The cerebellum is located at the rear of the brain above the brain stem. It is connected to the spinal cord, brain stem, and cerebral cortex. The function of the cerebellum has been the topic of classic studies of the effects of lesions of the cerebellum—such as discoordination and dysmetria—in animals and human patients. These symptoms suggest that the cerebellum helps us to move precisely, smoothly, quickly, and even without visual feedback. Because learning is essential to the acquisition of such skills, one can infer that the cerebellum learns by itself, although it is possible it connects to a learning mechanism elsewhere in the brain.
A detailed analysis of neuronal circuits in and around the cerebellum (see Figure 1) has yielded network models, modular models, and control-system models of the cerebellum to represent its computational capabilities. The cerebellar cortex is modeled by the three-layered simple perceptron, which, in combination with nuclear structures, forms a functional module of the cerebellum, a cerebellar corticonuclear microcomplex. The microcomplex provides an adaptive control mechanism to a spinal-cord/brain-stem function. In cortical sensorimotor functions, it provides an internal model that helps feedforward control. In mental thought processes in the association cortex as well, it seemingly plays a similar role.
Neuronal Network in the Cerebellar Cortex
The cerebellar cortex is made up of five principal types of neurons (Purkinje, basket, stellate, Golgi, and granule cells) that are arranged in a very uniform pattern. The cerebellar cortex receives mossy fibers from various precerebellar nuclei, which also supply excitatory synapses to cerebellar nuclei by way of their collaterals. Input from the mossy fibers is relayed to the granule cells, which send out bifurcating axons, called parallel fibers, that run in beams. The axons of the granule cells supply excitatory synapses to Purkinje cells, which provide the sole output of the cerebellar cortex. Each Purkinje cell receives as many as 175,000 inputs from the granule cells. Thus, the connections from mossy fibers to granule cells to Purkinje cells are the major signal-flow pathway, allowing the distribution of input information received by mossy fibers to numerous granule cells from which Purkinje cells select information to generate their output signals. Normally, Purkinje cells fire simple spikes at approximately 50 hertz. The discharge frequency of simple spikes are modulated in response to mossy-fiber input.
Each Purkinje cell also normally receives a single climbing fiber exclusively from the inferior olive. Climbing fibers project to the cerebellar cortex, forming strong synaptic connections with Purkinje cells, and send collaterals to the cerebellar nuclei. Climbing-fiber input occurs at a relatively low frequency (around 1 hertz) and over a narrow dynamic range but powerfully excites Purkinje cells, causing them to generate bursts of three to four high-frequency spikes (complex spikes).
A theory propounded in 1970 indicated that conjunctive activation of a mossy-fiber/parallel-fiber pathway and a climbing fiber onto a Purkinje cell either strengthens or weakens the transmission from the parallel fiber to the Purkinje-cell synapse. Climbing-fiber signals will thus modify the signal flow through the cerebellar cortex. This operation is analogous to that in which all connections from the second to the third layer are strengthened or weakened by an outside teacher who recognizes correct or incorrect performance of the simple perceptron. Indeed, long-term depression (weakening) of synaptic strength (LTD) was discovered a decade later. The occurrence of LTD suggests that the cerebellar cortex learns by means of weakening connections responsible for erroneous performance.
In the simple-perceptron model, the cerebellum processes spatial information; in the adaptive-filter model, it processes temporal patterns. In the adaptive-filter model, the inhibitory connection between Golgi cells and granule cells constructs an integrator with a long time constant, which activates granule cells with varied latencies. A mossy-fiber signal would then be converted to time-scattered parallel-fiber signals from which Purkinje cells select an appropriate temporal pattern learned through climbing-fiber signals.
Functional Module of the Cerebellum: Microcomplex
The cerebellar cortex is organized into seven longitudinal (A, B, C1, C2, C3, D1, and D2) zones. Each zone sends Purkinje cell axons to a certain cerebellar or vestibular nucleus. Thus, zone A projects to fastigial and vestibular nuclei, zone B to vestibular nuclei, zones C1 and C3 to the rostral part of the interpositus nucleus and zone C2 to the caudal part of it, and zones D1 and D2 to the medial and lateral parts of the lateral (dentate) nucleus.
Each longitudinal zone of the cerebellar cortex is composed of a number of microzones. Each microzone projects to a small group of vestibular or cerebellar nuclear neurons and receives climbing fibers from a small group of inferior olive neurons, which project collaterals to a small group of nuclear neurons projected by the microzone. A microcomplex is an interconnected set of a microzone, a small group of nuclear neurons, and a small group of inferior olive neurons. The microzones defined in the paravermis and flocculus may measure about 10 mm2. In rats, one microzone of this size contains about 10,000 Purkinje cells and 2,740,000 granule cells. The human cerebellum is about 50,000 mm2 wide, so that it may contain as many as 5,000 microzones as its functional unit.
