Professional Documents
Culture Documents
691682260 Mark L Latash Tarkeshwar Singh Neurophysiological Basis of Motor Control Human Kinetics 2023
691682260 Mark L Latash Tarkeshwar Singh Neurophysiological Basis of Motor Control Human Kinetics 2023
691682260 Mark L Latash Tarkeshwar Singh Neurophysiological Basis of Motor Control Human Kinetics 2023
Motor Control
Third Edition
Introduction
Chapter 1 History, Evolution, and Motor Control
1.1 Brief History of
Movement Studies
1.2 Evolution of
Movements and Nikolai
Bernstein’s Theory
1.3 Motor Control and
Laws of Nature
PROBLEM 1.1
What makes the soul command the body according to classics of
Greek philosophy? And what could be the origin of commands to
the soul?
Until the second century A.D., philosophers discussed only very
general features of biological movements without paying much
attention to organs within the body that were crucial for movements
to occur. The great Roman physician Galen (129-201) was arguably
the first to emphasize two important features of animal movements.
The first is the involvement of muscle pairs with opposing actions
(agonists and antagonists, for example flexors-extensors) in
voluntary movements of body segments. And the second was the
role of nerves that delivered animal spirits to muscles that led to their
force production.
The formulation of the main question of movement as that of soul-
body interaction persisted until relatively recently. A great
philosopher of the Renaissance, René Descartes (1596-1650), is
credited for introducing the basics of dualism, a branch of philosophy
that views two independent entities forming every human being, the
soul and the body (not very different from the ancient Greeks).
Following Galen, Descartes thought that the soul used animal spirits
to move the body, although some movements, for example the
beating of the heart, were viewed as independent of the soul.
Descartes also emphasized the importance of senses and the
central nervous system for some movements. Descartes and a
British anatomist Thomas Willis (1628-1678) viewed quick motor
reactions to sensory stimuli as building blocks for voluntary
movements—a view developed in the 20th century by Sherrington.
Later, Jean Astruc (1684-1766) introduced the term reflex for such
actions.
At about the same time, the science of biomechanics was born
with particularly impressive contributions by Giovanni Alfonso Borelli
(1608-1679), a disciple of Galileo. Borelli viewed muscles as elastic
structures controlled by the soul with droplets of nerve juice
delivered by nerves—a great insight into the role of chemical
processes and neuromuscular mediator (acetylcholine) in the
production of biological movements.
PROBLEM 1.2
Present examples of biological movements dependent and
independent of the soul according to Descartes.
The next level, Level B, which had two names, the level of
synergies and patterns and the thalamic-pallidar level, was
responsible for uniting numerous muscles that take part in most
functional actions, for example locomotion, into groups. Muscles
within each such group were expected to show parallel changes in
their activation levels, which is currently viewed as one of the
distinguishing features of multi-muscle synergies. Bernstein realized
the importance of movement stability in the continuously changing,
unpredictable world, and associated this function with Level B. It is
rather amazing that Bernstein linked this level to the thalamus and
globus pallidus—two structures within the brain involved in neural
loops that define movement-related output of cortical areas. As we
will see later, this insight maps well on the current understanding of
the role of the thalamus and basal ganglia (globus pallidus is part of
this system of subcortical structures) in the control and coordination
of movements.
Bernstein associated the next level, Level C, the level of the
spatial field, or the pyramidal-striatal level, with the concept of spatial
field, a portion of the external space accessible for specific actions,
with its geometry and metrics. This notion included not only motor
elements but also sensory signals relevant to that portion of the
surrounding space and its perception with the help of various
sensory modalities, including proprioception, touch, vision, and
vestibular system. The notion of spatial field is a close relative to the
notion of affordances introduced later by the great American
psychologist James Gibson (1904-1979) to describe possibilities for
action existing in the environment given the properties of the body,
location of targets, obstacles, and external forces (Gibson 1979).
The level of spatial field was viewed as responsible for actions
toward targets in the external space. Two sublevels were responsible
for identification of a target and ultimately reaching it (C2, pyramidal
sublevel) and for ensuring specific trajectories from the initial state to
the target (C1, striatal sublevel). Note that proper functioning of this
level always has to take into consideration actual and expected
forces acting from the environment. Since such expectations are
never 100% perfect, behaviors controlled by Level C are always
associated with motor variability and can be characterized with
accuracy—consider, for example, catching a ball. Bernstein
introduced one of his famous expressions with respect to actions
controlled at Level C: “Repetition without repetition.” He implied that,
when a person tried to perform the same task multiple times, the
process of solving this motor problem was repeated but actual
trajectories were not—a great insight considered in detail later.
