NEUROSCIENCE

LECTURE
SUPPLEMENT
Nachum Dafny, Ph.D., Professor
Department of Neurobiology and Anatomy
University of Texas Medical School at Houston

C. Motor System
 TABLE OF CONTENTS
Page

Overview of the Motor System................................................................................1

Motor Units and Muscle Receptors .......................................................................12

Spinal Reflexes ......................................................................................................23

Cerebellum.............................................................................................................33

Basal Ganglia .........................................................................................................43

Motor Cortex..........................................................................................................53

Integrated Motor System and Disorders of the Motor System ..............................61

i
 OVERVIEW OF THE MOTOR SYSTEM

James Knierim, Ph.D.

Much of the brain and nervous system is devoted to the processing of sensory input, in
order to construct detailed representations of the external environment. Through vision,
audition, somatosensation, and the other senses, we perceive the world and our relationship to it.
This elaborate processing would be of limited value, however, unless we had a way to act upon
the environment that we are sensing, whether that action consist of running away from a
predator; seeking shelter against the rain or wind; searching for food when one is hungry;
moving one’s lips and vocal cords in order to communicate with others; or performing the
countless other varieties of actions that make up our daily lives. In some cases the relationship
between the sensory input and the motor output are simple and direct; for example, touching a
hot stove elicits an immediate withdrawal of the hand (Fig. 1). Usually, however, our conscious
actions require not only sensory input but a host of other cognitive processes that allow us to
choose the most appropriate motor output for the given circumstances. In each case, the final
output is a set of commands to certain muscles in the body to exert force against some other
object or forces (e.g., gravity). This entire process falls under the category of motor control.

ENVIRONMENT Fig. 1. Sensory receptors provide information about the

environment, which is then used to produce action to change
the environment. Sometimes the pathway from sensation to
action is direct, as in a reflex. In most cases, however,
cognitive processing occurs to make actions adaptive and
appropriate for the particular situation.
Senses Action

Cognition

Some Necessary Components of Proper Motor Control

(1) Volition. The motor system must generate movements that are adaptive and that
accomplish the goals of the organism. These goals are evaluated and set by high-order areas of
the brain, including the prefrontal cortex. The motor system must transform these goals and
desired movements into the appropriate activations of muscles to perform the desired activity.

(2) Coordination of signals to many muscle groups. Few movements are restricted to
the activation of a single muscle. Rather, most movements result from the coordinated activity
of different muscle groups. The act of moving your hand from inside your pocket to a position
in front of you requires the coordinated activity of the shoulder, elbow, and wrist. Making the
same movement while removing a 10-lb weight from your pocket may result in the same
trajectory of your hand, but will require different sets of forces on the muscles that make the
movement. The task of the motor system is to determine the necessary forces and coordination
at each joint in order to produce the final, smooth motion of the arm.

1
 (3) Proprioception. In order to make a desired movement (e.g., raising your hand to ask
a question), it is essential for the motor system to know the starting position of the hand. Raising
one’s hand from a resting position on a desk, compared to a resting position on top of the head,
results in the same final position of the arm, but these two movements require a very different
pattern of muscle activation. The motor system has a set of sensory inputs (called
proprioceptors) that inform it of the length of muscles and the forces being applied to them; it
uses this information to calculate joint position and other variables necessary to make the
appropriate movement.

(4) Postural adjustments. In addition to the coordination of muscles necessary to

produce the desired output, the motor system must also constantly produce adjustments to the
body’s posture in order to compensate for the changes in the body’s center of mass as we move
our limbs, head, and torso. Without these automatic adjustments, the simple act of reaching for a
cup would cause us to fall forward, as the body’s center of mass shifts to a location in front of
the body axis.

(5) Sensory feedback. In addition to the use of proprioception to sense the position of
the body before a movement, the motor system must use other sensory information in order to
perform the movement accurately. By comparing desired activity with actual activity, sensory
feedback allows for corrections in movements as they take place, and it also allows
modifications to motor programs so that future movements are performed more accurately.

(6) Compensation for the physical characteristics of the body and muscles. To exert
a defined force on an object, it is not sufficient to know only the characteristics of the object
(e.g., its mass, size, etc.). The motor system must also account for the physical characteristics of
the body and muscles themselves. The bones and muscles have mass that must be considered
when moving a joint, and the muscles themselves have a certain degree of resistance to
movement.

(7) Unconscious processing. The motor system must perform many procedures in an
automatic fashion, without the need for high-order control. Imagine if walking across the room
required thinking about planting the foot at each step, paying attention to the movement of each
muscle in the leg and making sure that the appropriate forces and contraction speeds are taking
place. It would be hard to do anything else but that one task. Yet we can walk, talk, chew gum,
button a shirt, and think about how fascinating neuroscience is, all at the same time. A number
of motor tasks are performed in an automatic fashion that does not require effortful processing
by higher brain areas. Moreover, the motor system must be able to perform a number of tasks of
which the organism is completely unaware. For example, many of the postural adjustments that
the body makes during movement are performed without our awareness. These unconscious
processes allow higher-order brain areas to concern themselves with broad desires and goals,
rather than low-level implementations of movements.

(8) Adaptability. The motor system must adapt to changing circumstances. For
example, as a child grows and its body changes, different constraints are placed on the motor
system in terms of the size and mass of bones and muscles. The motor commands that work to

2
raise the hand of an infant would fail completely to raise the hand of an adult. The system must
adapt over time to change its output to accomplish the same goals. Furthermore, if the system
were unable to adapt, we would never be able to acquire motor skills, such as playing a piano,
hitting a baseball, or performing microsurgery.

These are some of the many components of the motor system that allow us to perform
complex movements in a seemingly effortless way. The brain has evolved exceedingly complex
and sophisticated mechanisms to perform these tasks, and researchers have only scratched the
surface in understanding the principles that underlie the brain’s control of movement.

Functional Segregation and Hierarchical Organization

The ease with which we make most of our movements belies the enormous sophistication
and complexity of the motor system. If you want to get a cup of coffee, you simply stand up,
walk to the coffee pot, and pour yourself a cup. Simple, right? Yet very smart engineers have
spent decades trying to get machines to perform such simple tasks, and the most advanced
robotic systems do not come close to emulating the precision and smoothness of movement,
under all types of conditions, that we achieve effortlessly and automatically. How does the brain
do it? Although many of the details are not understood, two broad principles appear to be key
concepts toward understanding motor control:

Functional Segregation. The motor system is divided into a number of different areas
that control different aspects of movement. These areas are located throughout the nervous
system. One of the key questions of research on motor control is to understand the functional
roles played by each area.

Hierarchical Organization. The different areas of the motor system are organized in a
hierarchical fashion. Lower levels of the hierarchy control the “nuts and bolts” of motor
processing, such as calculating the amount of force generated by a single muscle and
coordinating simple reflexes. Higher levels of the hierarchy calculate the trajectories of whole
limb movements and sequences of movements, and they evaluate the appropriateness of a
particular action given the current environmental context.

Because of the hierarchical nature of the motor system, the higher-order areas can
concern themselves with more global tasks regarding action, such as deciding when to act,
devising an appropriate sequence of actions, and coordinating the activity of many limbs. They
do not have to program the exact force and velocity of individual muscles, or coordinate
movements with changes in posture; these low-level tasks are performed by the lower levels of
the hierarchy.

The motor system hierarchy consists of the following parts (Fig. 2):

• Spinal Cord
• Brainstem
• Motor Cortex

3
 • Association Cortex
• Side Loops
o Basal Ganglia
o Cerebellum

Level 4: Association Cortex Fig. 2. Schematic

Side Loop 1:
Basal Ganglia
Level 3: Motor Cortex (Caudate Nucleus, Putamen,
Globus Pallidus, Substantia
Nigra, Subthalamic Nucleus)
representation of the different
levels and interconnections of
the motor system hierarchy.

Thalamus

Level 2: Brain Stem (VA,VL,CM)

(Red Nucleus, Reticular

Formation, Vestibular Nuclei, Side Loop 2:
Tectum, Pontine Nuclei, Inferior Cerebellum
Olive)

Level 1: Spinal Cord

Spinal Cord. The spinal cord is the first (lowest) level of the motor hierarchy (Fig. 3).

(1) The spinal cord is the site of motor neurons. These motor neurons reside in the anterior
horn of the spinal cord and synapse directly onto muscle fibers. Motor neurons are the
only way in which commands from higher areas can be transmitted to muscle. Damage to
motor neurons results necessarily in paralysis of the muscles that were innervated by
those neurons.
(2) The spinal cord contains the circuitry for many reflexes. These spinal reflexes include
the myotatic reflex (also called the stretch reflex or deep tendon reflex), the flexor reflex
(limb withdrawal away from a painful stimulus), and many others discussed in the next 2
lectures.
(3) The spinal cord controls many complex actions. The most obvious complex action that is
controlled by circuitry entirely within the spinal cord is gait. Although higher areas can
issue commands to begin walking, the circuitry that actually controls the rhythmic motion
of the legs resides in the spinal cord itself.
(4) The spinal cord is influenced by higher levels of the hierarchy. Although many reflexes
and complex actions are controlled by neural circuits within the spinal cord, these circuits
can be influenced in complex ways by higher brain centers. This top-down control is
necessary for ensuring that the low-level circuits within the spinal cord are utilized in an
adaptive manner.

4
 Fig. 3. A motor neuron in the spinal cord.

Brainstem. The brainstem is the second level of the motor hierarchy.

(1) Brainstem nuclei that are important for motor control include the red nucleus, the
vestibular nuclei, the pontine reticular formation, and the medullary reticular formation.
(2) These nuclei process selected afferent sensory inputs that are used to program and adapt
motor commands. As discussed later, the use of sensory input to guide motor output is a
necessary and ubiquitous feature of motor control.
(3) Brainstem nuclei modulate motor circuits that control posture, eye movements, and head
movements.

Motor cortex. The motor cortex is the third level of the motor hierarchy (Fig. 4).

(1) The motor cortex processes numerous task-related variables to produce the desired
action. Neurons in the motor cortex then give rise to descending motor commands.
Information encoded in these commands includes which part of the body should move, as
well as the force and direction of the movement.
(2) Motor cortex can be divided broadly into 3 areas: primary motor cortex, premotor
cortex, and the supplementary motor area.
(3) Descending pathways from the motor cortex can be divided into two systems:
a. The corticospinal system controls motor neurons and interneurons in the spinal
cord.
b. The corticobulbar system controls brainstem nuclei that innervate cranial muscles.

Fig. 4. The motor cortex, the third level of the

motor system hierarchy, is composed of three
major areas: the primary motor cortex (also
called M1), the premotor cortex, and the
supplementary motor area. Note that the
supplementary motor area extends onto the
medial wall of the cortex.

5
Association Cortex. The association cortex is the fourth (highest) level of the motor hierarchy
(Fig. 5).

(1) High-level association cortical areas create a frame of reference for directing
movements. This task is largely associated with the posterior parietal cortex. In addition
to other tasks, this brain region calculates transformations of sensory inputs from body-
centered (egocentric) coordinates to world-centered (allocentric) coordinates, allowing
movements to be directed to the external world regardless of the current orientation and
position of the body. Lower areas of the hierarchy transform the output of these areas
back to the appropriate egocentric coordinates necessary to generate the proper muscle
output. The posterior parietal cortex is also involved in directing attention to salient
objects in the world.
(2) Association cortex integrates behavior to produce goal-directed action appropriate for
the particular context. These poorly understood functions are largely associated with the
prefrontal cortex. Actions that are appropriate in one behavioral context (e.g., giving a
hard slap on the back to congratulate a basketball player who just made a critical 3-
pointer) may be completely inappropriate in another context (e.g., giving the same hard
slap to one’s grandmother to celebrate her 80th birthday). This type of processing,
although not traditionally motor processing per se, is important for producing movements
that are behaviorally adaptive.

Fig. 5. The fourth level of the motor

system hierarchy is the association
cortex, primarily the posterior
parietal cortex and the prefrontal
cortex.

Side loops. Two important brain regions are not part of the motor hierarchy, but they influence
the motor cortex profoundly through their connections with the motor thalamus.

(1) Basal ganglia. The basal ganglia comprise a number of distinct forebrain structures that
act as an integrative center for motor output. Damage to the basal ganglia produces
deficits of motor planning, speed of movement, and the ability to enable certain
stereotyped motor programs. Parkinson’s disease is the most well-known of basal
ganglia disorders.
(2) Cerebellum. The cerebellum modulates the activity of brainstem nuclei (through direct
connections) as well as motor cortex (through its thalamic connections). The cerebellum
is important for fine control and timing of movements; it is thought to be necessary for
motor learning (i.e., learning the precise patterns of movement necessary to achieve fine

6
 control of the body and limbs). Damage to the cerebellum can produce ataxia (lack of
coordination in movements) and inability to maintain balance, among other deficits.

Parallel and Serial Processing. Although the motor system is organized in the
hierarchical fashion described here, it is important to realize that the hierarchy is not a simple
chain of processing from higher to lower areas. Many routes enable the different levels of the
hierarchy to influence each other through multiple pathways (Fig. 6). Thus, the flow of
information through the motor system has both a serial organization (communication between
levels) and a parallel organization (multiple pathways between each level). This parallel
organization is critically important in understanding the various dysfunctions that can result from
damage to the motor system. If the motor hierarchy had a strictly serial organization, like a
series of links on a chain, then damage to any part of the system would produce severe deficits or
paralysis in almost all types of movements. However, because of the parallel nature of
processing, paralysis is actually a relatively rare outcome, produced by damage to the lowest
level of the hierarchy. Damage to higher levels results in deficits in motor planning, initiation,
coordination, and so forth, but movement is still possible. The parallel nature of organization is
also important for the ability of undamaged parts of the motor system to compensate (at least
partially) for injuries to other parts of the system.

Fig. 6. Parallel and serial organization of the motor system. Higher areas
can direct lower areas by multiple pathways (parallel organization) and
through polysynaptic processing chains (serial organization).

Motor Control Requires Sensory Input

One of the major principles of the motor system is that motor control requires sensory
input to accurately plan and execute movements. This principle applies to low levels of the
hierarchy, such as spinal reflexes, and to higher levels. The study of motor control has utilized
principles from the engineering field of control theory to characterize the interactions between
sensory input and motor output. There are three types of control systems in this framework:
open loop system with no sensory input, closed loop system with feedback control, and open
loop system with feedforward control.

7
 Open Loop System with No Sensory Input. Before describing the role of sensory
input, it is useful to describe the properties of a control system without such input. A system in
which there is no sensory feedback into the system is termed an open loop control system. In
such a system, a desired output is fed into a controller circuit, which in turns directs the effector
machinery (e.g., robotic arm, biceps muscle, etc.) to produce the output (Fig. 7). In this type of
system, there is no role of sensory information to direct, guide, or modulate the output;
movements are ballistic and, once initiated, cannot be modified. Such a system is rarely present
in a biological motor system. A nonbiological example would be a timer-controlled
heating/cooling system. In order to cool a room to a desired temperature, a control circuit with a
timer would turn on the air conditioner for a preset amount of time. Once the timer was set,
however, there would be no way to keep the air conditioner on longer if the desired temperature
was not reached, or to turn on the heater if the air conditioner made the room too cold. Perhaps
one of the few examples of a biological realistic example of such a ballistic system would be a
reflexive drop to the ground to avoid an incoming object.

DESIRED
CONTROLLER EFFECTOR OUTPUT
OUTPUT

Fig. 7. An open loop control system.

Closed Loop System with Feedback Control. An improvement on an open loop

control system is to incorporate sensory information during the execution of a task in order to
improve accuracy. One method of accomplishing this is a feedback control system (Fig. 8). In
such a system, a desired output is sent to a comparator, which compares the present state of the
system with the desired output. If there is a mismatch between the present and desired outputs,
an error signal is sent to the effector machinery that instructs it to bring the actual output closer to
the desired output. In order to detect the present state of the system, a sensor must be present to
measure the present state of the system. The flow of information, from the comparator to the
effector, to the sensor, and back to the comparator, is called a closed loop.

A common example of a feedback control system is the thermostat in your home. The
thermostat is set to a desired temperature (e.g., 72°), and a thermometer measures the current
temperature in the room. If the comparator detects that the room is cooler than the desired
temperature, it sends an error signal that turns on the furnace. If the comparator detects that the
room is warmer than the desired setting, its sends an error signal that turns on the air conditioner.