The microcomplex would function as a module of the cerebellum in the following manner. First, input signals from a precerebellar nucleus (except those from the inferior olive) drive the nuclear neurons, which generate output signals of the microcomplex under inhibitory influences of Purkinje cells. Second, the same input signals pass via mossy fibers to a microzone, where they are relayed by granule cells and in turn excite Purkinje cells and other cortical neurons, eventually evoking simple spikes in Purkinje cells. Simple spike discharges of Purkinje cells driven by moss-fiber signals produce a unique functional state of a microzone arising from concerted activities of excitatory and inhibitory synapses. Third, climbing fibers convey error signals pertaining to the operation of the neural system that includes the micrcomplex, as generated by various neuronal mechanisms in diverse preolivary structures. Fourth, climbing-fiber error signals induce LTD in the conjointly activated parallel fiber-Purkinje cell synapses (learning rule) and thereby modify the operation of the micro-complex until the error signals are minimized. Climbing-fiber signals evoke complex spikes in Purkinje cells and thereby induce conducting impulses in Purkinje cell axons, which eventually evoke IPSPs in the nuclear neurons. However, the effects of the IPSPs on nuclear neurons are counteracted by the EPSPs evoked via collaterals of olivocerebellar fibers. The signal content of complex spikes has been analyzed in various motor behaviors and is related partly to consequence errors and partly to internally computed errors.
The postulated operation of the microcomplex, including inhibitory neurons in the cerebellar cortex, has been computer-simulated. However, since several forms of synaptic plasticity other than conjunctive LTD were observed in the cerebellum, further studies are needed to reproduce the performance of a micro-complex that incorporate these forms of synaptic plasticity.
Roles in Neural Control
The microcomplex gives an extracerebellar system an adaptive mechanism. Reflexes are classic control systems in the spinal cord and brain stem, which are converted to adaptive control systems by an attached microcomplex. Typical examples are the vestibuloocular reflex and eyeblink conditioning. Compound-movement systems such as those for locomotion or saccades are control systems equipped with a function generator. Innate behavior such as food intake, drinking, and reproductive activity is a form of control involving a motor program. The microcomplex introduces adaptability into these control systems.
When a sensorimotor cortex develops, the micro-complex forms a cerebrocerebellar communication loop, which may provide an internal model of the skeletomuscular system to be controlled by the motor cortex. Command signals generated by the motor cortex may perform precise control using an internal feedback through this internal model of the cerebellum even without referring to the external feedback (see Figure 2A). In the Smith predictor model, the internal model not only represents dynamics of the control object but also incorporates the delay time to be spent in the external feedback, so that it reproduces exactly the same effect as the external feedback. The microcomplex may also be attached parallel to the motor cortex and provides an inverse model of the skeletomuscular system, which also allows the control to be performed without an external feedback (see Figure 2B).
The control with a forward or inverse model shown in Figure 2 can be generalized to the thought in which we move images, concepts, or ideas represented in the temporoparietal cortex by command signals generated by the prefrontal cortex. At the level of neuronal circuit, movement and thought could be controlled by the same mechanism. The thesis of the involvement of the cerebellum in mental activities was based on anatomical connections between the association cortex and cerebellar hemisphere. Brain-imaging studies have afforded evidence of an internal model.
See also:GUIDE TO THE ANATOMY OF THE BRAIN; LONG-TERM DEPRESSION IN THE CEREBELLUM, HIPPOCAMPUS, AND NEOCORTEX; NEURAL SUBSTRATES OF CLASSICAL CONDITIONING: DISCRETE BEHAVIORAL RESPONSES; VESTIBULOOCULAR REFLEX (VOR) PLASTICITY
Albus, A. (1971). A theory of cerebellar function. Mathematical Biosciences 10, 25-61.
Blakemore, S. J., Frith, C. D., and Wolpert, D. M. (2001). The cerebellum is involved in predicting the sensory consequences of action. NeuroReport 12, 1,879-1,884.
Dow, R. S., and Moruzzi, G. (1958). The physiology and pathology of the cerebellum. Minneapolis: University of Minnesota Press.
Fujita, M. (1982). Adaptive filter model of the cerebellum. Biological Cybernetics 45, 195-206.
Groenewegen, H. J., and Voogd, J. (1977). The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of cat cerebellum. Journal of Comparative Neurology 174, 417-488.
Groenewegen, H. J., Voogd, J., Freedman, S. L. (1979). The parasagittal zonation within the olivocerebellar projection. II. Climbing fiber distribution in the intermediate and hemispheric parts of cat cerebellum. Journal of Comparative Neurology 183, 551-601.
Hansel, C., Linden, D., and D'Angelo, E. (2001). Beyond parallel fiber LTD: The diversity of synaptic and nonsynaptic plasticity in the cerebellum. Nature Neuroscience 4, 467-475.