PROBLEM 1.3
Can spatial field be described with nonspatial variables? Present
examples.
Note that each of the two presented equations involves two types
of symbols, variables (F, a, ΔF, and Δx) and parameters (m and k).
Variables are constrained by each law, while parameters are not.
Indeed, mathematically it does not matter whether you write F = ma
or F = am. However, if you apply force to an object, it would lead to
proportional acceleration, not a change in mass. The same is true for
Hooke’s law: Changing force acting on a spring will lead to its
deformation, not a change in its stiffness. This is true even for
objects that are characterized by quickly changing parameters. For
example, m remains a parameter in Newton’s second law even if it
changes quickly as, for example, in a rocket burning lots of its fuel.
Classical laws of nature described in physics textbooks are
applicable to all objects, biological and inanimate. However, there is
a qualitative difference. Inanimate objects behave in a well-
predictable way if we know all the variables and parameters of the
relevant laws of nature. Biological objects do not violate those laws,
but their behavior is not prescribed, only constrained, by them.
Indeed, animals frequently show behaviors that are not expected
from inanimate objects. Examples include running uphill and
swimming against the current. These behaviors suggest that some of
the relevant laws of nature have not been discovered yet, which
makes the field of motor control very exciting. Further, we will try to
link specific features of neurophysiological systems and circuits in
the body to possible biology-specific laws of nature.
Figure 1.2 uses the well-known law of gravity to illustrate the
difference between the two main approaches, based on the control
theory (left panel) and on laws of nature (right panel). The law of
gravity was introduced by Isaac Newton in the following form:
where FG stands for the gravity force acting between two bodies, M1
and M2 for the masses of the two bodies, R for the distance between
the bodies, and γ is a constant. Imagine now that there is a
computational device on the Sun that receives signals from sensors
on the planets informing it of their masses and distances from the
Sun. This computational device also has information on the mass of
the Sun itself. Based on those data, it computes the requisite force
using equation 1.3 and sends signals to actuators that produce the
precomputed force on the planets.
Figure 1.2 An illustration of two approaches to the origin of gravity force acting
between the Sun and a planet. (a) The Sun gets information from sensors on the
mass (m) of and distance (R) to the planet, computes planned force (F) based also
on its own mass (M), and sends a signal to hypothetical actuators that put this
force into action. (b) The Sun creates gravity field G. Any object in this field
experiences force proportional to the field and the mass of the object. γ = constant.
CHAPTER 1 IN A NUTSHELL
Philosophers and researchers have been
thinking about problems of interaction
between the human mind (soul,
intention) and body literally for
millennia. Development of new methods
of observation, measurement, and
analysis played a major role in
leading the evolution of thinking
about the interactions between the
central nervous system and the rest of
the body. Nikolai Bernstein introduced
a multilevel scheme of the
construction of movements based on the
evolutionary approach; many of his
guesses have been confirmed in later
studies. There are two approaches to
motor control. One of them is based on
concepts from control theory and
engineering; the other is based on
ideas from natural science. According
to the latter approach, motor control
is a field of study trying to discover
laws of nature that define
interactions among body structures,
including the central nervous system,
and between the body and the
environment during natural movements.
Part I
Figure 2.4 An electric field creates a difference of potentials (U), which induces a
flow of charged particles (current, I). The current is proportional to the difference of
potentials. The inverse of the coefficient of proportionality is termed resistance, R.
PROBLEM 2.2
What will happen with the cell if it is placed in a solution containing
only permeable substances?
PROBLEM 2.3
In which direction will electric current flow across a membrane if
the potential inside the membrane is higher than Veq? Solve it for
Na+ and for Cl−.
The direction of the current is defined by the potential on the
membrane, while its magnitude is defined by Ohm’s law (see
equation 2.2). So, for example, electric current due to the movement
of K+ ions will be:
CHAPTER 2 IN A NUTSHELL
Biological membranes are unique
structures that allow cells to
interact with the environment and be
separate from it. Particles in
solutions can move among compartments
under the influence of differences in
pressure, differences in
concentration, and electrical field.
Osmosis is a process of movement of
the solvent, rather than solute,
across the membrane, in order to
equilibrate concentrations of all
particles. Equilibrium potential is a
potential on a membrane that creates
an electrical force acting on charged
particles, which is equal and opposing
the force due to the difference in
particle concentrations. It is defined
by the Nernst equation.