COMPARATOR

DESIRED OUTPUT
+ Error EFFECTOR OUTPUT
signal
-

Feedback signal
SENSOR

Fig. 8. A closed loop system with feedback control

8
 Feedback control systems can produce very accurate outputs; however, in general they
are slow. In order to change the output, the effector must wait until information is transmitted
from the sensor to the comparator and then to the effector. At this point, another comparison is
made, and the process continues. Consider further the thermostat example. If the temperature
reads 5° cooler than desired, the thermostat can instruct the furnace to turn on at a moderate heat.
It reads the new room temperature, and, if it is still too cool, it instructs the furnace to deliver
more heat, and so on. Although this will eventually produce an accurate room temperature at the
desired point, it takes a number of cycles to reach that point. One possible solution for quicker
results would be to turn an enormous furnace on full-blast, such that is heats the room very
quickly. This solution, however, can generate another problem. It will tend to cause the system
to oscillate if the feedback pathways are slow. For example, assume that the furnace can heat the
room at the rate of 5° per second, but that it takes 2 seconds for the thermometer to adjust to the
new temperature, and for the new error signal to turn the furnace off. In those 2 seconds, the
furnace has heated the room up 10°, and now it is too warm. So the error signal turns on the air
conditioner, and it cools the room at 5°/sec. Of course, it also takes 2 sec to receive the
feedback, and by the time it is told to shut off, it has cooled the room by 10°. You can see what
happens: the system will be sent into an endless oscillation of being 5° too hot and 5° too cold.
In order for a feedback system to work well, the transmission time of sensory information
through the comparator to the effector must be rapid compared to the time of the action.

Thus, the advantages of a feedback control system are:

(1) It can produce very accurate output.
(2) It is a very efficient mechanism in that all it requires of the operator is to set the desired
output level, and the system automatically adjusts itself to maintain that level. The
operator does not need to manually turn the effectors on or off to constantly adjust the
movement.

The disadvantages are

(1) It can be a slow system, often requiring many iterative cycles to produce the final desired
output.
(2) It is prone to oscillations.

As we shall see, many simple reflexes can be understood as feedback control systems, but such
systems are often too slow to control the accurate execution of many voluntary movements.

Open Loop System with Feedforward Control. A second manner in which sensory
information can be used to guide movements is by incorporating that information in advance of
the movement, during the planning and programming stages of motor execution. Such a system
is a feedforward control system (Fig. 9). In this type of system, when a desired output is sent to
the controller, the controller takes readings from sensors about the state of the environment and
about the current state of the effector itself. It then uses this information to program the best set
of instructions to accomplish the desired output. Importantly, in a pure feedforward system,
once the commands to the effector are sent, there is no feedback pathway to alter the effector
during the execution of the movement. This is why it is termed an “open loop” system. Note
that a feedforward system differs from the open loop system with no sensory control, as the
feedforward system uses sensory information to plan the movement.

9
 The major advantage of the feedforward control system over the feedback system is
speed of execution. Whereas the feedback system can be slow and prone to oscillations, the
feedforward system (when working well) can produce the precise set of commands for the
effector without needing to constantly check the output and make corrections during the
movement itself. The main disadvantage, however, is that the feedforward controller requires a
period of learning before it can function properly. In most biological systems, the environment
and conditions under which actions are made are constantly changing, and the feedforward
controller must be able to adapt its output commands to account for this variability; it is hard
(perhaps impossible) to pre-program all of the possible sensory conditions that the controller
may encounter during the life of the organism.

DESIRED
OUTPUT ADVANCE
SENSOR
INFORMATION

FEED-FORWARD
CONTROLLER
Feed-forward
control signal

EFFECTOR OUTPUT

Fig. 9. An open loop system with feedforward control uses sensory information in advance to output an appropriate
control signal to the effector.

Let us extend the thermostat example to see how a temperature controller operating as a
feedforward system would work to raise the temperature of a room from 70° to 75°. The
controller would use diverse sensory information about the environment before sending its
command to the furnace. For example, it would read the current temperature, the current
humidity level, the size of the room, the number of people in the room, and so forth. Based on
this information, it would direct the furnace to turn on for a pre-set period of time, and that’s it.
There would be no need to continually compare the current temperature with the desired setting,
as the system has predetermined how long the furnace needs to be working in order to achieve
the desired temperature. How did the controller obtain this information? A feedforward
controller requires a large amount of experience in order to learn the appropriate actions needed
for each set of environmental conditions. If on one trial it turns the furnace off too soon and the
room does not reach the desired temperature, it adjusts its programming such that the next time it
encounters the same environmental conditions, it turns the furnace on for a longer period of time.
Through many such instances of trial and error learning, the feedforward system creates a “look-
up table” that tells it how long the furnace needs to be active under the current conditions. The
key distinction between a feedback and feedforward system is that the feedback system uses
sensory information to generate an error signal during the control of a movement, whereas a
feedforward system uses sensory information in advance of a movement. Any error signal about

10
the final output is used by the feedforward system only to change its programming of future
movements.

Thus, the advantages of a feedforward control system are

(1) It is very accurate
(2) It is very fast
(3) It does not require constant monitoring of the output during the execution of the
movement

The disadvantages are

(1) It requires a period of learning before it can perform accurately

Motor deficits from sensory pathology. There are specific motor deficits seen with
particular sensory pathologies. The importance of feedback and feedforward control is evident
in the effects of loss of sensory input. This occurs in large fiber sensory neuropathy. A patient
with large fiber sensory neuropathy cannot sense their position nor detect motion of joints, since
input from muscle spindles and Golgi tendon organs is not present. Tactile information is also
impaired. Manual dexterity is devastated, since estimates of contact with objects cannot be made
precisely. However, pain and temperature sensation is preserved. These deficits are quite severe:
limb position can be maintained only if the patient can see them. In this case, patients must learn
visually guided, feedback and feedforward control strategies to compensate for the loss of
proprioceptive feedback and feedforward control signals. In severe cases, patients will collapse
the moment the lights are turned off, as they lose the visual feedback necessary to maintain
posture and are unable to make coordinated movements at all.

As we will encounter repeatedly, feedback and feedforward control are distributed among
the hierarchical levels of motor control. Many movements are the result of combinations of both
feedback and feedforward control.

11
 MOTOR UNITS AND MUSCLE RECEPTORS

The Spinal Cord: The First Hierarchical Level

The spinal cord is the first level of the motor hierarchy. It is the site where motor
neurons are located. It is also the site of many interneurons and complex neural circuits that
perform the “nuts and bolts” processing of motor control. These circuits execute the low-level
commands that generate the proper forces on individual muscles and muscle groups to enable
adaptive movements. Because this low level of the hierarchy takes care of these basic functions,
higher levels (such as the motor cortex) can process information related to the planning of
movements, the construction of adaptive sequences of movements, and the coordination of
whole-body movements, without having to encode the precise details of each muscle contraction.

Figure 1. Spinal cord with motor neuron in anterior horn

Motor Neurons

Alpha motor neurons (also called lower motor neurons) innervate skeletal muscle and
cause the muscle contractions that generate movement. Motor neurons release the
neurotransmitter acetylcholine at a synapse called the neuromuscular junction. When the
acetylcholine binds to acetylcholine receptors on the muscle fiber, an action potential is
propagated along the muscle fiber in both directions. The action potential triggers the
contraction of the muscle. If the ends of the muscle are fixed, keeping the muscle at the same
length, then the contraction results on an increased force on the supports (isometric contraction).
If the muscle shortens against no resistance, the contraction results in constant force (isotonic
contraction). The motor neurons that control limb and body movements are located in the
anterior horn of the spinal cord, and the motor neurons that control head and facial movements
are located in the motor nuclei of the brainstem. Even though the motor system is composed of
many different types of neurons scattered throughout the CNS, the motor neuron is the only way
in which the motor system can communicate with the muscles. Thus, all movements ultimately
depend on the activity of lower motor neurons. The famous physiologist Sir Charles Sherrington
referred to these motor neurons as the “final common pathway” in motor processing.
Motor neurons are not merely the conduits of motor commands generated from higher
levels of the hierarchy. They are themselves components of complex circuits that perform
sophisticated information processing. As shown in Figure 1, motor neurons have highly
branched, elaborate dendritic trees, enabling them to integrate the inputs from large numbers of
other neurons and to calculate proper outputs.

12
 Two terms are used to describe the anatomical relationship between motor neurons and
muscles: the motor neuron pool and the motor unit.

(1) Motor neurons are clustered in columnar, spinal nuclei called motor neuron pools (or
motor nuclei). All of the motor neurons in a motor neuron pool innervate a single muscle
(Figure 2), and all motor neurons that innervate a particular muscle are contained in the
same motor neuron pool. Thus, there is a one-to-one relationship between a muscle and a
motor neuron pool.

(2) Each individual muscle fiber in a muscle is innervated by one, and only one, motor
neuron. A single motor neuron, however, can innervate many muscle fibers. The
combination of an individual motor neuron and all of the muscle fibers that it innervates
is called a motor unit. The number of fibers innervated by a motor unit is called its
innervation ratio.

Figure 2. Motor
unit and motor
neuron pool

If a muscle is required for fine control or for delicate movements (e.g., movement of the
fingers or hands), its motor units will tend to have a small innervation ratio. That is, each motor
neuron will innervate a small number of muscle fibers (10-100), enabling many nuances of
movement of the entire muscle. If a muscle is required only for coarse movements (e.g., a thigh
muscle), its motor units will tend to have a high innervation ratio (i.e., each motor neuron
innervating 1000 or more muscle fibers), as there is no necessity for individual muscle fibers to
undergo highly coordinated, differential contractions to produce a fine movement.

Control of Muscle Force

A motor neuron controls the amount of force that is exerted by muscle fibers. There are
two principles that govern the relationship between motor neuron activity and muscle force: the
rate code and the size principle.
(1) Rate Code. Motor neurons use a rate code to signal the amount of force to be exerted by a
muscle. An increase in the rate of action potentials fired by the motor neuron causes an
increase in the amount of force that the motor unit generates. This code is illustrated in
Figure 3. When the motor neuron fires a single action potential, the muscle twitches slightly,
and then relaxes back to its resting state. If the motor neuron fires after the muscle has

13
 returned to baseline, then the magnitude of the next muscle twitch will be the same as the
first twitch. However, if the rate of firing of the motor neuron increases, such that a second
action potential occurs before the muscle has relaxed back to baseline, then the second action
potential produces a greater amount of force than the first (i.e., the strength of the muscle
contraction summates). With increasing firing rates, the summation grows stronger, up to a
limit. When the successive action potentials no longer produce a summation of muscle
contraction (because the muscle is at its maximum state of contraction), the muscle is in a
state called tetanus.

tetanus

Figure 3. Rate code

(2) Size Principle. When a signal is sent to the motor neurons to execute a movement, motor
neurons are not all recruited at the same time or at random. The motor neuron size principle
states that, with increasing strength of input onto motor neurons, smaller motor neurons are
recruited and fire action potentials before larger motor neurons are recruited. Why does this
orderly recruitment occur? Recall from previous lectures the relationship between voltage,
current, and resistance (Ohm’s Law): V = IR. Because smaller motor neurons have a smaller
membrane surface area, they have fewer ion channels, and therefore a larger input resistance.
Larger motor neurons have more membrane surface and correspondingly more ion channels;
therefore, they have a smaller input resistance. Because of Ohm’s Law, a small amount of
current will be sufficient to cause the membrane potential of a small motor neuron to reach
firing threshold, while the large motor neuron stays below threshold. As the amount of
current increases, the membrane potential of the larger motor neuron also increases, until it
also reaches firing threshold.

14
 Figure 4. Size principle

Figure 4 demonstrates how the size principle governs the amount of force generated by a
muscle. Because motor units are recruited in an orderly fashion, weak inputs onto motor
neurons will cause only a few motor units to be active, resulting in a small force exerted by
the muscle. With stronger inputs, more motor neurons will be recruited, resulting in more
force applied to the muscle. Moreover, different types of muscle fibers are innervated by
small and larger motor neurons. Small motor neurons innervate slow-twitch fibers;
intermediate-sized motor neurons innervate fast-twitch, fatigue-resistant fibers; and large
motor neurons innervate fast-twitch, fatigable muscle fibers. The slow-twitch fibers generate
less force than the fast-twitch fibers, but they are able to maintain these levels of force for
long periods. These fibers are used for maintaining posture and making other low-force
movements. Fast-twitch, fatigue-resistant fibers are recruited when the input onto motor
neurons is large enough to recruit intermediate-sized motor neurons. These fibers generate
more force than slow-twitch fibers, but they are not able to maintain the force as long as the
slow-twitch fibers. Finally, fast-twitch, fatigable fibers are recruited when the largest motor
neurons are activated. These fibers produce large amounts of force, but they fatigue very
quickly. They are used when the organism must generate a burst of large amounts of force,
such as in an escape mechanism. Most muscles contain both fast and slow-twitch fibers, but
in different proportions. Thus, the white meat of a chicken, used to control the wings, is
composed primarily of fast-twitch fibers, whereas the dark meat, used to maintain balance
and posture, is composed primarily of slow-twitch fibers.

Figure 5 demonstrates how the rate code principle and the size principle interact to signal
muscle force. In this classic experiment by Monster and Chan (1977), the firing rate of
individual motor neurons was measured as a function of the amount of force being generated by
a muscle. Each line on the graph corresponds to the firing rate of an individual motor neuron.
With small amounts of force, only a small number of motor neurons fire; these are the small
motor neurons that are recruited first. As the muscle generates increasing amounts of force,
more and more motor neurons fire; these additional neurons are the larger motor neurons. This
is the size principle. Notice that as each individual motor neuron fires more rapidly, it produces
a greater force on the muscle. This is the rate code principle.

15
 Figure 5. Interaction between
rate code and size principle in
determining muscle force.
Data from Monster AW &
Chan H (1977) Journal of
Neurophysiology 40:1432-
1443.

Muscle Receptors and Proprioception

As discussed in the previous lecture, the motor system requires sensory input in order to
function properly. In addition to sensory information about the external environment, the motor
system also requires sensory information about the current state of the muscles and limbs
themselves. Proprioception is the sense of the body’s position in space based on specialized
receptors that reside in the muscles and tendons. The muscle spindle signals the length of a
muscle and changes in the length of a muscle. The Golgi tendon organ signals the amount of
force being applied to a muscle.

Muscle Spindles

Muscle spindles are collections of 6-8 specialized muscle fibers that are located within
the muscle mass itself (Figure 6). These fibers do not contribute significantly to the force
generated by the muscle. Rather, they are specialized receptors that signal (a) the length and (b)
the rate of change of length (velocity) of the muscle. Because of the fusiform shape of the
muscle spindle, these fibers are referred to as intrafusal fibers. The large majority of muscle
fibers that actually contract and allow the muscle to do work are termed extrafusal fibers. Each
muscle contains many muscle spindles; muscles that are necessary for fine movements contain
more spindles than muscles that are used for posture or coarse movements.

Figure 6. Muscle spindle and Golgi tendon

organ

Intrafusal fibers

Extrafusal fibers

16
 Types of muscle spindle fibers. There are 3 types of muscle spindle fibers,
characterized by their shape and the type of information they convey (Figure 7).

(1) Nuclear Chain fibers. These fibers are so-named because their nuclei are aligned in
a single row (chain) in the center of the fiber. They signal information about the
static length of the muscle.
(2) Static Nuclear Bag fibers. These fibers are so-named because their nuclei are
collected in a bundle in the middle of the fiber. Like the nuclear chain fiber, these
fibers signal information about the static length of a muscle.
(3) Dynamic Nuclear Bag fibers. These fibers are anatomically similar to the static
nuclear bag fibers, but they signal primarily information about the rate of change
(velocity) of muscle length.

Figure 7. Muscle spindle detail

A typical muscle spindle is composed of 1 dynamic nuclear bag fiber, 1 static nuclear bag fiber,
and ~5 nuclear chain fibers.

Sensory innervation of muscle spindles. Because the muscle spindle is located in

parallel with the extrafusal fibers, it will stretch along with the muscle. The muscle spindle
signals muscle length and velocity to the CNS through two types of specialized sensory fibers
that innervate the intrafusal fibers. These sensory fibers have stretch receptors that open and
close as a function of the length of the intrafusal fiber.

(1) Group Ia afferents (also called primary afferents) wrap around the central portion of
all 3 types of intrafusal fibers; these specialized endings are called annulospiral
endings. Because they innervate all 3 types of intrafusal fibers, Group Ia afferents
provide information about both length and velocity.

(2) Group II afferents (secondary afferents) innervate the ends of the nuclear chain
fibers and the static nuclear bag fibers at specialized junctions termed flower spray
endings. Because they do not innervate the dynamic nuclear bag fibers, Group II
afferents signal information about muscle length only.