Imamizu, H., Miyauchi, S., Tamada, T., Sasaki, Y., Takino, R., Futz, B. Yoshioka, T., and Kawato, M. (2000). Human cerebellar activity reflecting an acquired internal model of a new tool. Nature 403, 192-195.
Ito, M. (1984). The cerebellum and neural control. New York: Raven Press.
—— (1993). Movement and thought: Identical control mechanisms by the cerebellum. Trends in Neuroscience 16, 448-450.
—— (2001). Long-term depression: Characterization, signal transduction and functional roles. Physiological Reviews 81, 1,143-1,195.
Kawato, M., Furukawa, K., and Suzuki, R. (1987). A hierarchical neuronal network model for control and learning of voluntary movement. Biological Cybernetics 57, 169-185.
Leiner, H. C., Leiner, A. L., and. Dow, R. S. (1986). Does the cerebellum contribute to mental skill? Behavioral Neuroscience 100, 443-453.
Marr, D. (1969). A theory of cerebellar cortex. Journal of Physiology 202, 437-470.
Miall, R. C., Weir, D. J., Wolpert, D. M., and Stein, J. F. (1993). Is the cerebellum a Smith Predictor? Journal of Motor Behavior 25, 203-216.
Oscarsson, O. (1976). Spatial distribution of climbing and mossy fibre inputs into the cerebellar cortex. In O. Creutzfeldt, ed., Afferent and intrinsic organization of laminated structures in the brain. Experimental Brain Research Supp. 1. Berlin: Springer-Verlag.
Rosenblatt, F. (1962). Principles of neurodynamics: Perceptron and the theory of brain mechanisms. Washington, DC: Spartan Books.
Schweighofer, N., Arbib, M., and Dominey, P. F. (1996). A model of the cerebellum in adaptive control of saccadic gain. I. The model and its biological substrate. Biological Cybernetics 75, 19-28.
Schweighofer, N., Arbib, M. and Kawato, M. (1998). Role of the cerebellum in reaching movements in humans. I. Distributed inverse dynamics control. European Journal of Neuroscience 10, 86-94.
That cerebellum's importance as a component of the motor system is clearly indicated by its anatomy and by the effects of lesions or pathology. Abnormal development of or injury to the cerebellum results in severe impairment of the ability to produce accurate and coordinated movements. Normal, everyday movements become uncoordinated and clumsy, hobbled by errors in direction, rate, amplitude, sequence, and precision. Depending on which portions of the cerebellum are damaged, these impairments can include abnormal voluntary movements of the limbs (e.g., reaching movements that miss the target), abnormal eye movements (e.g., saccades that miss the target), or abnormal vestibular (balance) reflexes and postural adjustments (e.g., patients struggling simply to stand without falling over).
How the neurons and synapses of the cerebellum contribute to accurate and coordinated movements has been the subject of much research and debate. The anatomy of the cerebellum indicates that some sort of sensory-motor integration is taking place. There is also evidence that the cerebellum contributes not only to the learning of and adaptation to movements based on experience, but also to the proper timing of movements. These various ideas about cerebellar function are all related to a relatively simple function that the cerebellum performs. It uses sensory information in a particular way (feedforward) to contribute to the accuracy of movements. Understanding what this means and how the cerebellum accomplishes this task requires a basic appreciation of the anatomy and connectivity of the cerebellum.
All vertebrates have a cerebellum (see Figure 1). It lies behind the forebrain and on top of the hind-brain. If you grab the back of your neck (fairly high), your hand surrounds the cerebellum. It is a small, smooth protuberance in most fish, amphibians, and reptiles, but in birds and mammals the surface of the cerebellum is folded into a complex arrangement of thin, elongated folia. The size, shape, and complexity of the cerebellum vary widely in different mammals; those that have more complex behavioral repertoires that are modifiable by learning are likely to have a larger and more convoluted cerebellum.
Figure 2 illustrates some of the unique anatomical features of the cerebellum. A stained section just lateral to the midline reveals the highly folded cerebellar cortex (each fold is a folia), which consists of a three-layer sheet of neurons. The cerebellar cortex is about one millimeter thick in all mammals. Near the base of the cerebellum are several clusters of neurons that make up the cerebellar deep nuclei, which provide the output of the cerebellum. The white matter beneath the cortex and around the deep nuclei consists of numerous axons, providing connections between the cortex, deep nuclei, and other parts of the central nervous system.