Chapter 3
Action Potential
Figure 3.1 A membrane is separating two areas, with and without ions of Na+
and Cl−. Diffusion of these ions may occur at different speeds. As a result, a new
equilibrium will be reached with different ion concentrations to the right and to the
left, when the electric force will exactly compensate for the concentration gradient
force.
PROBLEM 3.1
Find an error (an imprecise statement) in the previous paragraph.
PROBLEM 3.2
Calculate (approximately) equilibrium potentials for Na+, K+, and
Cl−.
Figure 3.3 shows typical concentrations of the three ions K+, Na+,
and Cl− inside and outside the membrane and the corresponding
equilibrium potentials. The difference in ion concentrations inside
and outside the membrane is maintained actively and requires
energy. This mechanism is commonly called the sodium–
potassium pump. Ion pumps involve the action of proteins that
carry ions against their concentration gradient. The sodium–
potassium pump involves proteins that carry Na+ outside the
membrane and carry K+ inside the membrane. The number of carried
ions is not balanced: The protein carries two K+ ions inside for every
three Na+ ions carried outside. As a result, the pump generates a net
negative potential inside the membrane or, in other words,
contributes to its hyperpolarization. Figure 3.4 shows schematically
how the pump works, receiving energy from adenosine triphosphate
(ATP) stored in mitochondria.
Figure 3.3 The differences in the concentration of the three most important ions
across the membrane. Equilibrium potentials for each ion are shown in
parentheses.
Figure 3.4 Maintaining the ion concentration gradients across the membrane
requires energy, which is provided by a chemical process that transforms ATP
(stored in mitochondria) into ADP. This mechanism is called the sodium–potassium
pump.
Let us imagine that a number of ions, for example K+, Na+, and
Cl−, can cross a membrane through the same channels and thus are
in a competition. Membrane potential will be defined according to
equation 3.3:
PROBLEM 3.3
Why is equation 3.4 inadequate during fast changes in membrane
potential? What has not been taken into account? Why, in
equation 3.3, are members for Cl− represented differently than
those for K+ and for Na+?
Figure 3.5 If you stimulate a membrane with relatively small electrical stimuli, its
resting potential will change somewhat in response to each stimulus, and then will
return to its resting level.
Figure 3.6 An increase in the stimulation current will lead, at low values, to a
gradual increase in the deviation of the membrane potential from its resting level.
At some value of the stimulus, when it reaches a certain threshold magnitude, an
action potential will be generated. Further increase in the strength of the
stimulation will not lead to a change in the membrane response.
PROBLEM 3.4
Suggest examples of the all-or-none law from everyday life.
Figure 3.7 A thin electrode is inserted into the cell without breaking the
membrane. Now we can apply electrical current to change the membrane resting
potential either toward lower (depolarization) or higher (hyperpolarization)
magnitudes of membrane polarization.
3.3 Mechanisms of Generating an
Action Potential
First, it is important to realize that an action potential emerges
because of the dependence of membrane permeability for certain
ions upon the membrane potential.
Let us consider an example (figure 3.8). There is only one ion that
can move through special channels in a membrane. Each channel is
being guarded by a demon who sometimes falls asleep. The
probability of the demon falling asleep depends on membrane
potential so that, at rest, all the demons are awake and do not let the
ions cross the membrane. We can apply short-lasting pulses of
stimulation to change the membrane potential. A depolarizing pulse
puts some of the demons to sleep so that some ions can cross the
membrane. The more demons that are asleep, the bigger is the
current created by the ions. Note, however, that the current itself will
change the membrane potential.
Figure 3.8 A nonscientific illustration. A demon is guarding each channel for Na+
in the membrane. Membrane depolarization makes some of the demons fall
asleep, so that their channels become open. Ions will cross the membrane and will
increase the depolarization, putting more demons to sleep.
Figure 3.9 A positive feedback process where (a) leads to a rapid amplification
of the effect, while a negative feedback process (b) quickly restores the original
state.
Figure 3.11 After a stimulus (St1) leading to an increase in gNa, another stimulus
is less able to turn it on for some time. For a short period of time this inactivation is
absolute—that is, gNa will not respond even to a very strong stimulus (St2, absolute
refractory period). Then, a stronger-than-usual stimulus can turn gNa on (St3,
relative refractory period).
So, the channels can close in response to a change of membrane
potential and also spontaneously, when they require some time to
recover. In the first case, you can open the channels with an external
depolarizing stimulus; in the second case, the only available method
is to wait.