17
 Because of their patterns of innervation onto the three types of intrafusal fibers, Group Ia
and Group II afferents respond differently to different types of muscle movements. Figure 8
shows the responses of each type of afferent to a linear stretch of the muscle. Initially, both
Group Ia and Group II fibers fire at a certain rate, encoding the current length of the muscle.
During the stretch, the two types differ in their responses. The Group Ia afferent fires at a very
high rate during the stretch, encoding the velocity of the muscle length; at the end of the stretch,
its firing decreases, as the muscle is no longer changing length. Note, however, that its firing
rate is still higher than it was before the stretch, as it is now encoding the new length of the
muscle. Compare the response of the Group Ia afferent to the Group II afferent. The Group II
afferent increases its firing rate steadily as the muscle is stretched. Its firing rate does not depend
on the rate of change of the muscle; rather, its firing rate depends only on the immediate length
of the muscle.

Figure 8. Responses of muscle spindles

Gamma motor neurons. Although intrafusal fibers do not contribute significantly to

muscle contraction, they do have contractile elements at their ends that are innervated by motor
neurons. Motor neurons are divided into two groups. Alpha motor neurons innervate extrafusal
fibers, the contracting fibers that supply the muscle with its power. Gamma motor neurons
innervate intrafusal fibers, which contract only slightly. The function of intrafusal fiber
contraction is not to provide force to the muscle; rather, gamma activation of the intrafusal fiber
is necessary to keep the muscle spindle taut, and therefore sensitive to stretch, over a wide range
of muscle lengths. This concept is illustrated in Figure 9. If a resting muscle is stretched, the
muscle spindle becomes stretched in parallel, sending signals through the primary and secondary
afferents. A subsequent contraction of the muscle, however, removes the pull on the spindle, and
it becomes slack, causing the spindle afferents to cease firing. If the muscle were to be stretched
again, the muscle spindle would not be able to signal this stretch. Thus, the spindle is rendered
temporarily insensitive to stretch after the muscle has contracted. Activation of gamma motor
neurons prevents this temporary insensitivity by causing a weak contraction of the intrafusal
fibers, in parallel with the contraction of the muscle. This contraction keeps the spindle taut at
all times and maintains its sensitivity to changes in the length of the muscle. Thus, when the
CNS instructs a muscle to contract, it not only sends the appropriate signals to the alpha motor
neurons, it also instructs gamma motor neurons to contract the intrafusal fibers appropriately;
this coordinated process is referred to as alpha-gamma coactivation.

18
 Figure 9. Gamma activation of intrafusal
fibers. (A) Muscle is at a certain length,
encoded by firing of Ia afferent. (B) When
muscle is stretched, muscle spindle stretches
and Ia afferent fires more strongly. (C) When
muscle is contracted again, muscle spindle
become slack, causing Ia afferent to fall silent.
The muscle spindle is rendered insensitive to
further stretches of muscle. (D) To restore
sensitivity, firing of gamma motor neurons
causes spindle to contract, thereby becoming
taut and able to signal.

Golgi Tendon Organ

The Golgi tendon organ is a specialized receptor that is located between the muscle and
the tendon (Figure 6). Unlike the muscle spindle, which is located in parallel with extrafusal
fibers, the Golgi tendon organ is located in series with the muscle and signals information about
the load or force being applied to the muscle.

A Golgi tendon organ is made up of a capsule containing numerous collagen fibers

(Figure 10). The organ is innervated by primary afferents called Group Ib fibers, which have
specialized endings that weave in between the collagen fibers. When force is applied to a
muscle, the Golgi tendon organ is stretched, causing the collagen fibers to squeeze and distort the
membranes of the primary afferent sensory endings. As a result, the afferent is depolarized, and
it fires action potentials to signal the amount of force.

Group Ib afferent

Figure 10. Golgi tendon organ detail

19
In summary,

(1) Muscle spindles signal information about the length and velocity of a muscle
(2) Golgi tendon organs signal information about the load or force applied to a muscle

Functions of Muscle Spindles and Golgi Tendon Organs: An Introduction to Spinal

Reflexes

As noted in the previous lecture, a sense of body position is necessary for adaptive motor
control. In order to move a limb toward a particular location, it is imperative to know the initial
starting position of the limb, as well as any force applied to the limb. Muscle spindles and Golgi
tendon organs provide this type of information. In addition, these receptors are components of
certain spinal reflexes that are important for both clinical diagnoses as well as for a basic
understanding of the principles of motor control.

Myotatic reflex

The myotatic reflex is illustrated in Figure 11. A waiter is holding an empty tray, when
unexpectedly a pitcher of water is placed on the tray. Because the waiter’s muscles were not
prepared to support the increased weight, the tray should fall. However, a spinal reflex is
automatically initiated to keep the tray relatively stable. When the heavy pitcher is placed on the
tray, the increased weight stretches the biceps muscle, which results in the activation of the
muscle spindle’s Ia afferents. The Ia afferents have their cell bodies in the dorsal root ganglia of
the spinal cord, send projections into the spinal cord, and make synapses directly on alpha motor
neurons that innervate the same (homonymous) muscle. Thus, activation of the Ia afferent
causes a monosynaptic activation of the alpha motor neuron that causes the muscle to contract.
As a result, the stretch of the muscle is quickly counteracted, and the waiter is able to maintain
the tray at the same position.

Figure 11. Myotatic reflex

The myotatic reflex is an example of a feedback control system (discussed in the previous
lecture). A comparator (the spindle) compares the desired output (hold the limb steady at a

20
particular position) with the current state of the system. Initially, when the tray is steady, there is
no difference between the desired output and the current state. When the pitcher is placed on the
tray, however, the muscle spindle is stretched, and this causes the Ia afferent to send an error
signal that the muscle is stretched more than the desired output. The alpha motor neuron then
causes the effector (the muscle) to contract, thereby realigning the current state of the system
with the desired output.

A major role of the myotatic reflex is the maintenance of posture. Recall that feedback
control mechanisms are slow, and are typically not useful for rapid, voluntary movements. For
maintaining static posture, however, feedback mechanisms are very efficient and accurate. If
one is standing upright and starts to sway to the left, muscles in the legs and torso are stretched,
activating the myotatic reflex to counteract the sway. In this way, the higher levels of the motor
system are able to send a simple command (“maintain current posture”) and then be uninvolved
in its implementation. The lower levels of the hierarchy implement the command with such
mechanisms as the myotatic reflex, freeing the higher levels to perform other tasks such as
planning the next sequence of movements.

The myotatic reflex is an important clinical reflex. It is the same circuit that produces the
knee-jerk, or stretch, reflex. When the physician taps the patellar tendon with a hammer, this
action causes the knee extensor muscle to stretch momentarily. This stretch activates the
myotatic reflex, causing an extension of the lower leg. (Because the physician taps the tendon,
this reflex is also referred to as the deep tendon reflex. Do not be confused, however, between
this terminology and the Golgi tendon organ. The myotatic reflex is initiated by the muscle
spindle, not the Golgi tendon organ.) As we will learn in the next lecture, spinal reflexes can be
modulated by higher levels of the hierarchy, and thus a hyperactive or hypoactive stretch reflex
is an important clinical sign to localize neurological damage.

Autogenic inhibition

The Golgi tendon organ is involved in a spinal reflex known as the autogenic inhibition
reflex (Figure 12). When tension is applied to a muscle, the Group Ib fibers that innervate the
Golgi tendon organ are activated. These afferents have their cell bodies in the dorsal root
ganglia, and they project into the spinal cord and synapse onto an interneuron called the Ib
inhibitory interneuron. This interneuron makes an inhibitory synapse onto the alpha motor
neuron that innervates the same muscle that caused the Ib afferent to fire. As a result of this
reflex, activation of the Ib afferent causes the muscle to cease contraction, as the alpha motor
neuron becomes inhibited. Because this reflex contains an interneuron between the sensory
afferent and the motor neuron, it is an example of a disynaptic reflex.

21
 +
Ib inhibitory
–

Figure 12. Autogenic inhibition

For many years, it was thought that the function of the autogenic inhibition circuit was to
protect the muscle from excessive amounts of force that might damage it. A classic example is
that of the weightlifter straining to raise a heavy load, when suddenly the autogenic inhibition
reflex is activated and the muscle loses power, causing the weight to fall to the ground. This
function was ascribed to the reflex because early work suggested that the Golgi tendon organ
was only activated when large amounts of force were applied to it. More recent evidence
indicates, however, that the Golgi tendon organ is sensitive to much lower levels of force than
previously believed. This finding suggests that the autogenic inhibition reflex may be more
extensively involved in motor control under normal conditions. One possibility is that this reflex
helps to spread the amount of work evenly across the entire muscle, so that all motor units are
working efficiently. That is, if some muscle fibers are bearing more of the load than others, their
Golgi tendon organs will be more active, which will tend to inhibit the contraction of those
fibers. As a result, other muscle fibers that are less active will have to contract more to pick up
the slack, thereby sharing the work load more efficiently. This hypothesized function is another
example of a feedback control system.

22
 SPINAL RELEXES AND DESCENDING MOTOR PATHWAYS

Spinal reflexes

The end of the last lecture introduced two simple spinal reflexes that are initiated by
proprioceptors: the myotatic (stretch) reflex, which is initiated by the muscle spindle, and the
autogenic inhibition reflex, which is initiated by the Golgi tendon organ. In each case, activation
of a muscle receptor causes a change in the alpha motor neurons that innervate the same muscle.
The production of adaptive, coordinated behaviors requires further sophistication in these simple
reflexes. For example, stretching the muscle spindle not only leads to the contraction of the
homonymous muscle by the myotatic reflex, but also leads to the contraction of synergist
muscles by collateral pathways. The following are further examples of more sophisticated reflex
pathways.

Reciprocal inhibition in the stretch reflex. Joints are controlled by two opposing sets of
muscles, extensors and flexors, which must work in synchrony. Thus, when a muscle spindle is
stretched and activates the stretch reflex, the opposing muscle group must be inhibited to prevent
it from working against the resulting contraction of the homonymous muscle (Fig. 1). This
inhibition is accomplished by an inhibitory interneuron in the spinal cord. The Ia afferent of the
muscle spindle bifurcates in the spinal cord. One branch innervates the alpha motor neuron that
causes the homonymous muscle to contract, producing the behavioral reflex. The other branch
innervates the Ia inhibitory interneuron, which in turn innervates the alpha motor neuron that
synapses onto the opposing muscle. Because the interneuron is inhibitory, it prevents the
opposing alpha motor neuron from firing, thereby reducing the contraction of the opposing
muscle. Without this reciprocal inhibition, both groups of muscles might contract together and
work against each other.

+
–
+

Figure 1: Reciprocal inhibition in stretch reflex

Reciprocal excitation in the autogenic inhibition reflex. Just as in the stretch reflex,
the autogenic inhibition reflex (initiated by the Golgi tendon organ) must coordinate the activity
of the extensor and flexor muscle groups (Fig. 2). The Ib afferent fiber bifurcates in the spinal
cord. One branch innervates the Ib inhibitory interneuron, as described in the last lecture. The
other branch innervates an excitatory interneuron that, in turn, innervates the alpha motor neuron

23
that controls the antagonist muscle. Thus, when the homonymous muscle is inhibited from
contracting, the antagonist muscle is caused to contract, allowing the opposing muscle groups to
work in synchrony.

+
+
+
–

Figure 2: Reciprocal excitation in the autogenic inhibition reflex

Flexor reflex. Spinal reflexes can be initiated by nonproprioceptive receptors as well as

by proprioceptors. An important reflex initiated by cutaneous receptors and pain receptors is the
flexor reflex. We have all experienced this reflex after accidentally touching a hot stove or a
sharp object, as we withdraw our hand even before we consciously experience the sensation of
pain. This quick reflex removes the limb from the damaging stimulus more quickly than if the
pain signal had to travel up to the brain, be brought to conscious awareness, and then trigger a
decision to withdraw the limb. The reflex circuit is illustrated in Figure 3. A sharp object
touching the foot causes the activation of Group III afferents of pain receptors. These afferents
enter the spinal cord and then travel up the cord. A branch of the afferent innervates an
excitatory interneuron in the lumbar region of the spinal cord, which then excites an alpha motor
neuron that causes contraction of the thigh flexor muscle. The Group III afferent also continues
upward to the L2 vertebra, where another branch innervates an excitatory interneuron at this
level. This interneuron excites the alpha motor neurons that excite the hip flexor muscle,
allowing the coordinated activity of two muscle groups to withdraw the whole leg away from the
painful stimulus. Thus, spinal reflexes work not only at a single joint; they can also coordinate
the activity of multiple joints simultaneously.

24
 Figure 3: Flexor reflex

Reciprocal inhibition in the flexor reflex. When the knee joints and the hip joints are
flexed, the antagonist extensor muscles must be inhibited (just as in the stretch reflex). This is
accomplished by the Group III afferents innervating inhibitory interneurons that in turn innervate
the alpha motor neurons controlling the antagonist muscle.

Crossed extension reflex. Although the flexor reflex as described works well to
synchronize the activity of multiple muscle groups to allow the coordinated movement of the
entire limb, further circuitry is needed to make the reflex adaptive. Because the weight of the
body is supported by both legs, the flexor reflex must coordinate the activity not only of the leg
being withdrawn but also of the opposite leg (Fig. 4). Imagine stepping on a tack, and having the
flexor reflex withdraw your right leg immediately. The left leg must simultaneously extend in
order to support the body weight that would have been supported by the left leg. Without this
coordination of the two legs, the shift in body mass would cause a loss of balance. Thus, the
flexor reflex incorporates a crossed extension reflex. A branch of the Group III afferent
innervates an excitatory interneuron that sends its axon across the midline into the contralateral
spinal cord. There it excites the alpha motor neurons that innervate the extensor muscles of the
opposite leg, allowing balance and body posture to be maintained.

25
 Figure 4: Crossed extension reflex

Recurrent inhibition of motor neurons (Renshaw cells). Axons of alpha motor

neurons bifurcate in the spinal cord and innervate a special inhibitory interneuron called the
Renshaw cell (Fig. 5). This interneuron innervates and inhibits the very same motor neuron that
caused it to fire. Thus, a motor neuron regulates its own activity by inhibiting itself when it fires.
This negative feedback loop is thought to stabilize the firing rate of motor neurons.

+
–

Figure 5: Renshaw cell

26
Descending Motor Pathways

The spinal reflex circuits described above demonstrate that quite sophisticated and
precise neural processing occurs at the lowest level of the motor hierarchy. These automatic
reflexes can be modulated, however, by higher levels of the hierarchy. For example, when
touching an iron to see if it is still hot, your flexor reflex may be even more sensitive than
normal. As a result, you pull your hand away repeatedly before even touching the iron,
anticipating that it may be hot. Conversely, if you are removing a hot dish from the oven and the
heat starts to go through the oven mitt, you will suppress the flexor response so that you do not
drop your dinner all over the floor as you rush to put it down on a counter top. These
modulations (both facilitatory and inhibitory) of the spinal reflexes arise from the descending
pathways from the brainstem and cortex. Additionally, voluntary movement, as well as
vestibular-, visual-, and auditory-driven reflex actions, are controlled by the descending
pathways.

Descending motor pathways arise from multiple regions of the brain and send axons
down the spinal cord that innervate alpha motor neurons, gamma motor neurons, and
interneurons. The motor neurons are topographically organized in the anterior horn of the spinal
cord according to two rules: the flexor-extensor rule and the proximal-distal rule.

Flexor-extensor rule: motor neurons that innervate flexor muscles are located posteriorly to
motor neurons that innervate extensor muscles.

Proximal-distal rule: motor neurons that innervate distal muscles (e.g., hand muscles) are
located lateral to motor neurons that innervate proximal muscles (e.g., trunk muscles).