All ideas about cerebellar function stem from its unique connectivity or wiring diagram. Although the cerebellum contains a great many neurons, there are only seven types of neurons in the cerebellar cortex that are interconnected in highly organized and specific ways (see Figure 3). Thus, the wiring diagram of the cerebellum lends itself to accessible depiction of the sort shown in Figure 4. Neurons of the deep nuclei provide the output of the cerebellum; they project to the motor nuclei such as the red nucleus and vestibular nuclei, and to the VL region of thalamus, which projects mostly to motor regions of cerebral cortex. The Purkinje cells are the sole output of the cerebellar cortex; they inhibit neurons of the deep nuclei. Mossy fibers represent one type of input to the cerebellum. They project directly to the deep nucleus neurons and to the cerebellar cortex. In the cortex, they branch diffusely to contact numerous granule and Golgi cells. In turn, the granule cell axons, the parallel fibers, make excitatory connections onto many Purkinje cells. Despite its many subtleties, the main properties of the cerebellum's wiring diagram are relatively simple. Output via the deep nucleus neurons is influenced by two parallel pathways: the direct excitatory pathway from mossy fibers and the more complex inhibitory pathway from the Purkinje cells. A simple numeric ratio illustrates the relative complexity of these pathways. For every deep-nucleus output neuron there are about 2 million parallel fiber synapses onto the Purkinje cells.
A second type of input to the cerebellum, the climbing fibers, are quite different from the mossy fibers. Climbing fibers project to the cerebellar cortex and contact only a few Purkinje cells. There are also weak collateral projections to the deep nuclei. Whereas each Purkinje cell has around 100,000 inputs from different granule cells, it gets input from only one climbing fiber. The climbing fiber branches profusely, "climbs" over the dendrites of the Purkinje cell, and makes numerous synapses. This spatially distributed input is so powerful that each action potential in the climbing fiber produces an all-or-none response in the Purkinje cell, which, in part, involves the widespread influx of calcium into the Purkinje cell. This calcium influx is important for the synaptic changes that mediate the learning in the cerebellum.
In 1969 David Marr published a theory of cerebellum that proposed that the wiring diagram described above is well suited to adapt movements via learning. There are three main components of Marr's theory:
- The mossy fiber/granule cell inputs encode what is happening for the Purkinje cells. Mossy fibers convey all sorts of sensory information, especially the position of the body, and information about motor commands from the cerebral cortex. Thus, each combination of motor commands and sensory input would produce a unique pattern of mossy-fiber activity and an even more complex pattern of granule-cell activity. These different patterns would allow Purkinje cells to respond differently under various circumstances.
- Climbing fibers, in contrast, seem to convey signals indicating that a movement should be different from each other.
- Marr then suggested that these climbing-fiber signals would alter the strength of the granule cell to Purkinje synapses that were active at that time. In this way the movement could be different next time, and this learning would be specific for the combination of circumstances encoded by that particular pattern of granule-cell activity.
The basic tenets of this theory have been supported by a wide variety of experimental evidence. There are several well-characterized examples of motor learning that require the cerebellum, and plasticity in the cerebellum controlled by climbing fibers has been firmly established. Results from eyelid-conditioning experiments provide a straightforward example. This simple example of motor learning involves training an animal by repeatedly presenting a relatively neutral stimulus like a tone paired with a reinforcing stimulus like a puff of air in the eye (see Figure 5A). Initially, the eyelids do not move during the tone, but with repeated training, the tone elicits a learned closure of the eyelids. Lesion, stimulation, and recording experiments have shown that the tone-activated mossy-fiber inputs to the cerebellum, the puff activates climbing-fiber inputs, and output from a cerebellar deep nucleus is responsible for the expression of the learned responses (see Figure 5B). Eyelid-conditioning experiments also reveal the temporal specificity or timing component of this cerebellar learning. The eyelids do not close at the onset of the tone (as they do for the puff); instead, the responses are delayed to peak at the time the puff is presented, optimizing the adaptive, protective nature of the response.
Figure 6 illustrates the sequence of events believed to occur during cerebellar learning—for example, eyelid conditioning. The tone activates a certain subset of mossy fibers, which, in turn, activate a subset of the granule cells. All these tone-activated mossy fibers may excite the deep nucleus neurons; eyelid responses are absent in part because the Purkinje cells are spontaneously active, and their inhibition of the nucleus cells prevents a response. Since the puff activates climbing fibers, the synapses made by the tone-activated granule cells onto the Purkinje cells are modified—they are made weaker. Eventually, the Purkinje cells learn to decrease activity during the tone, which releases the deep nucleus cells from inhibition and contributes to the expression of the conditioned responses. Learning seems to involve changes in the deep nuclei as well, which also contribute to the ability of the tone to excite the deep-nucleus neurons and produce a learned conditioned response.
Although such results are a clear illustration, the cerebellum did not evolve to mediate eyelid conditioning. Rather, such experiments reveal the basic capacity of the cerebellum for learning and reveal the basic properties of this learning. From such revelations the purpose of the cerebellum's capacity to learn becomes clearer. It is exactly the type of learning required to permit sensory input to improve the accuracy of movements. Making accurate movement necessarily requires input from sensory systems; this input is revealed by the severe motor impairments that result from sensory impairments like large-fiber neuropathies, wherein information about the position of the body does not reach the brain.
