Course Handout - The Pyramidal and Extrapyramidal Motor Systems
Copyright Notice: This material was written and published in Wales by Derek J. Smith (Chartered Engineer). It forms part of a multifile e-learning resource, and subject only to acknowledging Derek J. Smith's rights under international copyright law to be identified as author may be freely downloaded and printed off in single complete copies solely for the purposes of private study and/or review. Commercial exploitation rights are reserved. The remote hyperlinks have been selected for the academic appropriacy of their contents; they were free of offensive and litigious content when selected, and will be periodically checked to have remained so. Copyright © 2002-2018, Derek J. Smith.
First published online 11:35 BST 23rd August 2002, Copyright Derek J. Smith (Chartered Engineer). This version [2.0 - copyright] 09:00 BST 4th July 2018.
An earlier version of this material was contained in Smith (1997b). It is repeated here as a succession of extracts, supported with hyperlinks.
The Motor Cortex
Before proceeding, it may help to have to hand a copy of Kleist's (1934) cortical map. Click here to be transferred.
The motor cortex of each cerebral hemisphere controls voluntary skilled movement of the contralateral skeletal muscles. There has, however, been some controversy over how these areas are organised, and the following descriptions are from a variety of sources, including Crosby, Humphrey, and Lauer (1962), Wise (1985), and Goldberg (1985):
Primary Motor Area (M1 Cortex): M1 cortex maps reasonably well onto Brodmann's Area 4, and therefore occupies most of the precentral gyrus. When this M1 cortex is electrically stimulated, a movement takes place in the corresponding part of the body. This will always be a contralateral "natural" movement. Remembering that muscles work in opposed pairs (flexor-extensor), it is important when contracting one of these to relax its antagonist. Centrally elicited movements are complete in this respect, which means that the necessary coordination must take place subcortically. M1 cortex is Type 1 agranular cortex. The external band of Baillarger - the horizontal connections in Layer IV - is markedly underdeveloped in this area. Damage to the primary motor area results in flaccid (soft and relaxed) paresis (reduced power and fine control) of the voluntary musculature.
Supplementary Motor Area (SMA) (M2 Cortex): M2 cortex maps onto the medial and dorsal aspects of Brodmann's Area 6, thus placing it rostral to M1. However, it lacks the giant cells of Betz which characterise M1 cortex. M2 stimulation elicits more complicated bilateral movements such as walking. Damage to the SMA results in spasticity (rigidity and spasm) without paresis.
Premotor Cortex (PM Cortex): According to Wise (1985), this term was introduced by Hines (1929), and may safely be taken to be the "laterally situated part of the agranular frontal cortex, the region outside of the boundaries of both the primary motor cortex (M1) and the supplementary motor cortex (M2) ....." (Wise, 1985:3). This maps it onto the more lateral aspects of Brodmann's Area 6. Unlike M1 and M2, PM does not have a clear somatotopic representation. Indeed, it seems to be involved in the motor programming process (for more on which see Smith, 1997). It also has a higher density of Layer V pyramidal cells and a projection to the red nucleus. Broca's Area (see later this chapter) is a premotor type area.
The spinal tracts constitute the white matter of the spinal cord, and, as noted previously, occupy the vertical flutings in the grey matter. As a general rule, it is possible to locate a tract fairly precisely on the spinal cross-section by examining how its name is made up. With the lateral spinothalamic tract, for example, the key elements are lateral, spino-, and -thalamic. These elements combine to tell you that this tract is located laterally within the cross-section, comes from somewhere in the spinal cord, and runs to the thalamus. Moreover, because you already know that the thalamus is rostral to the spinal cord, you can then judge that this tract transmits upwards, and must accordingly be a sensory tract of some sort.
Because there are two major horns on the wings of the spinal grey matter , the spinal tracts fall naturally into quite distinct columns (known as funiculi). These are:
(a) Dorsal Column: The dorsal column lies in the groove between the left and right dorsal horns. It is comprised almost entirely of two major ascending fibre tracts, namely the fasciculus gracilis and the fasciculus cuneatus. These are sensory fibre tracts. They run all the way upwards to the brainstem, and thereby into the thalamus. Here, they synapse within the specific projection nuclei of the thalamus, before projecting to the sensory homonculus of the postcentral gyri of the cerebral cortex.