Figure 6: Flexor-extensor/proximal-distal rules

Descending motor pathways are organized into two major groups:

(1) Lateral pathways control both proximal and distal muscles and are responsible for most
voluntary movements of arms and legs. They include the
(a) lateral corticospinal tract
(b) rubrospinal tract

27
(2) Medial pathways control axial muscles and are responsible for posture, balance, and coarse
control of axial and proximal muscles. They include the
(a) vestibulospinal tracts (both lateral and medial)
(b) reticulospinal tracts (both pontine and medullary)
(c) tectospinal tract
(d) anterior corticospinal tract

Corticospinal tracts. The corticospinal tract originates in the motor cortex (Fig. 7). The
axons of motor projection neurons collect in the internal capsule, and then course through the
crus cerebri (cerebral peduncle) in the midbrain. At the level of the medulla, these axons form
the medullary pyramids on the ventral surface of the brainstem (hence, this tract is also called the
pyramidal tract). At the level of the caudal medulla, the corticospinal tract splits into two tracts.
Approximately 90% of the axons cross over to the contralateral side at the pyramidal
decussation, forming the lateral corticospinal tract. These axons continue to course through the
lateral funiculus of the spinal cord, before synapsing either directly onto alpha motor neurons or
onto interneurons in the ventral horn. The remaining 10% of the axons that do not cross at the
caudal medulla constitute the anterior corticospinal tract, as they continue down the spinal cord
in the anterior funiculus. When they reach the spinal segment at which they terminate, they cross
over to the contralateral side through the anterior white commissure and innervate alpha motor
neurons or interneurons in the anterior horn. Thus, both the lateral and anterior corticospinal
tracts are crossed pathways; they cross the midline at different locations, however.

Function. The corticospinal tract

(along with the corticobulbar tract) is the
primary pathway that carries the motor
commands that underlie voluntary
movement. The lateral corticospinal tract is
responsible for the control of the distal
musculature and the anterior corticospinal
tract is responsible for the control of the
proximal musculature. A particularly
important function of the lateral
corticospinal tract is the fine control of the
digits of the hand. The corticospinal tract is
the only descending pathway in which some
axons make synaptic contacts directly onto
alpha motor neurons. This direct cortical
innervation presumably is necessary to
allow the powerful processing networks of
the cortex to control the activity of the
spinal circuits that direct the exquisite
movements of the fingers and hands. The
percentage of axons in the corticospinal tract
that innervate alpha motor neurons directly
is greater in humans and nonhuman
Figure 7: Corticospinal tracts

28
 primates than in other mammals, presumably reflecting the increased manual dexterity of
primates. Damage to the corticospinal tract results in a permanent loss of the fine control of the
extremities. Although parallel descending pathways can often recover the function of more
coarse movements, these pathways are not capable of generating fine, skilled movements. In
addition to the fine control of distal muscles, the corticospinal tract also plays a role in the
voluntary control of axial muscles.

Rubrospinal tract. The rubrospinal tract

originates in the red nucleus of the midbrain (Fig.
8). The axons immediately cross to the
contralateral side of the brain, and they course
through the brainstem and the lateral funiculus of
the spinal cord. The axons innervate spinal
neurons at all levels of the spinal cord.

Function. The rubrospinal tract is an alternative

Figure 8: Rubrospinal tract
by which voluntary motor commands can be sent
to the spinal cord. Although it is a major pathway
in many animals, it is relatively minor in humans.
Activation of this tract causes excitation of flexor
muscles and inhibition of extensor muscles. The
rubrospinal tract is thought to play a role in
movement velocity, as rubrospinal lesions cause a
temporary slowness in movement. In addition,
because the red nucleus receives most of its input
from the cerebellum, the rubrospinal tract probably
plays a role in transmitting learned motor
commands from the cerebellum to the
musculature. The red nucleus also receives some
input from the motor cortex, and it is therefore
probably an important pathway for the recovery of
some voluntary motor function after damage to the
corticospinal tract.

Vestibulospinal tracts. The two vestibulospinal tracts originate in 2 of the 4 vestibular

nuclei (Fig. 9). The lateral vestibulospinal tract originates in the lateral vestibular nucleus. It
courses through the brainstem and through the anterior funiculus of the spinal cord on the
ipsilateral side, before exiting ipsilaterally at all levels of the spinal cord. The medial
vestibulospinal tract originates in the medial vestibular nucleus, splits immediately and courses
bilaterally through the brainstem via the medial longitudinal fasciculus and through the anterior
funiculus of the spinal cord, before exiting at or above the T6 vertebra.

Function. The vestibulospinal tracts mediate postural adjustments and head movements.
They also help the body to maintain balance. Small movements of the body are detected by the
vestibular sensory neurons, and motor commands to counteract these movements are sent

29

t
 through the vestibulospinal tracts to appropriate
muscle groups throughout the body. The lateral
vestibulospinal tract excites antigravity muscles
in order to exert control over postural changes
necessary to compensate for tilts and
movements of the body. The medial
vestibulospinal tract innervates neck muscles in
order to stabilize head position as one moves
around the world. It is also important for the
coordination of head and eye movements.

Figure 9: Vestibulospinal tracts

Reticulospinal tracts. The two reticulospinal tracts

originate in the brainstem reticular formation, a large,
diffusely organized collection of neurons in the pons and
medulla (Fig. 10). The pontine reticulospinal tract
originates in the pontine reticular formation, courses
ipsilaterally through the medial longitudinal fasciculus and
through the anterior funiculus of the spinal cord, and exits
ipsilaterally at all spinal levels. The medullary
reticulospinal tract originates in the medullary reticular
formation, courses mainly ipsilaterally (although some
fibers cross the midline) through the anterior funiculus of
the spinal cord, and exits at all spinal levels.

Function. The reticulospinal tracts are a major

alternative to the corticospinal tract, by which cortical
neurons can control motor function by their inputs onto
reticular neurons. These tracts regulate the sensitivity of
Figure 10: Reticulospinal tracts

flexor responses to ensure that only noxious stimuli elicit the responses. Damage to the
reticulospinal tract can thus cause harmless stimuli, such as gentle touches, to elicit a flexor

30
reflex. The reticular formation also contains circuitry for many complex actions, such as
orienting, stretching, and maintaining a complex posture. Commands that initiate locomotor
circuits in the spinal cord are also thought to be transmitted through the medullary reticulospinal
tract. Thus, the reticulospinal tracts are involved in many aspects of motor control, including the
integration of sensory input to guide motor output.

Tectospinal tract. The tectospinal tract originates in

the deep layers of the superior colliculus and crosses the
midline immediately. It then courses through the pons and
medulla, just anterior to the medial longitudinal fasciculus. It
courses through the anterior funiculus of the spinal cord,
where the majority of the fibers terminate in the upper cervical
levels.

Function. Little is known about the function of the

tectospinal tract, but because of the nature of the visual
response properties of neurons in the superior colliculus (the
optic tectum), it is presumably involved in the reflexive
turning of the head to orient to visual stimuli.

Figure 11: Tectospinal tract

Influences of descending pathways on spinal circuits

Voluntary movement. The most distinctive function of the descending motor pathways
is the control of voluntary movement. These movements are initiated in the cerebral cortex, and
the motor commands are transmitted to the musculature through a variety of descending
pathways, including the corticospinal tract, the rubrospinal tract, and reticulospinal tracts. How
voluntary movements are initiated and coordinated by the motor cortex is the subject of the next
lecture.

Reflex modulation. Another critical function of the descending motor pathways is to

modulate the reflex circuits in the spinal cord. The adaptiveness of spinal reflexes can change
depending on the behavioral context; sometimes the gain (strength) or even the sign (extension
vs. flexion) of a reflex must be changed in order to make the resulting movement adaptive. The
descending pathways are responsible for controlling these variables. For example, consider the
flexor reflex under two conditions.

31
 (a) Imagine a situation in which you want to pick up a dish from the stove top, but you
are uncertain whether it is hot or cold. You may attempt to lightly touch the surface, and this
will often lower the threshold of the flexor reflex, making you more likely to pull your hand
away even if the dish is not particularly hot. (You may even withdraw your hand numerous
times before even touching the dish!) Descending pathways have lowered the threshold for
producing the reflex in this case, making it easier for a weaker nociceptive input to trigger the
reflex; these pathways can also change the gain of the reflex, making the withdrawal response
greater than usual.

(b) Imagine now picking up the dish in order to move it to the table. As you hold the dish,
more of its heat begins to transfer to your hand, and it starts to get quite hot. Rather than
dropping the dish and spilling your dinner all over the floor, you rush to the table to put it down,
before withdrawing your hand and wishing you had used an oven mitt. In this case, the
descending pathways inhibited the flexor response.

Gamma bias. Recall from the previous lecture that there are two types of spinal motor
neurons. Alpha motor neurons innervate extrafusal muscle fibers, which provide the force for a
muscle contraction. Gamma motor neurons innervate the ends of intrafusal fibers and help to
maintain the tautness of muscle spindles, such that they are sensitive to changes of muscle length
over a wide range. In order to work adaptively, the activity of alpha and gamma motor neurons
must be coordinated. Thus, whenever motor commands are sent by descending pathways to
alpha motor neurons, the appropriate compensating commands are sent to gamma motor neurons.
This coordination of alpha-gamma motor commands is called alpha-gamma coactivation, and the
adjustment of spindle sensitivity by gamma activation is called gamma bias. Consider the
following two examples:

(a) When a command is given to a muscle to contract, the muscle spindles become slack,
thereby making them insensitive to further changes in muscle length. To compensate for this,
the gamma motor neurons that innervate these intrafusal muscle fibers are activated in concert
with the alpha motor neurons, allowing the intrafusal fibers to contract with the muscle. This
preserves the sensitivity of the muscle to unexpected stretches of the muscle.

(b) When a muscle contracts, the antagonist muscle is stretched during the movement.
An obvious problem arises when one considers the stretch reflex of the antagonist muscle. If
contraction of a muscle causes the activation of the stretch reflex of the antagonist muscle, the
antagonist muscle will contract to resist the movement of the limb. How is it possible to ever
flex a joint when the stretch reflex of the extensor muscle causes it to extend the joint instead?
Alpha-gamma coactivation solves this problem by relaxing the contraction of the intrafusal fibers
of the antagonist muscle, allowing the muscle to be stretched without triggering the stretch reflex
during a voluntary movement.

32
 CEREBELLUM

Overview: Functions of the cerebellum

The cerebellum (“little brain”) is a structure that is located at the back of the brain,
underlying the occipital and temporal lobes of the cerebral cortex (Fig. 1). Although the
cerebellum accounts for approximately 10% of the brain’s volume, it contains over 50% of the
total number of neurons in the brain. Historically, the cerebellum has been considered a motor
structure, because cerebellar damage leads to impairments in motor control and posture and
because the majority of the cerebellum’s outputs are to parts of the motor system. Motor
commands are not initiated in the cerebellum; rather, the cerebellum modifies the motor
commands of the descending pathways to make movements more adaptive and accurate. The
cerebellum is involved in the following functions:

Figure 1. Cerebellum

Maintenance of balance and posture. The cerebellum is important for making postural
adjustments in order to maintain balance. Through its input from vestibular receptors and
proprioceptors, it modulates commands to motor neurons to compensate for shifts in body
position or changes in load upon muscles. Patients with cerebellar damage suffer from balance
disorders, and they often develop stereotyped postural strategies to compensate for this problem
(e.g., a wide-based stance).

Coordination of voluntary movements. Most movements are composed of a number of

different muscle groups acting together in a temporally coordinated fashion. One major function
of the cerebellum is to coordinate the timing and force of these different muscle groups to
produce fluid limb or body movements.

Motor learning. The cerebellum is important for motor learning. The cerebellum plays a
major role in adapting and fine-tuning motor programs to make accurate movements through a
trial-and-error process (e.g., learning to hit a baseball).

Cognitive functions. Although the cerebellum is most understood in terms of its

contributions to motor control, recent studies have revealed that it is also involved in certain
cognitive functions, such as language. Thus, like the basal ganglia, the cerebellum is historically

33
considered as part of the motor system, but its functions extend beyond motor control in ways
that are not well understood.

Cerebellar gross anatomy

The cerebellum consists of two major parts (Fig. 2A). The cerebellar deep nuclei (or
cerebellar nuclei) are the sole output structures of the cerebellum. These nuclei are encased by a
highly convoluted sheet of tissue called the cerebellar cortex, which contains almost all of the
neurons in the cerebellum. A cross-section through the cerebellum reveals the intricate pattern
of folds and fissures that characterize the cerebellar cortex. Like the cerebral cortex, cerebellar
gyri are reproducible across individuals and have been identified and named. We will only be
concerned with some of the larger divisions of the cerebellar cortex.

Figure 2. (A) Cerebellar deep nuclei and cerebellar cortex in an idealized brain section. (B)
External morphology of the cerebellum

Divisions of the cerebellum. Two major fissures running mediolaterally divide the
cerebellar cortex into 3 primary subdivisions (Fig. 2B, Fig. 3). The posterolateral fissure
separates the flocculonodular lobe from the corpus cerebelli, and the primary fissure separates
the corpus cerebelli into a posterior lobe and an anterior lobe. The cerebellum is also divided
sagittally into 3 zones that run from medial to lateral (Figure 3). The vermis (from the Latin
word for worm) is located along the midsagittal plane of the cerebellum. Directly lateral to the
vermis is the intermediate zone. Finally, the lateral hemispheres are located lateral to the
intermediate zone (there are no clear morphological borders between the intermediate zone and
the lateral hemisphere that are visible from a gross specimen).

Cerebellar nuclei. All output from the cerebellum originates from the cerebellar deep
nuclei. Thus, a lesion to the cerebellar nuclei has the same effect as a complete lesion of the
entire cerebellum. It is important to know the inputs, outputs, and anatomical relationships
between the different cerebellar nuclei and the subdivisions of the cerebellum (Fig. 4).

34
 Figure 3. Divisions of cerebellum

(1) The fastigial nucleus is the most medially located of the cerebellar nuclei. It receives
input from the vermis and from cerebellar afferents that carry vestibular, proximal
somatosensory, auditory, and visual information. It projects to the vestibular nuclei and the
reticular formation.

(2) The interposed nuclei comprise the emboliform nucleus and the globose nucleus.
They are situated lateral to the fastigial nucleus. They receive input from the intermediate zone
and from cerebellar afferents that carry spinal, proximal somatosensory, auditory, and visual
information. They project to the contralateral red nucleus (the origin of the rubrospinal tract).

(3) The dentate nucleus is the largest of the cerebellar nuclei, located lateral to the
interposed nuclei. It receives input from the lateral hemisphere and from cerebellar afferents that
carry information from the cerebral cortex (via the pontine nuclei). It projects to the contralateral
red nucleus and the ventrolateral (VL) thalamic nucleus.

(4) The vestibular nuclei are located outside the cerebellum, in the medulla. Hence, they
are not strictly cerebellar nuclei, but they are considered to be functionally equivalent to the
cerebellar nuclei because their connectivity patterns are identical to the cerebellar nuclei. The
vestibular nuclei receive input from the flocculonodular lobe and from the vestibular labyrinth.
They project to various motor nuclei and originate the vestibulospinal tracts.

In addition to these inputs, all cerebellar nuclei and all regions of cerebellum get special
inputs from the inferior olive of the medulla (discussed below).

35
 It is convenient to remember that the anatomical locations of the cerebellar nuclei
correspond to the cerebellar cortex regions from which they receive input. Thus, the medially
located fastigial nucleus receives input from the medially located vermis; the slightly lateral
interposed nuclei receive input from the slightly lateral intermediate zone; and the most lateral
dentate nucleus receives input from the lateral hemispheres.

Figure 4. Inputs/outputs of cerebellum

36
 Cerebellar peduncles. Three major bundles of fibers carry the input and output of the
cerebellum.

(1) The inferior cerebellar peduncle (also called the restiform body) primarily contains
afferent fibers from the medulla, as well as efferents to the vestibular nuclei.

(2) The middle cerebellar peduncle (also called the brachium pontis) primarily contains
afferents from the pontine nuclei.

(3) The superior cerebellar peduncle (also called the brachium conjunctivum) primarily
contains efferent fibers from the cerebellar nuclei, as well as some afferents from the
spinocerebellar tract.

Thus, the inputs to the cerebellum are conveyed primarily through the inferior and middle
cerebellar peduncles whereas the outputs are conveyed primarily through the superior cerebellar
peduncle. The inputs arise from the ipsilateral side of the body, and the outputs also go to the
ipsilateral side of the body. Note that this is true even for the outputs to the contralateral red
nucleus. Recall from the lecture on descending motor pathways that the rubrospinal tract
immediately crosses the midline after the fibers leave the red nucleus. Thus, cerebellar output to
the red nucleus affects the ipsilateral side of the body by a double-crossed pathway. Unlike the
cerebral cortex, the cerebellum receives input from, and controls output to, the ipsilateral side of
the body, and damage to the cerebellum therefore results in deficits to the ipsilateral side of the
body.

Functional subdivisions of the cerebellum

The anatomical subdivisions described above correspond to three major functional

subdivisions of the cerebellum.