In principle, sensory input can be used in two ways to guide movements. One is via feedback, whereby a movement command is initiated; then sensory input is used during the execution of the movement to make the corrections required for proper performance. Feedback can be very effective and requires no learning, but it is slow. For this reason feedback cannot be used for most of our movements. The other alternative is feedforward. In this mode, sensory input is used at the beginning of a movement (when the motor command is issued) to guess what muscle forces will produce the proper movement. It can be schematized in this way: Given this command and the present sensory input, and based on previous experience, here are the forces that are likely to produce an accurate movement.
Notice that "based on previous experience" means learning. Indeed, it means the type of learning where errors in performance (like those that the climbing fibers convey) make changes so that later, given a similar situation (like those that the mossy fibers convey), performance will improve. If you reach for an object and miss, cerebellar learning makes adjustments so that the next time your hand naturally arrives at the right spot. The cerebellum is constantly using this type of learning to make small adjustments so that our movements remain accurate. We can see, then, that the inaccurate and uncoordinated movements arising from damage to the cerebellum reflect performance without the benefit of the cerebellum's reservoir of previous experience.
See also:GENETIC SUBSTRATES OF MEMORY: CEREBELLUM; LONG-TERM DEPRESSION IN THE CEREBELLUM, HIPPOCAMPUS, AND NEOCORTEX; NEURAL COMPUTATION: CEREBELLUM; NEURAL SUBSTRATES OF CLASSICAL CONDITIONING: DISCRETE BEHAVIORAL RESPONSES; VESTIBULOOCULAR REFLEX (VOR) PLASTICITY
Bloedel, J. R., Dichgans, J., and Precht, W., eds. (1985). Cerebellar functions. New York: Springer-Verlag.
Chan-Palay, V. (1977). Cerebellar dentate nucleus: Organization, cytology and transmitters. New York: Springer-Verlag.
Glickstein, M., Yeo, C., and Stein, J., eds. (1987). Cerebellum and neuronal plasticity. New York: Plenum Press.
Ito, M. (1984). The cerebellum and neural control. New York: Raven Press.
Linden D. J., and Connor, J. A. (1995). Long-term synaptic depression. Annual Review of Neuroscience 18, 319-57.
Marr, D. (1969). A theory of cerebellar cortex. Journal of Physiology 202, 437-470.
Mauk, M. D., and Donegan, N. H. (1997). A model of Pavlovian eyelid conditioning based on the synaptic organization of the cerebellum. Learning and Memory 4, 130-158.
Mauk, M. D., Medina, J. F, Nores, W. L. and Ohyama, T. (2000). Cerebellar function: Coordination, learning or timing? Current Biology 10, R522-R525.
Medina, J. F., Garcia, K. S., Nores, W. L., Taylor, N. M., and Mauk, M. D. (2000). Timing mechanisms in the cerebellum: Testing predictions of a large scale computer simulation. Journal of Neuroscience 20, 5,516-5,525.
Medina, J. F., and Mauk, M. D. (2000). Computer simulation of cerebellar information processing. Nature Neuroscience 3, 1,205-1,211.
Palay, S. L., and Chan-Palay, V. (1974). Cerebellar cortex: Cytology and organization. New York: Springer-Verlag.
Raymond, J., Lisberger, S. G., and Mauk, M. D. (1996). The cerebellum: A neuronal learning machine? Science 272, 1,126-1,132.
Thach, W. T., Goodkin, J. P., and Keating, J. G. (1992). Cerebellum and the adaptive coordination of movement. Annual Review of Neuroscience 15, 403-442.
The cerebellum makes up more than one-tenth of the volume of the human brain. The basic circuitry of nerves within it is essentially similar in all vertebrates and during evolution it has changed much less in size, relative to the body, than have the cerebral hemispheres. These facts suggest that it has some essential, basic function in all vertebrates. Although its exact mechanisms remain unclear, its fundamental role is in the control of movement. This was clearly recognized by the seventeenth-century physician Thomas Willis in his book Cerebri Anatome (1664) and the idea can be traced back to the observations and interpretations of Galen (c.130–210 ad).
The cerebellum comprises an outer, thin layer of grey matter — the cerebellar cortex — covering a core of white matter, within which lie three lumps of grey matter on each side of the midline (the deep cerebellar nuclei). Closest to the midline is the fastigial nucleus and furthest from it is the dentate nucleus, with the interpositus nucleus between.
The surface area of the cortex is greatly augmented by folds that run across from side to side — deep ones that divide the surface into ten lobules, and numerous shallower ones cutting each lobule into folia. If the cortex were flattened out, it would be a ribbon much longer than it is wide.