(b) Lateral Columns: The lateral columns (one left and one right) lie in the grooves between the dorsal and ventral horns. They contain the main descending tracts of the pyramidal and extrapyramidal motor systems (the lateral corticospinal tract for the former, and the reticulospinal and rubrospinal tracts for the latter), together with tracts ascending to the cerebellum.
(c) Ventral Column: The ventral column lies in the groove between the left and right ventral horns. It contains a variety of minor ascending and descending tracts, including the anterior corticospinal tract, a derivative of the pyramidal tract (see below).
There are also some shorter fibre tracts linking one vertebral level to its neighbours above and below (thus helping to integrate the activity of related myotomes), and some commissural fibres (helping to integrate the activity of the two halves of the body). Note, also, that the extrapyramidal tract is not a single discrete tract at all. It is more properly described as a system because it involves many lesser discrete tracts, such as the rubrospinal tract, a fibre tract descending from the red nucleus to spinal level and perhaps responsible, as part of the extrapyramidal system, for maintaining the tone of flexor muscles, and the vestibulospinal tract, which maintains the tone of extensor muscles.
This complex interweaving of ascending, descending, and local tracts, together with the amount of grey matter necessary to support them, means that both the diameter and the cross-sectional appearance of the spinal cord changes as one moves from segment to segment. The diameter is especially large where the nerves to the arms and legs emerge (the cervical and lumbar enlargements respectively). A typical spinal cord cross section is now shown .....
Spinal Cord Cross Section: A typical spinal cord section is shown (bottom centre). This consists of a vertically fluted column of grey matter (shaded), encased by a host of bilateral ascending and descending axon tracts (see Key). The cord is shown innervating a typical skeletal muscle. The alpha motor neuron is situated in the ventral (or "anterior") horn of the spinal grey, receives instructions from the pyramidal system, and innervates the muscle bulk. The neighbouring gamma motor neuron probably receives instructions from a number of extrapyramidal sources, including the vestibulospinal tract (as shown), and innervates muscle spindle fibres scattered relatively sparsely throughout the muscle bulk. Sensory fibres enter the spinal grey via its dorsal (or "posterior") horn. The main run of the spinal nerve (a bundling together of all the sensory and motor fibres which happen to be going the same way) is not shown. Set up in this way, the resulting "alpha-gamma system" will automatically maintain a given muscle tension despite external perturbations. Further detail in text. [See Smith (1997) for the cybernetic aspects of the alpha-gamma system, if interested.]
Key: FCun = Fasciculus Cuneatus; FGrac = Fasciculus Gracilis; LCS = Lateral Corticospinal Tract; LSThal = Lateral Spinothalamic Tract; MLF = Medial Longitudinal Fasciculus; MRS = Medullary Reticulospinal Tract; PRS = Pontine Reticulospinal Tract; RubS = Rubrospinal Tract; SCbl = Spinocerebellar Tract; SOT = Spino-Olivary Tract; STect = Spinotectal Tract; TectS = Tectospinal Tract; VestS = Vestibulospinal Tract; VSThal = Ventral Spinothalamic Tract
If this diagram fails to load automatically, it may be accessed separately at
Simplified from Carpenter (1991:98/108). This version Copyright © 2002, Derek J. Smith.
The spinal cord is involved in many of the body's reflexes. One way of classifying these reflexes is to count the number of neuron-to-neuron synapses involved. Monosynaptic reflexes have a single synapse between the sensory neuron and the motor neuron. Polysynaptic reflexes, on the other hand, involve one or more additional neurons (called interneurons) in the reflex pathway. The situation is further complicated by the fact that reflexes need to operate at more than one spinal segment. Those which enter and leave at the same level are known as intrasegmental reflexes, whereas those which enter and leave at different (higher or lower) levels are known as intersegmental reflexes.