(1) The vestibulocerebellum comprises the flocculonodular lobe and its connections
with the lateral vestibular nuclei. Phylogenetically, the vestibulocerebellum is the oldest part of
the cerebellum. As its name implies, it is involved in vestibular reflexes (such as the
vestibuloocular reflex; see below) and in postural maintenance.

(2) The spinocerebellum comprises the vermis and the intermediate zones of the
cerebellar cortex, as well as the fastigial and interposed nuclei. As its name implies, it receives
major inputs from the spinocerebellar tract. Its output projects to rubrospinal, vestibulospinal,
and reticulospinal tracts. It is involved in the integration of sensory input with motor commands
to produce adaptive motor coordination.

(3) The cerebrocerebellum is the largest functional subdivision of the human

cerebellum, comprising the lateral hemispheres and the dentate nuclei. Its name derives from its
extensive connections with the cerebral cortex, via the pontine nuclei (afferents) and the VL
thalamus (efferents). It is involved in the planning and timing of movements. In addition, the
cerebrocerebellum is involved in the recently discovered, yet poorly understood, cognitive
functions of the cerebellum.

37
Histology and connectivity of cerebellar cortex

The cerebellar cortex is divided into 3 layers (Fig. 5). The innermost layer, the granule
cell layer, is made of 5 x 1010 small, tightly packed granule cells. The middle layer, the Purkinje
cell layer, is only 1-cell thick. The outer layer, the molecular layer, is made of the axons of
granule cells and the dendrites of Purkinje cells, as well as a few other cell types. The Purkinje
cell layer forms the border between the granule and molecular layers.

Figure 5. Cerebellar circuitry. This basic pattern is repeated throughout all regions of the cerebellum.

Granule cells. Granule cells are very small, densely packed neurons that account for the
huge majority of neurons in the cerebellum. Indeed, cerebellar granule cells account for more
than half of the neurons in the entire brain. These cells receive input from mossy fibers and
project to the Purkinje cells.

Purkinje cells. The Purkinje cell is one of the most striking cell types in the mammalian
brain. Its apical dendrites form a large fan of finely branched processes. Remarkably, this
dendritic tree is almost two-dimensional; looked at from the side, the dendritic tree is flat and
almost dimensionless. Moreover, all Purkinje cells are oriented in parallel. This arrangement
has important functional considerations, as we shall see below.

Other cell types. In addition to the major cell types (granule cells and Purkinje cells),
the cerebellar cortex also contains various interneuron types, including the Golgi cell, the basket
cell, and the stellate cell.

Connectivity. The cerebellar cortex has a relatively simple, stereotyped connectivity

pattern that is identical throughout the whole structure. Figure 5 illustrates a simplified diagram
of the connectivity of the cerebellum. Cerebellar input can be divided into two distinct classes.

38
 (1) Mossy fibers originate in the pontine nuclei, the spinal cord, the brainstem reticular
formation, and the vestibular nuclei, and they make excitatory projections onto the cerebellar
nuclei and onto granule cells in the cerebellar cortex. They are called mossy fibers because of
the tufted appearance of their synaptic contacts with granule cells. There is a large degree of
divergence in the mossy fiber-granule cell connection, as each mossy fiber innervates hundreds
of granule cells. The granule cells send axons up toward the cortical surface. Each axon
bifurcates in the molecular layer, sending a collateral in opposite directions. These fibers, called
parallel fibers, run parallel to the folds of the cerebellar cortex, where they make excitatory
synapses with Purkinje cells along the way. The two-dimensional arbors of the Purkinje cell
dendrites are oriented perpendicular to the parallel fibers. Thus, the arrangement of Purkinje
cells and parallel fibers resembles telephone lines running between telephone poles. Each
parallel fiber makes contact with hundreds of Purkinje cells; because of the high degree of
divergence of the mossy fiber-granule cell synapses, the firing of each Purkinje cell can be
influenced (disynaptically) by thousands of mossy fibers.

(2) Climbing fibers originate exclusively in the inferior olive and make excitatory
projections onto the cerebellar nuclei and onto the Purkinje cells of the cerebellar cortex. They
are called climbing fibers because their axons climb and wrap around the dendrites of the
Purkinje cell like a climbing vine. Each Purkinje cell receives a single, extremely powerful input
from a single climbing fiber. In contrast to mossy fibers and parallel fibers, each climbing fiber
contacts only 10 Purkinje cells on average, making ~300 synapses with each Purkinje cell. Thus,
the climbing fiber is a restricted, but extremely powerful, excitatory input onto Purkinje cells.

The Purkinje cell is the sole source of output from the cerebellar cortex. It is
important to note that Purkinje cells make inhibitory connections onto the cerebellar nuclei.
(Note the distinction between the Purkinje cells, which constitute the sole output of the cerebellar
cortex, and the cerebellar nuclei, which constitute the sole output of the entire cerebellum.)
Almost all of the spikes generated by the Purkinje cell are caused by its parallel-fiber inputs.
These inputs cause the Purkinje cell to fire at a high resting rate (~70 spikes/sec), tonically
inhibiting its cerebellar nucleus targets. The powerful inputs from climbing fibers occur less
frequently (~1 spike/sec); thus, they have a minor influence on the overall firing rate of the
Purkinje cell. The Purkinje cell spikes that are generated by climbing fibers are calcium-spikes,
however, which allow the climbing fibers to initiate a number of calcium-dependent changes in
the Purkinje cell. As described below, one important change appears to be a long-lasting change
in the strength of the parallel-fiber inputs to the Purkinje cell.

Damage to cerebellum produces movement disorders

Much of what is known about cerebellar function comes from studies of patients with
cerebellar damage. In general, such patients display uncoordinated voluntary movements and
problems maintaining balance and posture. The following are some symptoms of cerebellar
damage:

(1) Decomposition of movement. Most of our movements involve the coordinated

activity of many muscle groups and different joints to produce a smooth trajectory of the body
part through space. Patients with cerebellar dysfunction are unable to produce these coordinated,

39
smooth movements. Instead, they often break the movements down into their component parts
in order to execute the desired trajectory. For example, touching one’s finger to one’s nose
requires the coordinated activity of shoulder, elbow, and wrist joints. Cerebellar patients must
first perform the shoulder movement, then the elbow movement, and finally the wrist movement
in sequence, rather than as one, uniform motion.

(2) Intention tremor. When making a movement to a target, cerebellar patients often
produce an involuntary tremor that increases as they approach closer to the target. For example,
if reaching for a cup, the hand starts out in a direct line toward the cup; as it gets closer, however,
the hand begins to move back and forth as it attempts to make contact with the cup.

(3) Dysdiadochokinesia. Patients have difficulty performing rapidly alternating

movements, such as hitting a surface rapidly and repeatedly with the palm and back of the hand.

(4) Deficits in motor learning. Experimental studies have demonstrated that cerebellar
damage causes deficits in motor learning in both human patients and experimental animals. One
prominent experimental model is the vestibuloocular reflex (VOR). This reflex allows us to
maintain gaze on an object when the head is rotated. Vestibular signals detect the head
movement, and send signals through the cerebellum to the eye muscles to precisely counter the
head rotation and maintain a stable center of gaze. The motor commands to the eyes must be
calibrated precisely with experience, and this calibration appears to be the job of the cerebellum.
Experiments have been performed in which subjects wore prisms that reversed the visual field.
When the subjects’ heads were moved, the VOR actually caused the visual image to rotate on the
retina rather than remaining stable. Over days, however, the VOR slowly adjusted, such that the
proper compensatory eye movements were made to keep the retinal image stable when the head
was rotated. In experimental animals, lesions to the cerebellum prevent this adjustment of the
VOR.

A second example of cerebellum-dependent motor learning involves the execution of

accurate, coordinated movements. Subjects wore prism goggles that shifted the visual image to
the right, and they were asked to then throw darts at a target. Because of the prisms, the
accuracy of the subjects was initially quite low, as the darts consistently hit to the left of the
target. With repeated practice, however, the subjects became more and more accurate at hitting
the target. When the goggles were removed, the subject now began to throw the darts to the right
of the target, because their motor programs had been recalibrated to use the shifted visual input.
Over time, once again, they gradually increased their accuracy. Patients with cerebellar damage
never learned to compensate for the prism, as their darts always landed to the left of the target
when the goggles were worn. When the goggles were removed, they were immediately accurate
at hitting the target, because they never made compensations for the earlier prism trials.

A third example involves the Pavlovian classical conditioning of the eye blink reflex. In
this task, a neutral stimulus (such as a tone) is paired with a noxious stimulus (such as a puff of
air to the eye) that causes a reflexive eye blink. Over time, experimental animals will learn to
close their eye when the tone occurs, in anticipation of the air puff. This learned eyelid closure is
remarkably well-timed to peak at the expected time of the puff. Animals with cerebellar damage
do not learn to produce the eyelid closure in response to the tone.

40
Cerebellum as feedforward and feedback controller

What do the various symptoms of cerebellar damage have in common that reveal the
function of the cerebellum? A number of different theories have been proposed. Recall the
discussion in Chapter 1 of feedback and feedforward controllers, and the ubiquitous use of
sensory information in motor control. The cerebellum appears to be acting as a feedback and
feedforward control system for motor control. In some circumstances, the cerebellum appears to
integrate ongoing sensory information in order to correct for errors in movements during the
movement (feedback control). In other situations, the cerebellum appears to act as a feedforward
controller. Recall that feedforward control mechanisms can act very fast, but that they require
learning. This requirement may explain the role of the cerebellum in motor learning.

For example, the cerebellar involvement in the VOR may be explained in terms of the
learning requirements of a feedforward controller. When the head moves, a compensatory eye
movement must be made to maintain a stable gaze. The cerebellum receives sensory input from
the vestibular system informing it that the head is moving. It also receives input from eye
muscle proprioceptors and other relevant sources of information about current conditions in
order to make an accurate compensatory eye movement. It evaluates all of this advance sensory
information and calculates the proper eye movement to exactly counterbalance the head
movement. What if the eye movement does not match the head movement, however, and the
visual image moves across the retina (such as in the experimental condition in which a prism was
worn, or in a real-life situation in which an individual wears new prescription eyeglasses)? The
retinal slip constitutes an error signal to tell the cerebellum that next time these conditions are
met, adjust the eye movement to decrease the retinal slip. This trial and error sequence will be
repeated until the movement is properly calibrated; moreover, these mechanisms will ensure that
the movements stay calibrated.

As another example, the coordination of movements requires that muscle groups be

activated in precise temporal sequence. Not only do the different joints need to be coordinated
temporally, but even antagonist muscles that control the same joint need precise temporal
coordination. For example, an extensor muscle needs to be activated to start a reaching
movement, and the corresponding flexor muscle needs to be activated at the end of the
movement to stop the movement appropriately. The precise timing of muscle contractions and
the force necessary for each contraction varies with the amount of load placed on a muscle, as
well as on the inherent properties of the muscle itself (e.g., elasticity). These variables are
constantly changing throughout life, as one grows, gains/loses weight, and ages. Moreover, a
similar movement will require different patterns of motor activity depending on the weight being
born by the muscle (for example, if an extended hand is empty or holding a heavy weight). The
cerebellum appears necessary for the proper timing and coordination of muscle groups, very
likely through a trial-and-error learning mechanism discussed previously. Such a role helps
explain the deficits seen in dysdiadochokinesia, in which patients cannot perform rapidly
alternating sequences of movements.

It is believed that the mossy fiber inputs to the cerebellum convey the sensory
information used to evaluate the overall “context” of the movement. Mossy fibers are known to

41
respond to sensory stimuli; they are also correlated with different movements (Fig. 6). These
fibers convey such information as: Where are the appropriate body parts (proprioceptors), what
is the current load on the muscle (proprioceptors, somatosensory receptors, etc.), what other
sensory information can predict a useful response (e.g., the tone in the eye blink conditioning),
what are the desired movements (motor cortex). The error signal is believed to be conveyed by
the climbing fiber inputs. Climbing fibers are known to be especially active when an unexpected
event occurs, such as when a greater load than expected is placed on a muscle or when a toe is
stubbed. Thus, the large divergence of input from the mossy fibers to the granule cells to the
parallel fibers is believed to create complex representations of the entire sensory context at
present and the desired motor output. When the desired output is not achieved, the climbing
fibers signal this error and trigger a calcium spike in the Purkinje cell. The influx of calcium
changes the connection strengths between parallel fibers and Purkinje cells, such that the next
time the same behavioral context occurs, the motor output will be modified to more closely
approximate the desired output. (As an exercise, review the example of feedforward control of
temperature in Chapter 1 and compare the components of that circuit with the feedforward
control of movement by the cerebellum.)

Figure 6. Cerebellum as a feedforward controller

42
 BASAL GANGLIA

The previous 4 lectures have described the anatomy and function of the 4 levels of the
motor system hierarchy: the spinal cord, the brainstem, the motor cortex, and the association
cortex. Two other brain structures can be considered as “side loops” in the motor hierarchy.
They influence the processing of motor control and modulate the output of the descending
pathways without directly causing motor output. Both of these structures—the basal ganglia and
the cerebellum—are now known to have other functions in addition to modulating motor control.
Because the most obvious clinical signs of damage to these areas are a wide variety of motor
impairments, they are still generally considered to be motor structures. Basal ganglia
dysfunction causes a set of symptoms that are quite different from damage to descending motor
pathways, and thus the basal ganglia were at one time considered to form an “extrapyramidal
motor system” that was distinct from the pyramidal tract pathways. It is now known that the
basal ganglia do not originate a separate motor pathway. Instead, they influence and modulate
the activity of motor cortex and the descending motor pathways in ways that cause distinct
symptoms when different basal ganglia structures are damaged.

Gross anatomy of the basal ganglia

The basal ganglia comprise a distributed set of brain structures in the telencephalon,
diencephalon, and mesencephalon (Fig. 1 and 2). The forebrain structures include the caudate
nucleus, the putamen, the nucleus accumbens (or ventral striatum) and the globus pallidus.
Together, these structures are named the corpus striatum. The caudate nucleus is a C-shaped
structure that is closely associated with the lateral wall of the lateral ventricle. It is largest at its
anterior pole (the head), and its size diminishes posteriorly as it follows the course of the lateral
ventricle (the body) all the way to the temporal lobe (the tail), where it terminates at the
amygdaloid nuclei. The putamen is also a large structure that is separated from the caudate
nucleus by the anterior limb of the internal capsule. The putamen is connected to the caudate
head by bridges of cells that cut across the internal capsule. Because of the striated appearance
of these cell bridges (Fig. 1b), the caudate and putamen are collectively referred to as the
striatum or neostriatum, and the nucleus accumbens is often called the ventral striatum.
Functionally, the caudate nucleus and the putamen are considered equivalent to each other;
indeed, most mammals have only a single nucleus called the striatum. It is unclear whether there
is any functional significance of the separation of the striatum into the caudate and putamen in
primates. The putamen and the globus pallidus are collectively called the lenticular nucleus, or
lentiform nucleus. The globus pallidus is divided into two segments: the internal (or medial)
segment and the external (or lateral) segment.

43
 Figure 1. Basal ganglia nuclei. (A) Location of basal ganglia components in idealized brain section. (B) Cell
bridges between the caudate and putamen give a striated appearance.

Basal Ganglia

Nucleus accumbens
Corpus
Caudate Striatum or Neostriatum
striatum
Lenticular Putamen
nucleus Globus pallidus
Subthalamic nucleus
Substantia nigra

Figure 2. Basal ganglia nomenclature

The subthalamic nucleus is part of the diencephalon; as its name implies, it is located just
below the thalamus. The substantia nigra is a midbrain structure, composed of two distinct parts:
the pars compacta and the pars reticulata. The substantia nigra is located between the red
nucleus and the crus cerebri (cerebral peduncle) on the ventral part of the midbrain. The pars
compacta is the source of a clinically important dopaminergic pathway to the striatum; loss of
neurons in this area is the cause of Parkinson’s disease (see below). An area that is functionally
analogous to the substantia nigra pars compacta is the ventral tegmental area, which is located
nearby and makes a dopaminergic projection to the nucleus accumbens.

Basal ganglia afferents

The striatum is the main recipient of afferents to the basal ganglia (Fig. 3). These
excitatory afferents arise from the entire cerebral cortex and from the intralaminar nuclei of the
thalamus (primarily the centromedian nucleus and parafascicularis nucleus). The projections
from different cortical areas are segregated, such that the frontal lobe projects predominantly to
the caudate head and the putamen; the parietal and occipital lobes project to the caudate body;
and the temporal lobe projects to the caudate tail. The primary motor cortex and the primary

44
somatosensory cortex project mainly to the putamen, while the premotor cortex and
supplementary motor areas project to the caudate head. Other cortical areas project primarily to
the caudate. Thus, along the C-shaped extent of the caudate nucleus, the caudate cells receive
their input from the cortical regions that are close by. The enlarged head of the caudate reflects
the large projection from the frontal cortex to the caudate. In addition, the nucleus accumbens
(ventral striatum) receives a large input from limbic cortex.