The cortex is divided up functionally into longitudinal (i.e. fore-and-aft) strips or zones, each interconnected with a particular deep nucleus. The vermis, running down the middle, connects with the right and left fastigial nuclei. This is flanked on each side by a paravermal cortical zone related to nucleus interpositus; and most lateral is the pair of large cerebellar hemispheres, linked to the dentate nuclei. Since the 1960s studies in animals have shown that each cortical zone comprises many narrower micro-zones, each relating to a particular ‘private’ portion of the corresponding deep nucleus.
The fine structure of the cortex and the circuits that link it with the deep nuclei vary little from place to place, which suggests that all parts of the cerebellum perform a similar basic ‘computation’ or operation. If different parts of the cerebellum have different functional roles, this must be due to differences in their input and output connections rather than their internal wiring.
Damage to part or even all of the human cerebellum, on its own, does not lead to clear impairment of intellect, emotion, or vegetative functions (such as the control of the heart and breathing). But there is abundant evidence that the control of movements is markedly disordered. Typically, patients with cerebellar damage are unsteady on their feet, and their hands shake as they try to point or lift objects (‘intention tremor’); their eyes swing uncontrollably from side to side (nystagmus); and even their speech can be jerky (‘scanning speech’). These three typical signs, described by the great nineteenth-century French neurologist Charcot, are known as ‘Charcot's Triad’.
The movements most affected vary somewhat depending on the location of the damage. No type of movement is completely lost, but movements ranging in complexity from simple reflex actions to walking, speech, and highly skilled manipulations may all be defective in rate, range, force, and timing. Extremely rarely, individuals are born with little or no cerebellum, and although some of its functions may be taken over by other parts of the brain, movements are permanently clumsy and poorly co-ordinated, suggesting that the learning of motor skills is impaired.
The 600 000 nerve cells in the deep nuclei send messages out of the cerebellum along their fibres (or axons), which run through the peduncles to a number of nuclei in the brain stem and thalamus. These in turn are connected to the spinal cord and to regions of the cerebral cortex concerned with the control of movement.
Studies of the activity of nerve cells in animals have been the main source of knowledge of how the cerebellum works. Even when movements are not being made, neurons of the deep nuclei are continuously active, producing impulses at rates of 30–50 per second. This continuous background firing arises because the huge number of excitatory nerve fibres that enter the cerebellum, carrying information to the cortex, send side branches into the deep nuclei. In addition, they receive the axons of the 15 million Purkinje cells, the largest cells in the cortex. These are all inhibitory, using gamma-amino-butyric acid (GABA) as their transmitter. So, the variation of firing of cells in the deep nuclei, which constitutes the output of the cerebellum and hence modulates movement, is dependent on the relationship between incoming activity and the resulting firing of Purkinje cells.
The incoming nerve fibres, which ultimately control the firing of Purkinje cells, are of two types. The first are the axons of cells in a nucleus in the medulla of the brain stem that glories in the name inferior olive, and which receives signals, indirectly, from parts of the cerebral cortex concerned with movement. Each of these axons wraps itself around the huge bush of processes (dendrites) of just one Purkinje cell (hence their name, ‘climbing fibres’), ending in around 2000 synapses. As a result, even a single impulse in a climbing fibre will make its Purkinje cell fire an impulse.
The other class of incoming axons are called ‘mossy fibres’. They are the fibres of several different kinds of nuclei in the brain stem and spinal cord. Some 40 million of them arise from cells in a region of the pons called the pontine nuclei. Some mossy fibres carry signals from the eyes, inner ears, skin, muscles, and tendons, providing information about the state and posture of the body. Because movements inevitably generate sensory stimulation, these messages must include ‘feedback’ information regarding current patterns of movement. Other mossy fibres (the majority) carry signals originating in various areas of the cerebral cortex, probably including copies of the current ‘commands-to-move’ emanating from motor areas of the cortex. They inform the cerebellum about movement intentions even before any motion has begun, enabling it to modify movements before errors have started to occur. This essentially ‘predictive’ revision is thought to reduce the extent to which the control of movement depends on feedback from sensory receptors about actual, achieved movement. This is very useful because feedback obviously cannot begin until movement has started, and the delay in a control system causes oscillations and other errors, as engineers well know.
The mossy fibres do not contact Purkinje cells directly: they end mainly on the 50 billion tiny granule cells in the cortex, whose long parallel fibres each form synaptic connections on many Purkinje cells. About 95% of the impulses produced by Purkinje cells result from the stream of signals from granule cells.
But how does it all work? Although we know more about the micro-anatomy of the cerebellum than of any other area of the brain, there is still intense debate about exactly what it does. One of the complicating factors is that the strength of each synapse between any parallel fibre and the large number of Purkinje cells that it contacts can be changed, in ways that are invisible even under the microscope. Technically elegant experiments, involving recording from Purkinje cells in slices of cerebellum, maintained alive in vitro, show that when the cell is activated by its climbing fibre, the synapses of any parallel fibres that are simultaneously active are decreased in effectiveness, and that this ‘long-term depression’ lasts a very long time. This implies that the climbing fibre can, in effect, ‘teach’ the Purkinje cell to alter its response to any recurrence of the particular pattern of mossy fibre input it was experiencing (representing a particular sensory and motor state of the body) when the climbing fibre was activated. This line of thinking is not universally accepted but has prompted attempts to identify the circumstances (behavioural contexts) in which the climbing fibres increase their activity. At present, the slim available evidence suggests that this occurs when a mismatch develops between the commands-to-move issued to the muscles by the central nervous system and the movements that actually ensue. Climbing fibres may, therefore, function (at least in part) as error-detectors in movement control.