One in particular of these - the stretch reflex - plays an especially powerful role in maintaining posture in the face of the unexpected. In essence, what happens is that an accidental or unwanted lengthening of a muscle will result in it immediately contracting more forcefully until the lengthening stops. The lengthening is detected by receptor cells within the muscle bulk. These are known as muscle spindles, and they form part of the kinaesthetic system. They send information back to the spinal segment, and raise the output of the lower motor neuron concerned. This makes for a very useful antigravity device because the system, once programmed, will automatically maintain itself in the same position for as long as it has the energy to do so. According to Bowsher (1970), all the "jerk" reflexes are monosynaptic, that is to say, there is a direct axon link between the proprioceptive and lower motor neurons concerned. Another useful example is the crossed extensor reflex. This involves the extending of limb A as a result of flexing the contralateral limb B. The spinal segment is also involved in the phenomenon known as alpha-gamma muscle control, a complex biological servomechanism for the fine control of skeletal muscles.
Pyramidal Motor System
The body's motor systems are traditionally dealt with under two headings according to whether the main descending pathway passes through that area of the ventral medulla known as the pyramids. Those which do are known as the pyramidal tract, and those which do not are known as the extrapyramidal tract. We now discuss each system in turn (although considerably more effort needs to go into the latter).
The pyramidal tract is a descending fibre tract, a large proportion of which is made up of corticospinal fibres arising from the giant cells of Betz, that is to say, the pyramidal cells of Layer V of the primary motor area (Area 4). These are the upper motor neurons, and their axons travel without interruption down through the corona radiata to the internal capsule, then down through the cerebral peduncles of the midbrain to the medulla. According to Carpenter (1991), the pyramidal tract consists of a million or so fibres at this point, 90% of which decussate in the medulla and travel down into the contralateral (or "crossed") lateral corticospinal tract of the spinal cord. The remaining 10% of the fibres remain ipsilateral (or "uncrossed"). 2% travel down the ipsilateral lateral corticospinal tract, and 8 % down the ipsilateral anterior corticospinal tract.
As far as the main (crossed) pathway is concerned, the fibres descend to the appropriate spinal segment where they synapse either directly with the appropriate lower motor neuron at the appropriate spinal level, or else with a local interneuron connected to that lower motor neuron. Bowsher (1970) states that the synapsing is direct for the lower motor neurons controlling the fingers, but indirect otherwise. According to Moyer (1980), approximately 55% of the pyramidal fibres terminate in the cervical spine, 20% in the thoracic spine, and 25% below that. As he puts it, this allows "much better innervation and consequently better motor control in the upper body and extremities" (p166). As far as the minor branches are concerned, the anterior tract mainly terminates in the cervical spine, where it decussates "late", that is to say, the fibres cross over within the destination spinal segment rather than in the brainstem. The uncrossed lateral tract does not decussate in either location.
Key Concept - Interneuron: An interneuron is an intermediate neuron in a neural circuit. It is excited or inhibited by an input neuron, and in turn excites or inhibits an output neuron. Depending upon the particular circuit involved, this allows various things to happen, including (a) an excitatory activity can excite an inhibitory interneuron, thus allowing it to suppress a competing excitatory activity (or, indeed, its own activity - see Renshaw inhibition below), and (b) an excitatory activity can excite an excitatory interneuron which feeds back onto itself thus keeping it active (usually termed "reverberating") long after the initial stimulation has died away.
Discharge of the lower motor neuron is provided with an automatic safety device by a phenomenon known as Renshaw inhibition. This involves excitation from the motor neuron being fed back via a collateral fibre onto an inhibitory interneuron known as a Renshaw cell, and thence back onto the body of the motor neuron itself. The more the lower motor neuron fires, the more the Renshaw cell intervenes to "cool it down". This makes what is called a "negative feedback" circuit, and presumably protects the body from accidental overactivity of the motor neuron concerned, and therefore damage to the muscle it is connected to.
The pyramidal system is consistently described as mediating voluntary movement, a suggestion which is supported by comparative studies showing that the tract is phylogenetically relatively new, being found only in mammals. Animals which have no "free will" have no pyramidal tract (or perhaps animals which have no pyramidal tract have no free will). The system also involves some widely distributed structures. Moyer (1980) reminds us that the human pyramidal tract is some 60cm long, whilst that of the whale can be up to 10 metres. It is thus highly vulnerable to damage, and the typical result of this damage is a contralateral flaccid hemiplegia. The flaccidity results from the fact that the pyramidal tract is essentially an excitatory system. It exists to tension muscles, and when it is damaged that tension simply disappears and the muscle eventually wastes away. This contrasts strongly with the result of lesions to the extrapyramidal tract (see next). The pyramids themselves can be damaged by a brainstem CVA involving the anterior spinal artery. If the resulting lesion is unilateral, the patient presents with the usual contralateral flaccid hemiplegia, commonly combined (because the hypoglossal nerve nucleus CNN.XII is in the same general area) with contralateral flaccid paralysis of the tongue. If the lesion is bilateral, then the paralyses, too, are bilateral.