Figure 3. Basal ganglia afferents. For diagram simplicity, in this and subsequent figures, the
caudate and putamen are represented by the putamen only, as the two regions have similar
connections.
Basal ganglia efferents

The major output structures of the basal ganglia are the globus pallidus internal segment
(GPint) and the substantia nigra pars reticulata (SNr) (Fig. 4). Both of these structures make
GABAergic, inhibitory connections on their targets. The GPint projects to a number of thalamic
structures by way of two fiber tracts: the ansa lenticularis and the lenticular fasciculus. The loop
that processes sensorimotor information from the motor cortex and the somatosensory cortex
projects to the ventral anterior (VA) and ventral lateral (VL) nuclei. The loop that processes
other neocortical information projects to the dorsomedial nucleus (DM), intralaminar nuclei, and
parts of the VA nucleus. The SNr projects to the superior colliculus, which is involved in eye
movements, as well as to the VA/VL thalamic nuclei.

45
 Figure 4. Basal ganglia efferents

Basal ganglia intrinsic connections

A number of intrinsic pathways interconnect various basal ganglia structures (Fig. 5).

(1) the striatopallidal pathway is a GABAergic, inhibitory connection between the striatum and
both segments of the globus pallidus.
(2) the striatonigral pathway is a GABAergic, inhibitory connection between the striatum and the
SNr.
(3) the globus pallidus external segment makes a GABAergic, inhibitory connection to the
subthalamic nucleus.
(4) the subthalamic nucleus makes glutamatergic, excitatory connections onto both segments of
the globus pallidus and the SNr. This pathway is the only purely excitatory pathway among
the intrinsic pathways of the basal ganglia.
(5) The nigrostriatal pathway makes a dopaminergic synapse onto striatal neurons. As we will
see below, this is a mixed pathway, with excitatory effects on some striatal neurons and
inhibitory effects on others.

46
 Fig. 5. Basal ganglia intrinsic connections

Two pathways process signals in the basal ganglia

There are two distinct pathways that process signals through the basal ganglia: the direct
pathway and the indirect pathway. These two pathways have opposite net effects on thalamic
target structures. Excitation of the direct pathway has the net effect of exciting thalamic neurons
(which in turn make excitatory connections onto cortical neurons). Excitation of the indirect
pathway has the net effect of inhibiting thalamic neurons (rendering them unable to excite motor
cortex neurons). The normal functioning of the basal ganglia apparently involves a proper
balance between the activity of these two pathways. One hypothesis is that the direct pathway
selectively facilitates certain motor (or cognitive) programs in the cerebral cortex that are
adaptive for the present task, while the indirect pathway simultaneously inhibits the execution of
competing motor programs. An upset of the balance between the direct and indirect pathways
results in the motor dysfunctions that characterize the extrapyramidal syndrome (see below).

Although the connectivity patterns of the direct and indirect pathways are relatively
straightforward, the predominance of inhibitory connections in the system can make an
understanding of the functional circuitry complicated and nonintuitive (Fig. 6). The direct
pathway starts with cells in the striatum that make inhibitory connections with cells in the GPint.
The GPint cells in turn make inhibitory connections on cells in the thalamus. Thus, the firing of
GPint neurons inhibits the thalamus, making the thalamus less likely to excite the neocortex.
When the direct pathway striatal neurons fire, however, they inhibit the activity of the GPint
neurons. This inhibition releases the thalamic neurons from inhibition (i.e., it disinhibits the
thalamic neurons), allowing them to fire to excite the cortex. Thus, because of the “double
negative” in the pathway between the striatum and GPint and the GPint and thalamus, the net
result of exciting the direct pathway striatal neurons is to excite motor cortex.

47
 Think of it as a multiplication equation, with an excitatory connection (E) equal to +1 and
an inhibitory connection (I) equal to –1:

E I I
Cortex Æ striatum Æ GPint Æ thalamus
+1 x –1 x –1 = +1

because the two negative numbers cancel each other out.

Fig. 6. Direct/indirect pathways. Thick (green) lines represent excitatory connections and thin (red)
lines represent inhibitory connections.

The indirect pathway starts with a different set of cells in the striatum. These neurons
make inhibitory connections to the external segment of the globus pallidus (GPext). The GPext
neurons make inhibitory connections to cells in the subthalamic nucleus, which in turn make
excitatory connections to cells in the GPint. (Remember that the subthalamic-GPint pathway is
the only purely excitatory pathway within the intrinsic basal ganglia circuitry.) As we saw
before, the GPint neurons make inhibitory connections on the thalamic neurons. To see the net
effects of activation of the indirect pathway, let us work backwards from the GPint. When the
GPint cells are active, they inhibit thalamic neurons, thus making cortex less active. When the
subthalamic neurons are firing, they increase the firing rate of GPint neurons, thus increasing the
net inhibition on cortex. Firing of the GPext neurons inhibits the subthalamic neurons, thus
making the GPint neurons less active and disinhibiting the thalamus. However, when the
indirect pathway striatal neurons are active, they inhibit the GPext neurons, thus disinhibiting the
subthalamic neurons. With the subthalamic neurons free to fire, the GPint neurons inhibit the
thalamus, thereby producing a net inhibition on the motor cortex.

48
 Again, think of a multiplication analogy:

E I I E I
Cortex Æ striatum Æ GPext Æ Subthalamic Nucleus Æ GPint Æ thalamus
+1 x –1 x –1 x +1 x –1 = –1

Because there are 3 negative numbers in the equation, the net effect is negative.

Thus, as a result of the complex sequences of excitation, inhibition, and

disinhibition, the net effect of the cortex exciting the direct pathway is to further excite the
cortex (positive feedback loop), whereas the net effect of cortex exciting the indirect
pathway is to inhibit the cortex (negative feedback loop). Presumably, the function of the
basal ganglia is related to a proper balance between these two pathways. Motor cortex neurons
have to excite the proper direct pathway neurons to further increase their own firing, and they
have to excite the proper indirect pathways neurons that will inhibit other motor cortex neurons
that are not adaptive for the task at hand.

The nigrostriatal projection

An important pathway in the modulation of the direct and indirect pathways is the
dopaminergic, nigrostriatal projection from the substantia nigra pars compacta to the striatum.
Direct pathways striatal neurons have D1 dopamine receptors, which depolarize the cell in
response to dopamine. In contrast, indirect pathway striatal neurons have D2 dopamine
receptors, which hyperpolarize the cell in response to dopamine. The nigrostriatal pathway
thus has the dual effect of exciting the direct pathway while simultaneously inhibiting the
indirect pathway. Because of this dual effect, excitation of the nigrostriatal pathway has the net
effect of exciting cortex by two routes, by exciting the direct pathway (which itself has a net
excitatory effect on cortex ) and inhibiting the indirect pathway (thereby disinhibiting the net
inhibitory effect of the indirect pathway on cortex). The loss of these dopamine neurons in
Parkinson’s disease causes the poverty of movement that characterizes this disease, as the
balance between direct pathway excitation of cortex and indirect pathway inhibition of cortex is
tipped in favor of the indirect pathway, with a subsequent pathological global inhibition of motor
cortex areas.

Functions of the basal ganglia

Motor functions

The function of the basal ganglia in motor control is not understood in detail. It appears
that the basal ganglia is involved in the enabling of practiced motor acts and in gating the
initiation of voluntary movements by modulating motor programs stored in the motor cortex and
elsewhere in the motor hierarchy. Thus, voluntary movements are not initiated in the basal
ganglia (they are initiated in the cortex); however, proper functioning of the basal ganglia

49
appears to be necessary in order for the motor cortex to relay the appropriate motor commands to
the lower levels of the hierarchy.

What is the function of the tonic inhibitory output of the basal ganglia? Recall from the
“Motor Cortex” lecture that stimulating the motor cortex of monkeys at various locations results
in stereotyped sequences of movements, such as bringing the hand to the mouth or adopting a
defensive posture. It appears that a number of “primitive” motor programs are stored in the
cortex, and motor control may require the activation of these elemental motor programs in the
precise temporal order to accomplish a sophisticated motor plan. It is important that only one
motor program be active at a given time, however, such that one motor act (e.g., use hand to
bring food to the mouth) is not competing with a conflicting motor act (e.g., use hand to shield
face from dangerous object). It is thought that the basal ganglia is normally active in suppressing
inappropriate motor programs, and that activation of the direct pathway temporarily releases one
motor program from inhibition, enabling it to be executed by the organism. Thus, the basal
ganglia act as a gate that enables the execution of automatic programs in the hierarchy.

Which motor programs should be released from inhibition at a given moment? The basal
ganglia may have a major role in learning what motor acts result in rewards for the organism.
This information is provided by the dopaminergic neurons of the SNc and ventral tegmental
nucleus. Recordings from these neurons in monkeys have shown that they tend to respond when
the monkey receives an unexpected reward, and they tend to be inhibited when the monkey fails
to receive an expected reward (Fig. 7). Because the net effect of activation of the nigrostriatal
pathway is to excite the direct pathway and inhibit the indirect pathway, this pattern of
dopaminergic firing may be involved in tuning the relative balance of direct/indirect pathway
activity to enhance the firing of cortical motor programs that produce rewarding outcomes and to
suppress the activity of motor programs that do not result in reward. In this way, motor habits
can be constructed that tend to reward the animal.
Figure 7. Dopaminergic neurons signal
unexpected reward or unexpected
absence of reward. (A) If a reward
occurs unexpectedly, the dopaminergic
neuron fires briskly. This may
strengthen the cortical motor programs
that led up to the reward. (B) If a
reward occurs that the monkey
previously learned was predicted by
stimulus, the neuronal firing is not
altered. This may signal that all is
proceeding normally. (C) If the
reward-predicting stimulus does not
produce a reward, the neuronal firing is
inhibited. This may weaken the
cortical motor programs that did not
produce the expected reward.
Cognitive functions

As mentioned earlier, there are a number of cortical loops through the basal ganglia that
involve prefrontal association cortex and limbic cortex. Through these loops, the basal ganglia

50
are thought to play a role in cognitive function that is similar to their role in motor control. That
is, the basal ganglia are involved in selecting and enabling various cognitive, executive, or
emotional programs that are stored in these other cortical areas. Moreover, the basal ganglia
appear to be involved in certain types of learning. For example, in rodents the striatum is
necessary for the animal to learn certain stimulus-response tasks (e.g., make a right turn if
stimulus A is present and make a left turn if stimulus B is present). Recordings from rat striatal
neurons show that early in training, striatal neurons fire at many locations while a rat learns such
a task on a T-shaped maze (Fig. 8). This suggests that initially the striatum is involved
throughout the execution of the task. As the animal learns the task and becomes exceedingly
good at its performance, the striatal neurons change their activity patterns, firing only at the
beginning of the trial and at the end. It appears that the learned programs to solve this task are
now stored elsewhere; the firing of the striatal neurons at the beginning of the maze is presumed
to reflect its enabling of the appropriate motor/cognitive plan in the cortex, and the firing at the
end of the maze is involved in evaluating the reward outcome of the trial.

Figure 8. Habit learning in striatum. (A) A rat is trained to run down a T-shaped maze and make a left turn for food
reward if it hears a high-pitch tone or make a right turn for food reward if it hears a low-pitch tone. (B) Early in
training, as the rat is beginning to learn the task, striatal neurons fire at locations all over the maze, especially at the
choice point. (C) Late in training, when the rat has mastered the task and performs very quickly and accurately, the
striatal neurons now fire only at the start and ends of the maze.

In humans, the basal ganglia appear to be necessary for certain forms of implicit memory
tasks. Like motor habit learning discussed above, many types of cognitive learning require
repeated trials and are often unconscious. An example is probabilistic classification. In this type
of task, people have to learn to classify objects based on the probability of belonging to a class,
rather than on any explicit rule. In one experiment, subjects were shown a deck of cards with
different symbols. Each symbol was associated with a certain probability (60-85%) of predicting
rain or sunshine, and the subjects had to say on each trial whether the symbol was a predictor of
rain or sunshine. Because the same symbol sometimes predicted sunshine and other times
predicted rain, the subjects could not devise a simple rule, and they made many errors at first.
Over time, however, they began to get better at classifying the symbols appropriately, although
they still often claimed to be guessing. Patients with basal ganglia disorders were impaired at
this task, suggesting that the processing of the cognitive loops of the basal ganglia are somehow
involved in our ability to subconsciously learn the probabilities of predicted outcomes associated
with particular stimuli.

51
Disorders of the basal ganglia

A number of neurological disorders result from damage to the basal ganglia. Two of
these disorders (Parkinson’s disease and Huntington’s disease) will be briefly discussed here to
relate the concepts learned in this chapter to the symptoms of the disorders. More thorough
treatment of these disorders will be given in Chapter 7.

Nigrostriatal pathway and Parkinson’s disease

Parkinson’s disease is characterized by slowness or absence of movement (bradykinesia

or akinesia), rigidity, and a resting tremor (especially in the hands and fingers). Patients have
difficulty initiating movements, and once initiated the movements are abnormally slow. The
cause of Parkinson’s disease is the loss of the dopaminergic neurons in the substantia nigra pars
compacta. From one’s knowledge of the effects of the nigrostriatal pathway on the direct and
indirect pathways, it becomes straightforward to see why the loss of this pathway results in the
poverty of movement symptomatic of Parkinson’s disease. Because the nigrostriatal pathway
excites the direct pathway and inhibits the indirect pathway, the loss of this input tips the balance
in favor of activity in the indirect pathway. Thus, the GPint neurons are abnormally active,
keeping the thalamic neurons inhibited. Without the thalamic input, the motor cortex neurons
are not as excited, and therefore the motor system is less able to execute the motor plans in
response to the patient’s volition.

Indirect pathway and Huntington’s disease

The symptoms of Huntington’s disease are in many respects the opposite of the
symptoms of Parkinson’s disease. Huntington’s disease is characterized by choreiform
movements: involuntary, continuous movement of the body, especially of the extremities and
face. Often these movements resemble pieces of adaptive movements, but they occur
involuntarily and without behavioral significance. Huntington’s disease results from the
selective loss of striatal neurons in the indirect pathway. Thus, the balance between the direct
and indirect pathways becomes tipped in favor of the direct pathway. Without the normal
inhibitory influence on the thalamus that is provided by the indirect pathway, thalamic neurons
can fire randomly and inappropriately, causing the motor cortex to execute motor programs with
no control by the patient.

52
 MOTOR CORTEX

The previous chapters discussed the lower levels of the motor hierarchy (the spinal cord
and brainstem), which are involved in the low-level, “nuts and bolts” processing that
controls the activity of individual muscles. Individual alpha motor neurons control the
force exerted by a particular muscle, and spinal circuits can control sophisticated and
complex behaviors such as walking and reflex actions. The types of movements
controlled by these circuits are not initiated consciously, however. Voluntary movements
require the participation of the third and fourth levels of the hierarchy: the motor cortex
and the association cortex. These areas of the cerebral cortex plan voluntary actions,
coordinate sequences of movements, make decisions about proper behavioral strategies
and choices, evaluate the appropriateness of a particular action given the current
behavioral or environmental context, and relay commands to the appropriate sets of lower
motor neurons to execute the desired actions.

Motor cortex comprises the primary motor cortex, premotor cortex, and
supplementary motor area

The motor cortex comprises three different areas of the frontal lobe, immediately anterior
to the central sulcus. These areas are the primary motor cortex (Brodmann’s area 4), the
premotor cortex, and the supplementary motor area (Fig. 1). Electrical stimulation of
these areas elicits movements of particular body parts. The primary motor cortex, or M1,
is located on the precentral gyrus and on the anterior paracentral lobule on the medial
surface of the brain. Of the three motor cortex areas, stimulation of the primary motor
cortex requires the least amount of electrical current to elicit a movement. Low levels of
brief stimulation typically elicit simple movements of individual body parts. Stimulation
of premotor cortex or the supplementary motor area requires higher levels of current to
elicit movements, and often results in more complex movements than stimulation of
primary motor cortex. Stimulation for longer time periods (500 msec) in monkeys results
in the movement of a particular body part to a stereotyped posture or position, regardless

Figure 1. Motor cortex areas. Note that the

supplementary motor area continues onto the
medial wall of the cortex.

of the initial starting point of the body part. Thus, the premotor cortex and
supplementary motor areas appear to be higher level areas that encode complex patterns
of motor output and that select appropriate motor plans to achieve desired end results.