This hypothesis also implies that the cerebellar cortex is the repository of many learned responses or ‘motor memories’ that help to ensure the prompt and accurate execution of skilled movements. Whenever these memories prove to be inadequate, either because a novel movement command is required or because they are fading, control errors will be made and the teaching effect of the climbing fibres will automatically come into play, gradually reducing the errors and improving the skill.
David M. Armstrong
See nervous system.See also brain; cerebral cortex; memory; movement, control of.
The cerebellum is a cauliflower-shaped brain structure located just above the brainstem, beneath the occipital lobes at the base of the skull.
The word cerebellum comes from the Latin word for "little brain." The cerebellum has traditionally been recognized as the unit of motor control that regulates muscle tone and coordination of movement. There is an increasing number of reports that support the idea that the cerebellum also contributes to non-motor functions such as cognition (thought processes) and affective state (emotion).
The cerebellum comprises approximately 10% of the brain's volume and contains at least half of the brain's neurons. The cerebellum is made up of two hemispheres (halves) covered by a thin layer of gray matter known as the cortex. Beneath the cortex is a central core of white matter. Embedded in the white matter are several areas of gray matter known as the deep cerebellar nuclei (the fastigial nucleus, the globise-emboliform nucleus, and the dentate nucleus). The cerebellum is connected to the brainstem via three bundles of fibers called peduncles (the superior, middle, and inferior).
The cerebellum is a complex structure. At the basic level, it is divided into three distinct regions: the vermis, the paravermis (also called the intermediate zone), and the cerebellar hemispheres. Fissures, deep folds in the cortex that extend across the cerebellum, further subdivide these regions into 10 lobules, designated lobules I–X. Two of these fissures in particular, the posterolateral fissure and the primary fissure, separate the cerebellum into three lobes that have different functions: the flocculonodular lobe, or the vestibulocerebellum (lobule X); the anterior lobe (lobules I–V); and the posterior lobe (lobules VI–IX).
The cerebellum plays an important role in sending and receiving messages (nerve signals) necessary for the production of muscle movements and coordination. There are both afferent (input) and efferent (output) pathways. The major input pathways (also called tracts) include:
- dorsal spinocerebellar pathway
- ventral spinocerebellar pathway
- corticopontocerebellar pathway
- cerebo-olivocerebellar pathway
- cerebroreticulocerebellar pathway
- cuneocerebellar pathway
- vestibulocerebellar pathway
The major output pathways include the following:
- globose-emboliform-rubral pathway
- fastigial reticular pathway
- dentatothalamic pathway
- fastigial vestibular pathway
There is a network of fibers (cells) within the cerebellum that monitors information to and from the brain and the spinal cord. This network of neural circuits links the input pathways to the output pathways. The Purkinje fibers and the deep nuclei play key roles in this communication process. The Purkinje fibers regulate the deep nuclei, which have axons that send messages out to other parts of the central nervous system .
The flocculonodular lobe helps to maintain equilibrium (balance) and to control eye movements. The anterior lobe parts of the posterior lobe (the vermis and paravermis) form the spinocerebellum, a region that plays a role in control of proximal muscles, posture, and locomotion such as walking. The cerebellar hemispheres (part of the posterior lobe) are collectively known as the cerebrocerebellum (or the pontocerebellum); they receive signals from the cerebral cortex and aid in initiation, coordination, and timing of movements. The cerebrocerebellum is also thought to play a role in cognition and affective state.
The cerebellum has been reported to play a role in psychiatric conditions such as schizophrenia , autism , mood disorders, dementia , and attention deficit hyperactivity disorder (ADHD). Currently, the relationship between the cerebellum and psychiatric illness remains unclear. It is hoped that further research will reveal insights into the cerebellar contribution to these conditions.
There are a variety of disorders that involve or affect the cerebellum. The cerebellum can be damaged by factors including:
- toxic insults such as alcohol abuse
- paraneoplastic disorders; conditions in which autoantibodies produced by tumors in other parts of the body attack neurons in the cerebellum
- structural lesions such as strokes, multiple sclerosis , or tumors
- inherited cerebellar degeneration such as in Friedreich ataxia or one of the spinocerebellar ataxias
- congenital anomalies such as cerebellar hypoplasia (underdevelopment or incomplete development of the cerebellum) found in Dandy-Walker syndrome , or displacement of parts of the cerebellum such as in Arnold-Chiari malformation
Typical symptoms of cerebellar disorders include hypotonia (poor muscle tone), movement decomposition (muscular movement that is fragmented rather than smooth), dysmetria (impaired ability to control the distance, power, and speed of an act), gait disturbances (abnormal pattern of walking), abnormal eye movement, and dysarthria (problems with speaking).