Extrapyramidal Motor System
The extrapyramidal motor system is a highly complex system of major ganglia and tracts responsible for the control of involuntary muscle excitation. We now look at this system in greater detail, and begin by considering why it is needed in the first place. And the answer is that it solves some very real biomechanical problems for us. The main culprits are:
Problem No 1 - Gravity: This is the problem of maintaining posture and balance against the forces of gravity. The body solves this problem by maintaining most of its skeletal muscles in a state of resting tension. That is to say, your muscles are never totally relaxed: they are helping to keep your body rigid against the pull of gravity even when it is stationary. The resting tension is known as tonus (see panel), and maintaining tonus requires a lot of complex and continuous neural computation.
Key Concept - Muscle Tone: Muscle tone (or tonus) is the resting tension within a muscle. This can be anything between zero and full power, but is usually carefully maintained at a value in between these two extremes. Tone is maintained at a "background" level even when the organism is at rest. Additionally, the tone of antagonistic muscle pairs - especially those involved in maintaining posture and balance - is carefully maintained to balance each other out. This gives the body the stiffness it needs to resist gravity even when otherwise at rest. Tone is controlled by the relative number of motor units being activated, and can momentarily be increased considerably for protective purposes should a heavy impact be foreseen. Much spinal cord traffic is concerned (subconsciously and automatically) with the afferent and efferent aspects of postural control.
Problem No 2 - Prior Learning: This is the problem of motor skill. The point is that all but the simplest organisms show improvement with practice. Indeed, when it comes to motor behaviour they practice until they are perfect. This reaches its peak in skilled behaviours such as riding bicycles, playing games, juggling, and even playing the piano. What characterises this sort of behaviour is that through practice the brain develops a large repertoire of "pre-planned" movements which can be "run off" more or less involuntarily at high speed. This, too, requires a lot of complex and continuous computation, to say nothing of a memory store capable of looking after the skills in that repertoire while they are not actively being used.
Problem No 3 - Compensating Movements: This is the closely associated problem of what to do when the NS somehow "knows" during the execution of either a skilled or an unskilled movement that it is about to become unbalanced. Marsden, Merton, and Morton (1978) arranged for kneeling subjects to flex the elbow of one arm at a constant rate against a load attached to the wrist. The subjects steadied themselves by holding onto a table top with their free arm. Electromyograph readings were taken from the muscles of the supporting arm when the load on the moving arm was unexpectedly altered. They found that the muscles of the supporting arm began to compensate for the resulting imbalance within 55 msec. Such rapid adjustments to the tension of muscles far removed from the site of the disturbance are known as anticipatory postural reflexes. Traub, Rothwell, and Marsden (1980) describe such reflexes in Parkinson's disease, noting carefully where these were inferior to normals. The disturbance they used was a tug on the arm of the standing subject, and the normal anticipatory reflex was a rapid automatic contraction of the calf muscles. However, this reflex was "absent or greatly reduced in many patients with Parkinson's disease" (op cit:411).
Problem 4 - Merging the Voluntary with the Involuntary: This is the problem of suddenly imposing a voluntary excitation upon the background of ongoing involuntary excitation. It is the problem of one process interrupting or overriding another. It is not known how the switching backwards and forwards is managed biologically, but it has to be done or else your muscles would end up quite literally in knots.
And how is all this complex computation handled? Well the short answer is "with difficulty". Most movement takes a long time to get right - remember that it takes the human infant a year or so just to get upright, let alone ride a unicycle or walk a tightrope! This is probably because so many distinct NS subsystems need to be integrated. Suffice it here to note that by the time you have integrated the cerebral cortex, the proprioceptive senses, the two motor tracts, the basal ganglia, the cerebellum, the midbrain, and the spinal reflexes, you have involved a very sizeable percentage of the NS's available grey matter. That is to say, a very sizeable percentage of the brain's processing power is dedicated to the prosaics of getting about, most of which is carried out quite automatically and unconsciously. Do not underestimate the importance of locomotion and the motor systems, therefore. We now look at the role of the specific components one by one:
(a) Secondary Motor Area: The supplementary motor area (Area 6) assists the primary motor area by "pre-planning" sequences of movements. As such, it is closely involved in the learning and execution of skilled behaviours. It may also be involved in the initiation of voluntary movements, because lesions in this area are associated with initiation problems (Carpenter, 1991).
