction time is plotted versus the number of choices the subject is exposed to for six different age groups. A simple experimental situation with two choices takes an adult 0.3 s to initiate, but a 6-year-old needs around 1.0 s. A 10-year-old falls in between at around 0.7 s. An increase in the number of choices causes a marked increase in reaction time. From a practical standpoint, this provides important information when one considers the possibility for a child to cope with different demanding situations, such as moving around in a modern city. Data replotted from Conolly (1970).

The more complex the situation is, the longer the reaction time. If a choice is involved, such as responding by pressing different buttons to light stimuli of different colors, the time delays become much longer than in the simple reaction time task. The choice reaction task time increases in proportion to the level of complexity and the number of choices (Fig. 27.9).

When young children are tested, their reaction times are much longer than in adults. The simple reaction time for a 6-year-old can be three times that of a 14-year-old, and the difference with choice reaction times may be even larger. In everyday life this condition may have severe consequences, such as in an unexpected traffic situation. A 6-year-old will need at least three times more time to interpret what he or she sees. This is the basis for recommending that children below 11 years of age should not ride a bicycle in open traffic. It is noteworthy that this is a gradual maturation process, and training will not shorten the reaction time. During aging the choice reaction time tends to increase again, a factor that can be important when driving a car.

The cerebellum is also involved in the coordination of movement. Although it is smaller than the cerebral cortex in volume, it is thought to contain as many nerve cells as the latter. It has a very stereotyped neuronal organization with two types of inputs from mossy and from climbing fibers. These inputs not only carry information from the spinal cord about ongoing movements in all different parts of the body, but they also carry information from the different motor centers about planned movements even before a movement has been executed. The cerebellum also interacts with practically all parts of the cerebral cortex. This means that it is updated continuously about what goes on in all parts of the body with regard to movement and also about the movements that are planned in the immediate future.

Lesions of the cerebellum lead initially to great problems with postural stability and with a lack of accuracy of movements. Most movements can be carried out, but their quality is reduced drastically. The cerebellum was originally thought to be involved exclusively in the coordination of movement. In addition to motor control, evidence has accumulated that the lateral parts of cerebellum may also be involved in different cognitive tasks.

The cerebellum is subdivided into a great number of different microregions that each process different kinds of information. These different regions respond via their output neurons, called Purkinje cells. The cerebellar cortex contains only inhibitory interneurons, and the two input systems, climbing fibers and mossy fibers (via granular cells), provide the excitatory components. All Purkinje cells are inhibitory and project to the cerebellar nuclei, which in turn are excitatory and project to different motor centers in the brainstem and to the cortex via the thalamus.

Each granular cell receives very specific input from only a few mossy fibers, which are able to make them fire. The granular cell axons form parallel fibers in the cerebellar cortex, and thousands of parallel fibers impinge on each Purkinje cell that has a very extensive dendritic tree. It is believed that only a small portion of these parallel synapses are functional in any given moment but that they are available to become recruited in learning situations. In contrast, there is only one climbing fiber for each Purkinje cell. The climbing fibers can serve as “error detectors,” at least under some conditions. Through the effects of climbing fibers on the Purkinje cells, the synaptic efficacy of the parallel fiber input to the Purkinje cell can be modified (depressed). This change in efficacy can last over many hours and possibly much longer and is referred to as long-term depression (LTD). It is generally thought that this LTD can contribute to motor learning (see later). With regard to basic movements such as posture, walking, and the much-studied eye blink reflex, the cerebellum is clearly involved through each movement phase. It is likely that the fine tuning and modification of movements that are required in different behavioral contexts involve the cerebellum. With regard to the eye blink reflex, motor learning has been demonstrated in terms of associating an unconditioned stimulus with a conditioned one. It requires that the cerebellar cortex be intact, which strongly suggests that the cerebellar cortex contributes importantly to this type of motor learning.

There is still, however, much to learn about the actual processing that goes on in the cerebellum, a structure that regulates the quality of motor performance and is involved in some forms of motor learning. A variety of complex motor tasks, including speech and handwriting, can still be performed, but with less accuracy. Thus, the motor programs for these learned tasks cannot be stored exclusively in the cerebellum. When new types of movements are learned, such as riding a bicycle, playing an instrument, or typing at a keyboard, the new stored programs for sequences of motor acts must also involve other structures in the brain, most likely the basal ganglia and the cerebral cortex.

Motor Learning

All vertebrate and invertebrate species with a nervous system have a motor infrastructure that is determined evolutionarily. In the case of primates, including humans, there are preformed circuits (e.g., CPGs) that allow performance of a basic movement repertoire (e.g., locomotion, posture, breathing, eye movements), as well as basic circuits that underlie reaching, hand and finger movements, and sound production, as in speech. This constitutes the motor infrastructure that is available to a given individual after maturation of the nervous system has occurred.

The motor system is in addition characterized by its ability to learn, in what is referred to as motor or procedural learning. The cortex and basal ganglia play an important role for learning of new motor sequences and skills, and the cerebellum is indispensable for the quality of the different motor patterns and the ability to associate two stimuli, as during conditioned reflexes. When one is able to perform a movement with great precision, like hitting a ball, this may lead to an activation of the dopamine system and a feeling of satisfaction (reward). It has a marked ability to facilitate the development of synaptic plasticity, which is learning. It facilitates the particular synapses that were active as the successful movement was performed.

One can also learn motor sequences, as when playing the flute. The particular finger settings that produce a given tone can be retained in memory, along with the sequence of tones that produce a certain melody. Thus particular muscle combinations that produce a sequence of well-timed motor patterns are stored. Similarly, when learning to pronounce different sounds (phonemes) and combine them into words, the muscle activation patterns are stored. A simpler case occurs when learning to recombine basic movement patterns to be able to maintain equilibrium while bicycling. There are a large number of examples involving different types of motor learning and parts of the body.

It is characteristic of a given learned motor program that one can “call” upon it to perform a given motor act over and over again. Despite this, there is generally little awareness of the way in which the movement is actually performed in terms of the muscles involved and their timing. Information has thus been stored in the nervous system with regard to which parts of the infrastructure of the motor system should be used to produce a given learned motor pattern, as well as the exact timing between the different components. However, the details of the complex storage process are not yet fully understood.

Conclusion

After this overview of the motor control system, it should be apparent that biological evolution has succeeded in refining remarkable sensorimotor machinery capable of executing the motor tasks required for the full behavior of a given species. The degree of difficulty involved is apparent when watching the clumsiness and lack of flexibility in robots, characteristic of even the most advanced models that have been developed. Details of the different neural subsystems involved in motor control are dealt with in the following chapters.

Chapter 27

Fundamentals of Motor Systems