53
Like the somatosensory cortex of the postcentral gyrus, the primary motor cortex is
somatotopically organized (Figure 2). Stimulation of the anterior paracentral lobule
elicits movements of the contralateral legs. As the stimulating electrode is moved across
the precentral gyrus from dorsomedial to ventrolateral, movements are elicited
progressively from the torso, arms, hands and face (most laterally). The representation of
body parts that perform precise, delicate movements, such as the hands and face, is
disproportionately large compared to the representation of body parts that perform only
coarse, unrefined movements, such as the trunk or legs. The premotor cortex and
supplementary motor area also contain somatotopic maps of body parts.

Figure 2. Somatotopic representation of motor

outputs in motor cortex.

One might predict that the motor cortex “homunculus” arises because neurons that
control individual muscles are clustered together in the cortex. That is, all of the neurons
that control the biceps muscle may be located together, and all of the neurons that control
the triceps may be clustered nearby, and the neurons that control the soleus muscle may
be clustered in a region further removed. Electrophysiological recordings have shown
that this is not the case, however. Movements of individual muscles are correlated with
activity from widespread parts of the primary motor cortex. Similarly, stimulation of
small regions of primary motor cortex elicits movements that require the activity of
numerous muscles. Thus, the primary motor cortex homunculus does not represent the
activity of individual muscles. Rather, it apparently represents the movements of
individual body parts, which often require the coordinated activity of large groups of
muscles throughout the body.

Cortical afferents and efferents

The motor cortex exerts its influence over muscles by a variety of descending routes
(Figure 3). Some of the descending pathways reviewed in the last chapter can be
influenced by motor cortex output. Thus, in addition to the direct cortical innervation of
alpha motor neurons via the corticospinal tract, the following cortical efferent pathways
influence the remaining descending tracts:

54
 1) the corticorubral tract allows cortex to modulate the rubrospinal tract
2) the corticotectal tract allows cortex to modulate the tectospinal tract
3) the corticoreticular tract allows cortex to modulate the reticulospinal tracts

Figure 3. Parallel pathways from the

motor cortex allow the cortical motor
areas to influence the processing of all
descending motor tracts and side loops
of the m otor system.

Figure 4. Major afferents of motor cortex.

Motor cortex cytoarchitecture

Like all parts of the neocortex, the primary motor cortex is made of six layers (Fig. 5).
Unlike primary sensory areas, primary motor cortex is agranular cortex; that is, it does
not have a cell-packed granular layer (layer IV). Instead, the most distinctive layer of
primary motor cortex is its descending output layer (Layer V), which contains the giant
Betz cells. These pyramidal cells and other projection neurons of the primary motor
cortex make up ~30% of the fibers in the corticospinal tract. The rest of the fibers come

55
from the premotor cortex and the supplementary motor area (~30%), the somatosensory
cortex (~30%), and the posterior parietal cortex (~10%).

Fig. 5. Pyramidal and non-pyramidal neurons in motor

cortex. The cerebral cortex is organized into six layers.
These layers contain different proportions of the two
main classes of cortical neurons, pyramidal and non-
pyramidal cells. Pyramidal cells send long axons down
the spinal cord and are the major output neurons. They
are abundant in layer V. Non-pyramidal cells have axons
which terminate locally.

Encoding of movement by motor cortex

Primary motor cortex

Alpha motor neurons in the spinal cord encode the force of contraction of groups of
muscle fibers using the rate code and the size principle. As discussed above, the primary
motor cortex does not generally control individual muscles directly, but rather appears to
control individual movements or sequences of movements that require the activity of
multiple muscle groups. Thus, in accordance with the concept of hierarchical
organization of the motor system, the information represented by motor cortex is a higher
level of abstraction than the information represented by spinal motor neurons.

What is encoded by the neurons in primary motor cortex? Clues have come from
recording the activity of these neurons as experimental animals perform different motor
tasks. In general, primary motor cortex encodes the parameters that define individual
movements or simple movement sequences.

(1) Primary motor cortex neurons fire 5-100 msec before the onset of a movement.
Thus, rather than firing as the result of muscle activity, these neurons are involved in
relaying motor commands to the alpha motor neurons that eventually cause the
appropriate muscles to contract.

(2) Primary motor cortex encodes the force of a movement. The amount of force
required to raise the arm from one location to another is much greater if one is holding a
bowling ball than if one is holding a balloon. Many neurons in primary motor cortex
encode the amount of force that is necessary to make such a movement. Note the
distinction between movement force and muscle force. Primary motor cortex neurons
encode the amount of force necessary for a particular movement, regardless of which
individual muscles are used. Alpha motor neurons, in turn, translate the commands of the
motor cortex neurons and control the amount of force generated by individual muscles to
accomplish that movement, under the principles of the rate code and the size principle.

56
(3) Primary motor cortex encodes the direction of movement. Many neurons in the
primary motor cortex are selective for a particular direction of movement. For example,
one cell may fire strongly when the hand is moved to the left, whereas it will be inhibited
when the hand is moved to the right.

(4) Primary motor cortex encodes the extent of movement. The firing of some
neurons is correlated with the distance of a movement. A monkey was trained to move
its arm to different target locations that varied in direction and distance from the center.
The firing of many neurons was correlated with the direction of movement (as in Point
3), whereas the firing of other neurons was correlated with the distance of the movement.
Interestingly, some neurons were correlated with the interaction of a particular distance
and direction; that is, they were correlated with a particular target position.

(5) Primary motor cortex neurons encode the speed of movement. Almost all targeted
movements follow a typical bell-shaped curve of velocity as a function of distance (Fig.
6). For example, when the hand reaches out to an object, the hand accelerates during the
first half of the movement, reaches a peak velocity approximately halfway to the target,
and then decelerates until it reaches the target object. The firing rate of some primary
motor cortex neurons in monkeys correlates with this bell-shaped speed profile,
demonstrating that information about movement speed is contained in the spike trains of
these neurons.

Figure 6. Voluntary movements typically

follow a bell-shape speed profile. The firing of
some neurons in primary motor cortex is
correlated with this speed profile.

Premotor Cortex

The premotor cortex sends axons to the primary motor cortex as well as to the spinal cord
directly. It performs more complex, task-related processing than primary motor cortex.
Stimulation of premotor areas in the monkey at a high level of current produces more
complex postures than stimulation of the primary motor cortex. The premotor cortex
appears to be involved in the selection of appropriate motor plans for voluntary
movements, whereas the primary motor cortex is involved in the execution of these
voluntary movements.

(1) Premotor cortex neurons signal the preparation for movement (Fig. 7). Monkeys
were trained to make a particular movement in response to a visual signal, with a variable
delay between the onset of the signal and the onset of the movement. Recordings from
premotor cortex have shown that many neurons fire selectively in the delay interval, for
many seconds before the onset of the movement. A particular neuron will fire when the

57
monkey is preparing to make a movement to the left, for example, but will be silent when
the monkey is preparing to make a movement to the right. Thus, the firing of this type of
neuron does not cause the movement itself, but appears to be involved in preparing the
monkey to make the correct movement when the “Go” signal is given.

Figure 7. Some premotor cortex neurons fire for many seconds in anticipation of a planned
movement.

(2) Premotor cortex neurons signal various sensory aspects associated with
particular motor acts. Some premotor neurons fire when the animal is performing a
particular action, such as breaking a peanut. Interestingly, the same neuron fires
selectively when the animal sees another monkey or person breaking the peanut. It also
fires selectively to the sound of a peanut shell being broke, even without any visual or
motor activity. These neurons are called “mirror” neurons, because they respond not
only to a particular action of the monkey but also to the sight (or sound) of another
individual performing the same action.

Supplementary Motor Area

The supplementary motor area (SMA) is involved in programming complex sequences of

movements and coordinating bilateral movements. Whereas the premotor cortex appears
to be involved in selecting motor programs based on visual stimuli or on abstract
associations, the supplementary motor area appears to be involved in selecting
movements based on remembered sequences of movements.

(1) SMA responds to sequences of movements and to mental rehearsal of sequences

of movements (Fig. 8). Brain activity was measured in a PET scanner while subjects
made simple and complex sequences of movement. When the movements were simple,
such as a repetitive movement of a single digit, the primary motor cortex and the primary
somatosensory cortex were activated on the contralateral hemisphere. When the subject
was asked to perform a complex sequence of finger movements, the SMA was activated
bilaterally, in addition to the contralateral primary motor and somatosensory cortex
activation. Finally, when the subject was asked to remain still but to mentally rehearse
the complex sequence of activity, the SMA was still active, even though the primary

58
motor and somatosensory cortex areas were silent. Thus, the SMA appears to be
involved in bilateral movements and in the mental rehearsal of these movements.

Fig. 8. Positron emission tomography (PET) study of simple vs. complex finger movements. Ovals
indicate regions of increased blood flow. Data from Roland PE, et al. (1980) J. Neurophysiol. 43: 118-
136.

(2) SMA is involved in the transformation of kinematic to dynamic information.

Movements can be defined in terms of dynamics (the amount of force necessary to make
a movement) and kinematics (the distance and angles that define a particular movement
in space). Many movement plans are represented in kinematic terms (e.g., Move the
hand to the left). However, the motor system must eventually translate this to a
representation based on dynamics, in order to instruct the appropriate muscles to contract
with the appropriate force. Recordings from monkeys have shown that during the
preparatory delay before a monkey makes an instructed movement, some SMA neurons
change their firing correlates from a kinematic-based representation to a dynamics-based
representation, suggesting that SMA plays a vital role in this transformation.

59
 Association Cortex

The fourth level of the motor hierarchy is the association cortex, in particular the
prefrontal cortex and the posterior parietal cortex. These brain areas are not motor areas
in the strict sense. Their activity does not correlate precisely with individual motor acts,
and stimulation of these areas does not result in motor output. However, these areas are
necessary to ensure that movements are adaptive to the needs of the organism and
appropriate to the behavioral context.

(1) Posterior parietal cortex is involved in ensuring that movements are targeted
accurately to objects in external space. This area is involved in processing spatial
relationships of objects in the world and in constructing a representation of external space
that is independent of the observer’s eye position or body position. Such representations
allow a stable percept of the world that is independent of viewer orientation, as well as
the representation of desired trajectories in space that are independent of body position.
Damage to the posterior parietal cortex can result in a number of apraxias, that is, the
inability to make complex, coordinated movements. For example, a patient with
constructional apraxia is unable to replicate the configuration of a set of blocks in the
proper sequence, even though the patient is able to maneuver each block individually
with dexterity.

(2) Prefrontal cortex is involved in the selection of appropriate actions for a

particular behavioral context. It is also involved in the evaluation of the consequences
of a particular course of action. Patients with damage to the prefrontal cortex have
problems in executive processing. They make inappropriate behavioral decisions, and
often cannot anticipate the likely consequences of their actions. They display impulsive
behavior, often showing an inability to delay instant gratification for a long-term larger
reward.

60
 THE INTEGRATED MOTOR SYSTEM AND
DISORDERS OF THE MOTOR SYSTEM

The previous motor system lectures have deconstructed the motor system into its
component parts, in an effort to portray how the brain’s “divide and conquer” strategy assigns
different motor control tasks to different brain regions. This chapter describes the types of
disorders that result from damage or disease to different parts of the motor system. In the
process, the different components of the motor system are reviewed to see how they work
together to produce the fluid, effortless body movements that we take for granted. An emphasis
will be placed on trying to explain the causes and symptoms of motor system disorders in terms
of the basic principles of neuroanatomy and neuronal function that you learned in the earlier
lectures.

Lower Motor Neuron Syndrome

The first level of the motor system hierarchy is the spinal cord, the location of the alpha
motor neurons that constitute the “final common pathway” of all motor commands. Alpha motor
neurons are the neurons that directly innervate skeletal muscle, causing the contractions that
produce all movements. Reflex circuits and other neural circuitry within the spinal cord underlie
the automatic processing of many of the direct commands to the muscles (the “nuts and bolts”
processing), thereby freeing higher-order areas to concentrate on more global, task-related
processing.

Motor system dysfunction can result from damage or disease at any level of the motor
system hierarchy and side-loops. Differences in the symptoms that result from damage at
different levels allow the clinician to localize where in the hierarchy the damage is likely to be.
Damage to alpha motor neurons results in a characteristic set of symptoms called the lower
motor neuron syndrome (lower motor neurons refer to alpha motor neurons in the spinal cord
and brain stem; all motor system neurons higher in the hierarchy are referred to as upper motor
neurons). This damage usually arises from certain diseases that selectively affect alpha motor
neurons (such as polio) or from localized lesions near the spinal cord. Lower motor neuron
syndrome is characterized by the following symptoms:

(1) The effects can be limited to small groups of muscles. Recall that a motor neuron
pool is a nucleus of alpha motor neurons that innervate a single muscle. Furthermore, nearby
motor neuron pools control nearby muscles. Thus, restricted damage to lower motor neurons,
either within the spinal cord or at the ventral roots, will affect only a restricted group of muscles.

(2) Muscle atrophy. When alpha motor neurons die, the muscle fibers that they
innervate become deprived of trophic factors necessary for their survival. Eventually, the muscle
itself atrophies.

(3) Weakness. Because of the damage to alpha motor neurons and the atrophy of
muscles, weakness is profound in lower motor neuron disorders.

61
 (4) Fasciculation. When alpha motor neurons degenerate, they can produce spontaneous
action potentials. These spikes cause the muscle fibers that are part of that neuron’s motor unit
to fire, resulting in a visible twitch of the affected muscle. This twitch is called a fasciculation.

(5) Fibrillation. With further degeneration of the alpha motor neuron, only remnants of
the axons near the muscle fibers remain. These individual axon fibers can also generate
spontaneous action potentials; however, these action potentials will only cause individual muscle
fibers to contract. This spontaneous twitching of individual muscle fibers is called a fibrillation.
Fibrillations are too small to be seen as a visible muscle contraction. They can only be detected
with electrophysiological recordings of the muscle activity (an electromyogram).

(6) Hypotonia. Because alpha motor neurons are the only way to stimulate extrafusal
muscle fibers, the loss of these neurons causes a decrease in muscle tone.

(7) Hyporeflexia. The myotatic (stretch) reflex is weak or completely absent with lower
motor neuron disorders, because the alpha motor neurons that cause the muscle to contract are
damaged.

Upper motor neuron syndrome

Damage to any part of the motor system hierarchy above the level of alpha motor neurons
(not including the side loops) results in a set of symptoms termed the upper motor neuron
syndrome. Some of these symptoms are opposite of those experienced with lower motor neuron
disorders. Thus, one of the critical determinations a clinician must make is whether a patient
presenting with motor problems has an upper motor neuron disorder or a lower motor neuron
disorder.

Upper motor neuron disorders typically arise from such causes as stroke, tumors, and
blunt trauma. For example, strokes to the middle cerebral artery, lateral striate artery, or the
medial striate artery can cause damage to the lateral surface of cortex or to the internal capsule,
where the descending axons of the corticospinal tract collect. The symptoms of upper motor
neuron syndrome are:

(1) The effects extend to large groups of muscles. Recall from the “Motor Cortex”
lecture that muscles from different body parts are activated by stimulation of parts of motor
cortex, consistent with the notion that motor cortex represents movements that are controlled by
many joints, rather than individual muscles. Thus, a stroke in a particular part of motor cortex
will affect the activation of many muscles in the body. Likewise, a stroke that affects the
internal capsule or crus cerebri could affect muscles on the entire contralateral side of the body.

(2) Atrophy is rare. Because alpha motor neurons are present, muscles will continue to
receive trophic agents necessary for their survival. A mild amount of atrophy may result from
disuse, but it will not be as pronounced as that resulting from a lower motor neuron disorder.

62
 (3) Weakness. Upper motor neuron disorders produce a graded weakness of movement
called paresis, which is different from the complete loss of muscle activity caused by paralysis
(plegia).

(4) Absence of fasciculations. Because alpha motor neurons themselves are spared,
fasciculations do not occur.

(5) Absence of fibrillations. Likewise, fibrillations do not occur, as the alpha motor
neurons are not damaged.