Figure 3.10 also shows the dependence of potassium
conductance (gK) on membrane potential. Note that gK starts to
increase with depolarization but does not turn off spontaneously. It
goes down to its original value only when the membrane potential
returns to its resting level. This means that there is no channel
inactivation, and no refractory period for potassium channels. Note
also that gK increases more slowly than gNa, which means that early
in the process of membrane depolarization, the open sodium
channels will play a bigger role.
As can be seen from the figures already shown, both gNa and gK
behave “smoothly” with membrane voltage (i.e., they do not show
any threshold effects). In order to understand the mechanisms giving
rise to the all-or-none action potential, we need to remove the
voltage clamp and allow the potential to change.
Note that opening channels for sodium and for potassium leads to
different consequences for the membrane potential because of the
difference in the concentrations of Na+ and K+ ions inside and
outside the cell. An increase in gK, for example, induced by a short
depolarizing pulse, leads to a flow of K+ out of the cell. The loss of
positive ions leads to a drop in the membrane potential (remember,
membrane potential is measured inside the cell with respect to the
outside!)—that is, to a decrease in depolarization or to
hyperpolarization. This, in turn, will lead to a drop in gK. So we have
a system with a negative feedback that will quickly restore the
original resting potential. An increase in gNa, however, will lead to an
inflow of Na+ inside the cell (i.e., to further depolarization). Here we
deal with a system with a positive feedback which, as is well known,
loves to go berserk. Different dependencies of gNa and gK on
membrane potential, together with the property of sodium channels
to inactivate, lead to the generation of the action potential.
Remember that the direction of flow of an ion depends on its
equilibrium potential such that the difference between actual
membrane potential and the equilibrium potential of an ion defines
the direction in which the ion will flow. On the other hand, an ion with
a higher permeability plays a bigger role in defining the overall
membrane potential than ions with smaller permeabilities. This
means that changes in gNa and gK can lead to changes in the resting
membrane potential.
Figure 3.12 shows an action potential and the changes in gNa and
gK in different phases of the potential. The sequence of the events is
as follows:
1. The initial depolarization (created by an external stimulus)
increases gNa so that the membrane potential tries to reach
sodium equilibrium potential.
2. The sodium channels are quickly inactivated, leading to a drop
in gNa; as a result, the membrane potential has no time to
reach the sodium equilibrium potential.
3. After a brief delay, gK increases and draws membrane
potential to its equilibrium potential (i.e., repolarizes the
membrane).
4. There is a rather long period of hyperpolarization (the
afterpotential) after which membrane potential returns to its
resting value.
Figure 3.12 Changes in Na+ and K+ conductance during an action potential.
Note that the peak of the action potential is positive, and after the action potential
the membrane remains hyperpolarized for some time.
PROBLEM 3.5
Why does the membrane potential drop below the resting level?
Can you imagine a situation in which the afterpotential would be
higher than the resting potential?
CHAPTER 3 IN A NUTSHELL
The action potential is the unit of
information transmission within the
bodies of higher animals. Membrane
potential is created by a small number
of unbalanced ions. Movement of ions
through the membrane occurs at special
sites called ion channels. An active
molecular mechanism, a sodium–
potassium pump, maintains the
difference in the concentrations of
sodium and potassium ions across the
membrane. The dependence of the sodium
ions’ conductance on the membrane
potential leads to the generation of
an action potential when membrane
depolarization reaches its threshold.
After an action potential, the
membrane stays in a short-lasting
state of insensitivity due to the
inactivation of sodium channels.
Chapter 4
Figure 4.1 Action potential travels along a neural fiber. For calculation of the
velocity of its propagation (V), the action potential has to be recorded at different
times (t1 and t2) at different coordinates (s1 and s2).
PROBLEM 4.1
If a neural fiber is stimulated strongly at some point in the middle,
in what direction will the action potential propagate?
PROBLEM 4.2
What would happen in response to a single strong stimulus
applied to the membrane of a long neural fiber if there were no
sodium inactivation mechanism?
PROBLEM 4.3
What will happen if a myelinated fiber suddenly loses its sheath?
What can you expect from such a fiber in a hot bath and in a cold
bath? Note that ion diffusion proceeds much more quickly at high
temperatures.
Table 4.1 compares the velocities of different processes. Note that
speeds of conduction in our body are rather high, but not extremely
high. Certainly, they are not comparable to the speed of light, which
is equal to the speed of propagation of an electromagnetic field.
PROBLEM 4.4
The nature of an action potential is electric. Why is the speed of
its propagation so much lower that the speed of electric events
like electric current?