(b) Basal Ganglia: The basal ganglia have important but highly complex connections to the cerebral cortex, the diencephalon, the brainstem, and the spinal cord (see Chapter 8 of Passingham, 1993, for details if interested), and because of this complexity it is difficult to state precisely what each component nucleus actually does. Generally speaking, however, the basal ganglia seem to take care of subconscious bodily movements, integrating the necessary movements with the wished for. The putamen is involved in subconscious movements of skeletal muscles, and the globus pallidus in regulating muscle tone. Damage to the basal ganglia results in a variety of movement problems such as tremor, rigidity, and decomposition (see under cerebellum below), and is strongly implicated in Parkinson's disease and Huntington's disease. Cauterisation of the globus pallidus can alleviate the "flailing" movements found in some sufferers from Parkinson's disease. Another of the main symptoms of Parkinson's disease - bradykinesia - has also been studied. As described by Hallett and Khoshbin (1980), bradykinesia has several distinctive features including prolonged movement initiation time, prolonged time to arrest a false movement, weakness, and rapid fatigue. Above all, however, bradykinetic movement is slow. Hallett and Khoshbin saw this slowness as reflecting an inability of Parkinson's disease patients to execute ballistic movements, that is to say, movements whose end position has to be "computed" in advance, so that the limbs concerned can be "thrown" into the required position by a short and powerful pre-planned contraction of the agonist muscles concerned. And whilst EMG studies in normals show that the muscle contractions are finished well before the limb arrives in its required position, in Parkinson's disease patients there seems to be an inability to produce initial contractions of sufficient power. The authors conclude "that a normal role of the basal ganglia in movement is to energise the appropriate muscles required" (op cit:311).
(c) Thalamus: The thalamus functions as a highly elaborate "switching centre" for the NS, and its contribution to the extrapyramidal system reflects this. For more on its connections and role, see Crosson (1985).
(d) The Midbrain: Disease in the substantia nigra of the midbrain gives rise to Parkinson's disease, seemingly because there is a reduction in one of the important neurotransmitters there, namely dopamine. This is why injections of substances similar to dopamine - a monoamine neurotransmitter - markedly alleviate the symptoms.
(e) The Cerebellum: The cerebellum receives afferent input from skeletal muscles (tendons and joints), from the visual and auditory systems, and also from the vestibular system. It also receives efferent input from the motor cortex, the basal ganglia, and the brain stem. Its outputs descend to exercise a largely inhibitory influence on both the pyramidal and extrapyramidal tracts. The cerebellum, too, has a characteristic damage pattern, being characterised by a variety of motion problems but not paralysis proper. These problems include decreased muscle tone (hypotonia), ataxia, disturbed posture, intention tremor (zigzagging towards a target), rebound (overshooting a target when a resistance is suddenly removed), and dysarthria (disturbances of the smooth flow of movement) including dysarthria of speech. Inhoff, Diener, Rafal, and Ivry (1989) credit Holmes (1917) with having identified the cardinal clinical sign of cerebellar damage as a "decomposition of movement". This means that the once-coordinated activity of many sets of muscles can no longer be supported. To confirm this, they studied the ability of patients with cerebellar disease to carry out simple movement sequences. They concluded that the cerebellum helps "schedule" a previously programmed sequence of responses, and that it does this before the onset of the movement.