(6) Hypertonia. Upper motor neuron disorders result in an increase in muscle tone.
Recall that descending motor pathways can modulate the intrinsic circuitry that is present in the
spinal cord. This modulatory input can be either inhibitory or excitatory. Through mechanisms
that are not well understood, the loss of descending inputs tends to result in an increased firing
rate of alpha and/or gamma motor neurons. The higher firing rate causes an increase in the
resting level of muscle activity, resulting in hypertonia.

(7) Hyperreflexia. Because of the loss of inhibitory modulation from descending

pathways, the myotatic (stretch) reflex is exaggerated in upper motor neuron disorders. The
stretch reflex is a major clinical diagnostic test of whether a motor disorder is caused by damage
to upper or lower motor neurons.

(8) Clonus. Sometimes the stretch reflex is so strong that the muscle contracts a number
of times in a 5-7 Hz oscillation when the muscle is rapidly stretched and then held at a constant
length. This abnormal oscillation, called clonus, can be felt by the clinician.

(9) Initial contralateral flaccid paralysis. In the initial stages after damage to motor
cortex, the contralateral side of the body shows a flaccid paralysis. Gradually, over the course of
a few weeks, motor function returns to the contralateral side of the body. This gradual recovery
of function results from the ability of other motor pathways to take over some of the lost
functions. Recall that there are multiple descending motor pathways by which high-order
information can reach the spinal cord. Thus, descending pathways such as the rubrospinal and
the reticulospinal tracts, which receive direct or indirect cortical input, can begin to take over the
function lost by the damage to the corticospinal tract. Moreover, primary motor cortex itself is
known to be capable of reorganizing itself to recover some lost function. Thus, if the part of
motor cortex that controls a certain body movement is damaged, neighboring parts of the motor
cortex that are undamaged can, to some extent, alter their function to help compensate for the
damaged areas. The one major exception to the recovery of function is that fine control of the
distal musculature will not be regained after a lesion to the corticospinal tract. Recall that there
are direct connections from primary motor cortex neurons to alpha motor neurons controlling the
fingers. These connections presumably underlie our abilities to manipulate objects with great
precision and to do such tasks as playing a piano and performing microsurgery. None of the
other descending pathways have direct connections onto spinal motor neurons, and none of them
can compensate for the loss of fine motor control of the hands and fingers after damage to the
corticospinal tract.

63
 (10) Spasticity. A clinical sign of upper motor neuron disorder is a velocity dependent
resistance to passive movement of the limb. If the clinician moves a patient’s limb slowly, there
may be little resistance to the movement. As the passive movement becomes quicker, however,
at a certain point the muscle will sharply resist the movement. This is referred to as a “spastic
catch.” The mechanism for this spasticity is not entirely known, but altered firing rate of gamma
motor neurons and their regulating interneurons may be involved, as well as an increase in alpha
motor neuron activity, causing an inappropriately powerful stretch reflex to a fast stretch of the
muscle. Sometimes, the resistance becomes so great that the autogenic inhibition reflex is
initiated, causing a sudden drop in the resistance; this is referred to as the clasp-knife reflex.

(11) Babinski sign. A classic neurological test for corticospinal tract damage is the
Babinski test. In this test, the clinician strokes the sole of the foot firmly with an instrument.
This elicits a normal plantar response in normal individuals, as the toes curl inward. In patients
with an upper motor neuron disorder, however, an abnormal extensor plantar response is elicited,
as the big toe extends upward and the remaining toes fan out. This is called a positive Babinski
sign. Interestingly, the positive Babinski sign is normal in infants for the first 2 years of life.
During development, however, the reflex changes to the normal adult pattern, presumably as
corticospinal circuits mature.

In addition to the above symptoms, damage to the motor cortex and association cortex
can result in impairments in motor planning and strategies and in an inability to perform complex
motor tasks. Performance of simple tasks is intact, but patients are unable to perform complex,
practiced tasks. This symptom is known as apraxia. For example, patients may be unable to
arrange a set of blocks to match an example block-structure in front of them. They can move the
blocks individually, but cannot come up with a motor plan to arrange them properly. This
disorder is known as constructional apraxia. Other apraxias include dressing apraxia (inability to
dress oneself) and verbal apraxia (inability to coordinate mouth movements to produce speech).

Paralysis

A section or crush of the spinal cord will result in paralysis of all parts of the body below
the damaged region. Even though such an injury occurs in the spinal cord, it is not considered a
lower motor neuron disorder, as the alpha motor neurons themselves are not directly damaged.
If the damage occurs at the cervical level, then all four limbs will be paralyzed (quadriplegia). If
the damage occurs below the cervical enlargement, then only the legs are paralyzed (paraplegia).
Other terms used to describe patterns of paralysis are hemiplegia (paralysis to one side of the
body) and monoplegia (paralysis of a single limb).

Disorders of the basal ganglia

The basal ganglia have historically been considered part of the motor system because of
the variety of motor deficits that arise when they are damaged. The types of symptoms that
result from basal ganglia disorders can be divided into two classes: dyskinesias, which are
abnormal, involuntary movements, and akinesias, which are abnormal, involuntary postures.
Because the basal ganglia were once considered to form a separate, “extrapyramidal” motor
system, these symptoms are called extrapyramidal disorders.

64
 Dyskinesia

The dyskinesias that result from basal ganglia disorders are described below:

(1) Resting tremors are most often associated with Parkinson’s disease. When the
patient is at rest, certain body parts will display a 4-7 Hz tremor. For example, the thumb and
forefingers will move back-and-forth against each other in a characteristic tremor called “pill-
rolling tremor.” The tremor stops when the body part engages in active movement.

(2) Athetosis is characterized by involuntary, writhing movements, especially of the

hands and face.

(3) Chorea, which derives from the Greek word for “dance,” is characterized by
continuous, writhing movements of the entire body. It is viewed by some as an extreme form of
athetosis. Chorea is most closely identified with Huntington’s disease.

(4) Ballismus is characterized by sudden, involuntary, ballistic movements of the

extremities.

Akinesia

The following describes the akinesias that result from basal ganglia disorders:

(1) Rigidity is a resistance to passive movement of the limb. Unlike spasticity, rigidity
does not depend on the speed of the passive movement. In some patients, this resistance is so
great that it is referred to as lead-pipe rigidity, because moving the patient’s limb feels like
bending a lead pipe. In some patients, this rigidity is coupled with tremors and is called
cogwheel rigidity, as moving the limb feels to the clinician like the catching and release of gears.
As with spasticity, the mechanism is not entirely understood, but may result from continuous
firing of alpha motor neurons causing a continual contraction of the muscle.

(2) Dystonia is the involuntary adoption of abnormal postures, as agonist and antagonist
muscles both contract and become so rigid that the patient cannot maintain normal posture.

(3) Bradykinesia refers to a slowness, or poverty, of movement.

A number of well-known movement disorders are associated with basal ganglia

dysfunction. We shall concentrate on 3 of the most well-understood: Parkinson’s disease,
Huntington’s disease, and hemiballismus. To understand how these disorders result in the
specific symptoms, it is necessary to review the circuit anatomy of the basal ganglia that was
presented in the “Basal Ganglia” lecture.

65
 Parkinson’s disease

Parkinson’s disease results from the death of dopaminergic neurons in the substantia
nigra pars compacta. It is characterized by a resting tremor, but the most debilitating symptom is
severe bradykinesia or akinesia. In advanced cases, patients have difficulty in initiating
movements, although involuntary, reflexive movements can be normal. It is as if the loss of the
substantia nigra neurons has put a brake on the output of motor cortex, inhibiting voluntary
motor commands from descending to the brain stem and spinal cord and activating alpha motor
neurons.

Although the cause of Parkinson’s disease is still not known, much has been learned in
the past 15 years from the development of an animal model of Parkinson’s disease. This model
was discovered by accident when a number of young patients presented with symptoms
remarkably similar to Parkinson’s disease. These patients were all drug addicts who had been
taking an artificially manufactured drug called MPTP (1-methyl-4-phenyl-1, 2, 3 ,6-
tetrahydropyradine). This drug destroyed the dopaminergic neurons in the substantia nigra,
leading to a Parkinsonian disorder. Laboratory animals injected with MPTP have since become
a leading model for understanding the disease and developing treatments.

How does the loss of the dopaminergic neurons cause the poverty of movements
associated with Parkinson’s disease (Fig. 1)? Recall from the “Basal Ganglia” lecture that the
substantia nigra pars compacta projects to both direct pathway and indirect pathways neurons in
the striatum. Because there are two different types of dopamine receptors, substantia nigra
activity excites the direct pathway and inhibits the indirect pathway. The net effect of the direct
pathway is to excite motor cortex, and the net effect of the indirect pathway is to inhibit motor
cortex. Thus, the loss of the nigrostriatal dopaminergic pathway upsets the fine balance of
excitation and inhibition in the basal ganglia and reduces the excitation of motor cortex. In ways
that are not understood, this reduction of thalamic excitation interferes with the ability of the
motor cortex to generate commands for voluntary movement, resulting in the poverty of
movement of Parkinsonian patients. It is as if all of the motor programs stored in cortex are
constantly inhibited by the indirect pathway, with not enough excitation of the direct pathway for
the desired motor program to become activated.

There is no cure for Parkinson’s disease, but a number of effective treatments exist. The
earliest effective treatment was developed when it was first discovered that Parkinson’s disease
was caused by a loss of dopaminergic neurons. It was reasoned that replacement of the
dopamine might alleviate these symptoms. Because dopamine itself does not cross the blood-
brain barrier, L-Dopa, a chemical precursor to dopamine, was used instead. Amazingly, flooding
the system with L-Dopa resulted in profound improvements in the symptoms of patients.
Unfortunately, this improvement is temporary, and typically symptoms return after a number of
years. Surgical intervention, such as making lesions to the globus pallidus internal segment
(pallidotomy), has shown effectiveness in some patients. In recent years, a new therapy, deep
brain stimulation of the subthalamic nucleus, has been gaining in popularity. In this treatment,

66
 Figure 1. Loss of dopaminergic neurons in the substantia nigra pars compacta causes Parkinson’s
disease.

an electrical stimulator is implanted in the subthalamic nucleus. When the electrical current is
turned on to stimulate the nucleus, the patient’s symptoms disappear immediately. It is not
known why this procedure works, or what its long-term efficacy is. Because the projection from
the subthalamic nucleus is excitatory onto globus pallidus neurons, which inhibit the thalamus, it
is paradoxical that such stimulation should increase motor cortex activity. One thought is that
the stimulation might actually overload the subthalamic nucleus, thereby inhibiting it and
disinhibiting the thalamus.

Huntington’s disease

Huntington’s disease (also known as Woody Guthrie Disease) is a genetic disorder that is
caused by an abnormally large number of repeats of the nucleotide sequence CAG on
chromosome 4. Normal individuals have 9-35 repeats of this sequence; mutations that cause
larger repeats give rise to Huntington’s disease. It is an autosomal dominant mutation, such that
the offspring of a patient with Huntington’s disease has a 50% chance of inheriting the mutation.
Individuals with the mutated gene will invariably develop Huntington’s disease, usually near
middle age. The affected gene codes for a protein known as huntingtin, the function of which is
not known. The effect of the mutated version of the gene, however, is to kill the indirect
pathway neurons in the striatum, particularly those of the caudate nucleus.

Huntington’s disease is also known as Huntington’s chorea because it is characterized by

a continuous, choreiform movements of the body (especially the limbs and face). In addition, the
disease in advanced stages is associated with dementia. There is at present no cure or effective
treatment for Huntington’s disease.

Why does the loss of indirect pathway neurons in the striatum cause the dyskinesias of
Huntington’s disease (Fig. 2)? Recall that the net effect of the indirect pathway is to inhibit

67
motor cortex. With the loss of these neurons, the excitatory effect of the direct pathway is no
longer kept in check by the inhibition of the indirect pathway. Thus, the motor cortex gets too
much excitatory input from the thalamus, disrupting its normal functioning and sending
involuntary movement commands to the brain stem and spinal cord. Because inappropriate
motor programs are not inhibited normally, the cortex continuously sends involuntary commands
for movements and movement sequences to the muscles.

Figure 2. An autosomal dominant mutation causes Huntington’s disease, in which the indirect
pathway cells of the striatum are selectively destroyed.

Hemiballismus

Hemiballismus results from a unilateral lesion to the subthalamic nucleus, usually caused
by a stroke. This lesion results in ballismus on the contralateral side of the body, while the
ipsilateral side is normal (hence the term hemiballismus). The involuntary, ballistic movements
result from the loss of the excitatory subthalamic nucleus projection to the globus pallidus (Fig.
3). Because the globus pallidus internal segment normally inhibits the thalamus when excited,
the loss of the subthalamic component lessens the inhibition of the thalamus, making it more
likely to send spurious excitation to the motor cortex. Some surgical operations have been
performed to relieve the symptoms of hemiballismus, and new pharmacological treatments are in
use to relieve the disorder.

68
 Figure 3. Hemiballismus is caused by a unilateral stroke to the subthalamic nucleus.

Disorders of the cerebellum

Like the basal ganglia, the cerebellum has historically been considered part of the motor
system because damage to it produces motor disturbances. Some of these symptoms were
discussed in Chapter 6. Unlike the basal ganglia, damage to the cerebellum does not result in
lack of movement or poverty of movement. Instead, cerebellar dysfunction is characterized by a
lack of coordination of movement. Also unlike basal ganglia (and motor cortex), damage to the
cerebellum causes impairments on the ipsilateral side of the body.

(1) Ataxia is a general term used to describe the general impairments in movement
coordination and accuracy that accompany cerebellar damage. There are two major forms of
cerebellar ataxia.

A. Disturbances of posture or gait result from lesions to the vestibulocerebellum.

Patients have difficulties in maintaining posture because of the loss of the fine-control
mechanisms programmed by cerebellar circuits that translate vestibular signals into
precise, well-timed muscle contractions to counter small sways in the body. As a result,
patients often develop abnormal gait and stances to compensate. For example, the feet
are often spaced widely apart when the patient stands still, as this provides a more stable
base to maintain balance. In addition, patients display a staggering gait, with a tendency
to fall toward the side of the lesion. This gait resembles that of a drunken individual;
indeed, alcohol is known to affect the firing of Purkinje cells, which may explain the loss
of coordination that accompanies inebriation.

B. Decomposition of movement results from the loss of the cerebellum’s ability to

coordinate the activity and timing of many muscle groups to produce smooth, fluid
movements. Instead, cerebellar patient must decompose each movement into its
component parts, performing them in serial, rather than all at once in a coordinated
fashion.

69
 (2) Dysmetria refers to the inappropriate force and distance that characterizes target-
directed movements of cerebellar patients. For example, in attempting to grab a cup, they may
move their hand outward with too much force or may move it too far, with the result of knocking
over the cup instead of grabbing it.

(3) Dysdiadochokinesia refers to the inability of cerebellar patients to perform rapidly

alternating movements, such as hitting a surface alternately with the palm and back of the hand.
This diagnostic sign results from the lack of the cerebellum’s ability to coordinate the timing of
muscle groups, alternately contracting and inhibiting antagonistic muscles, to produce the
rhythmic movements.

(4) Scanning speech refers to the often staccato nature of speech of cerebellar patients.
The production of speech is a motor act, as muscles of the jaw, tongue, and larynx need to work
in unison to produce words and sounds. Cerebellar patients have difficulty in coordinating these
muscle groups appropriately, and therefore their speech tends to be slow and disjointed.

(5) Hypotonia is another symptom of cerebellar damage. There is a decreased,

pendulous myotatic reflex, as the decreased muscle resistance tends to cause the limb to swing
back and forth after the initial reflex contraction.

(6) Intention tremor refers to the increasingly oscillatory trajectory of a cerebellar

patient’s limb in a target-directed movement. For example, the hand will start out on a straight
path toward the target, but as it gets closer, the hand begins to move back and forth, and the
patient must slow down the movement and very carefully approach the target. Note that this
tremor contrasts with the resting tremor of Parkinson’s disease, which disappears when the
movement is made. Intention tremor is absent when the hand is still, but appears toward the end
of a target-directed movement.

(7) Nystagmus is an oscillatory movement of the eyes, resulting from damage to the
vestibulocerebellum. Recall that one function of the cerebellum is to fine-tune the gain of the
vestibuloocular response. Damage to the cerebellum can disrupt this circuitry, resulting in a
continuing oscillation of the eyes.

(8) Delay in initiating movements. Cerebellar patients take longer to initiate

movements, often because they must actively plan sequences of movements that are performed
effortlessly by normal individuals.

(9) In addition to movement disorders, cerebellar patients also demonstrate subtle

cognitive deficits, such as an impaired ability to estimate time intervals.

70