De Zeeuw, C. I., P. Strata, and J. Voogd, eds. The Cerebellum: From Structure to Control. St Louis, MO: Elsevier Science, 1997.
Daum, I., B. E. Snitz, and H. Ackermann. "Neuropsychological Deficits in Cerebellar Syndromes." International Review of Psychiatry 13 (2001): 268–275.
Desmond, J. E. "Cerebellar Involvement in Cognitive Function: Evidence from Neuroimaging." International Review of Psychiatry 13 (2001): 283–294.
Leroi, I., E. O'Hearn, and R. Margolis. "Psychiatric Syndromes in Cerebellar Degeneration." International Review of Psychiatry 13 (2001): 323–329.
O'Hearn, E., and M. E. Molliver. "Organizational Principles and Microcircuitry of the Cerebellum." International Review of Psychiatry 13 (2001): 232–246.
Rapoport, M. "The Cerebellum in Psychiatric Disorders." >International Review of Psychiatry 13 (2001): 295–301.
Schmahmann, J. D. "The Cerebrocerebellar System: Anatomic Substrates of the Cerebellar Contribution to Cognition and Emotion." International Review of Psychiatry 13 (2001): 247–260.
Shill, H. A., and M. Hallett. "Cerebellar Diseases." International Review of Psychiatry 13 (2001): 261–267.
"BrainInfo Web Site." Cerebellum Information Page. Neuroscience Division, Regional Primate Research Center, University of Washington, 2000. (May 22, 2004.) <http://braininfo.rprc.washington.edu>.
The Cerebellum Database Site. (May 22, 2004). <http://www.cerebellum.org/8home/>.
The National Institute of Neurological Disorders and Stroke (NINDS). Cerebellar Degeneration Information Page. PO Box 5801 Bethesda, MD, 2003. (May 22, 2004). <http://www.ninds.nih.gov/health_and_medical/disorders/cerebellar_degeneration.htm>.
The National Institute of Neurological Disorders and Stroke (NINDS). Cerebellar Hypoplasia Information Page. PO Box 5801 Bethesda, MD, 2003. (May 22, 2004). <http://www.ninds.nih.gov/health_and_medical/disorders/cerebellar_hypoplasia.htm>.
National Institute of Mental Health. 6001 Executive Boulevard, Room 8184, MSC 9663, Bethesda, MD 20892-9663. (301) 443-4513 or (866) 615-6464; TTY: (301) 443-8431; Fax: (301) 443-4279. [email protected] <http://www.nimh.nih.gov/>.
National Institute of Neurological Disorders and Stroke (NINDS), NIH Neurological Institute. P.O. Box 5801, Bethesda, MD 20824. (301) 496-5751 or (800) 352-9424; TTY: (301) 468-5981. <http://www.ninds.nih.gov/>.
cerebellum (sĕr´əbĕl´əm), portion of the brain that coordinates movements of voluntary (skeletal) muscles. It contains about half of the brain's neurons, but these particular nerve cells are so small that the cerebellum accounts for only 10% of the brain's total weight. The cerebellum operates automatically, without intruding into consciousness; motor impulses from the cerebrum are organized and modulated before being transmitted to muscle. As the muscle tissue responds, its sensory nerve cells return information to the cerebellum. Thus, throughout periods of muscular activity, the cerebellum adjusts speed, force, and other factors involved in movement. The overall effect is a smooth, balanced muscular activity. If the cerebellum is injured, an activity like walking becomes spasmodic: the muscles involved contract too much or too little and operate out of sequence. Maintaining muscle tone is also a function of the cerebellum. Filling most of the skull behind the brain stem and below the cerebrum, the human cerebellum approximates an orange in size and consists of two hemispherical lobes. The grooved surface of the cerebellum is gray matter, composed chiefly of nerve cells. The interior, dense with nerve fibers, makes up the white matter. Five different nerve cell types make up the cerebellum: stellate, basket, Purkinje, Golgi, and granule cells. The Purkinje cells are the only ones to send axons out of the cerebellum. Three main nerve tracts link the cerebellum with other brain areas. Injury to the cerebellum usually results in disruption of eye movements, balance, or muscle tone.
cer·e·bel·lum / ˌserəˈbeləm/ • n. (pl. -bel·lums or -bel·la / -ˈbelə/ ) Anat. the part of the brain at the back of the skull in vertebrates. Its function is to coordinate and regulate muscular activity. DERIVATIVES: cer·e·bel·lar adj.