(f) The Vestibular System: The vestibular system signals information relating to bodily posture and balance to the CNS. The receptor cells are hair cells situated in the semicircular canals of the vestibular apparatus, the non-auditory part of the "labyrinth" of the inner ear. These are each sensitive to rotational acceleration in a given two-dimensional plane, and, since there are three of them at mutual right angles, the full set is sensitive to movement in all three dimensions. The vestibular system is thus a biological example of an inertial guidance system. Two smaller cavities, the utricle and the saccule, are similarly sensitive to linear acceleration. The hair cells communicate to bipolar first order neurons in the vestibular ganglion close by, and their signals are transmitted along the vestibular branch of the vestibulocochlear nerve to the (several) vestibular nuclei of the pons. Because of the need to input to so many other skeleto-muscular systems, the connections from the vestibular nuclei are complex. To begin with, there is a direct descending pathway known as the vestibulospinal tract, which can excite extensor and inhibit flexor motor neurons. Then there are tracts leading to and from the cerebellum, which gives that structure the information it needs to adjust any motor skills it happens to be controlling at the time. And finally, there are tracts leading to and from CNNs III, IV, VI, and XI, controlling the head, neck, and eye muscles. This allows vestibular response to intended movements of the head to be "ignored", as well as compensating movements of those structures to be initiated when movement elsewhere makes it necessary.
(g) Spinal Reflexes: See above.
(h) Motor Aspects of Sensory Systems: Unfortunately, the visual and auditory systems are complicated by the fact that they are not entirely sensory. There are, in fact, several motor systems involved in bringing them to peak efficiency. It is convenient to deal with these motor aspects under three headings as follows:
(i) Protective Reflexes: These are reflexes designed to protect the receptor cells in some way. The blink reflex is a good case in point, and is triggered automatically whenever the retinal cells detect a rapidly expanding retinal stimulus. The tympanic reflex is another. The tympanic muscles are small muscles anchoring the malleus and stapes to the skull. If a dangerously loud sound is detected, they contract very rapidly to stiffen these ossicles, thus preventing them from damage.
(ii) Orienting Reflexes: An orienting response is an instinctive attentional mechanism for turning a receptor apparatus in the direction of a sudden stimulus. Perhaps the easiest illustration is the orienting of the head (in man and other animals with fixed ears) or external ear (in animals - cats, dogs, etc - with directable ears) towards a sudden sound. What is significant is that after a few repetitions of the novel stimulus, it loses its novelty and ceases to elicit the orienting response. The mechanisms for this are believed to be located in the brainstem.
(iii) Fine Tuning: There are other motor systems designed to maximise the quality of a sensory input. The automatic focusing of the lens, and the automatic control of binocular convergence are cases in point. The mechanisms for these are located in peripheral ganglia and the brainstem.
The structures of the extrapyramidal system are all individually at risk from stroke, trauma, etc. Interpreting the symptoms is far harder than for the pyramidal system, however, reflecting the greater complexity and interconnection of the structures involved.
Bowsher, D. (1970). Introduction to the Anatomy and Physiology of the Nervous System (2nd Ed.). Oxford: Blackwell.
Carpenter, M.B. (1991). Core Text of Neuroanatomy. Baltimore, MD: Williams and Wilkins.
Crosby, E.C., Humphrey, T., and Lauer, E.W. (1962). Correlative Anatomy of the Nervous System. New York: MacMillan.
Goldberg, G. (1985). Supplementary motor area structure and function: Review and hypotheses. Behavioural and Brain Sciences, 8:567-616.
Hallett, M. and Khoshbin, S. (1980). A physiological mechanism of bradykinesia. Brain, 103:301-314.
Hines, M. (1929). On cerebral localisation. Physiological Review, 9:462-574.
Inhoff, A.W., Diener, H.C., Rafal, R.D., and Ivry, R. (1989). The role of cerebellar structures in the execution of serial movements. Brain, 112:565-581.
Marsden, C.D., Merton, P.A., and Morton, H.B. (1978). Anticipatory postural responses in the human subject. Journal of Physiology (London), 275:47-48.
Moyer, K.E. (1980). Neuroanatomy. New York: Harper and Row.
Passingham, R.E. (1993). The Frontal Lobes and Voluntary Action. Oxford: Oxford University Press.
Smith, D.J. (1997a). Human Information Processing. Cardiff: UWIC.
Smith, D.J. (1997b). Neuroanatomy for Students of Communication. Cardiff: UWIC.
Traub, M.M., Rothwell, J.C., and Marsden, C.D. (1980). Anticipatory postural reflexes in Parkinson's disease. Brain, 103:393-412.
Wise, S.P. (1985). The primate premotor cortex: Past, present, and preparatory. Annual Reviews of Neuroscience, 8:1-19.