PROBLEM 4.5
The speed of neural conduction is comparable to the highest
movement velocities observed in athletes. Does this mean that
there is an upper limit for movement velocity set by action
potential conduction speed?
The axon is typically a long, rather thick branch that carries the
output signals generated by the cell. At its end, the axon splits into a
whole bunch of smaller, thin branches (terminal branches) that make
contacts with other cells and transduce information to these cells.
These branches are commonly much shorter than the main part of
the axon. Axons can be very long, up to 1 meter, as in the case of
the axon of a motoneuron with its soma located in the spinal cord
that sends signals to a muscle in a foot. There are long axons of
neurons within the central nervous system as well, including axons
of neurons in the cortex of the large hemispheres that send their
signals to neurons in the lower parts of the spinal cord. The place
where the axon exits the soma is called the axon hillock. At this
place, the density of sodium channels is very high, and so this is the
place where action potentials are typically being generated. The
axons of groups of neural cells, typically united by a functional or
anatomical feature, commonly run together over relatively large
distances. In such cases, the groups of axons are addressed as
neural tracts (if they travel from one place to another within the
central nervous system) or as nerves (if they connect the central
nervous system with peripheral structures such as muscles and
sensory organs).
Dendrites form a tree around the soma and serve as sites of
inputs into the cell. Terminal branches of the axons of other cells
make connections (synapses) on the dendrites as well as on the
soma itself.
To summarize, dendrites and soma serve as sites where
information comes to the neuron from other neurons and is
integrated (assessed, compared, and put together); the axon hillock
is the place where action potentials are generated in response to the
incoming information, and the axon serves to conduct action
potentials to distant places and to transmit information to other cells.
Dendrites’ properties have attracted much attention in relation to
the ability of the dendrite membranes to show steady depolarization.
In other words, dendrites have been shown both theoretically
(Gutman 1991) and experimentally (Schwindt and Crill 1977, 1981;
Heckman et al. 2003, 2005) to have long-lasting changes in the
membrane potential induced by persistent inward currents. A
persistent inward current is a depolarizing current produced by
voltage-sensitive channels that do not show the phenomenon of
inactivation (this is why they can be long-lasting or persistent). These
channels on dendrites are specialized for Ca++ ions. Effectively,
persistent inward currents may be seen as the means of reducing
the threshold for action potential generation and, hence, reducing the
distance from the membrane resting potential to this threshold. So
these currents seem to facilitate the process of generating action
potentials. On the other hand, persistent inward currents tend to
sharply limit the efficacy of additional synaptic inputs into the neuron,
so their overall effects on the excitability of the neuron may not be
easily predictable.
Figure 4.8 illustrates the dependence between the voltage and
current over a membrane in the absence of persistent inward
currents (thin dashed curve) and when these currents are present
(thick solid curve). Note that when the current is zero, the
membrane, by definition, is at its resting potential. The thin line
shows only one value of the membrane potential when the current is
zero (i.e., only one value of the resting potential). The thick line
shows three crossings of the abscissa axis. The first and third
crossings are stable, while the intermediate one is not. An amazing
feature of the second resting potential (point 3 in figure 4.8) is that it
can be above the threshold for the generation of an action potential.
If the resting potential of a dendrite membrane is at that second
state, the dendrite starts generating action potentials and would
continue doing so without any external stimuli as long as the
membrane potential stays above the threshold value and the neuron
does not run out of energy. We will discuss the implications of this
phenomenon for voluntary movements in later chapters.
Figure 4.8 Current–voltage characteristics of a dendrite membrane in the
absence (thin lines) and in the presence (thick lines) of persistent inward currents
(PIC). In the second case, potentially there may be two stable values of the
membrane potential (points 1 and 3). Point 3 may be over the membrane threshold
for the generation of action potentials.
PROBLEM 4.6
What is the minimum frequency of presynaptic action potentials
that can theoretically lead to temporal summation and to a
postsynaptic action potential?
Figure 4.12 Spatial summation occurs when several action potentials AP1, AP2,
and AP3 arrive simultaneously at different synapses on the same presynaptic
membrane so that their individual EPSPs sum up and can induce an action
potential. The insert drawing shows possible locations of the three synapses on
the target neuron.
PROBLEM 4.7
What will happen if action potentials to synapses 1, 2, and 3 come
not simultaneously but at a delay?
CHAPTER 4 IN A NUTSHELL
Passive spread of local currents from
an action potential leads to
depolarization of adjacent segments of
the membrane to the threshold and to
the generation of a new action
potential. Speed of conduction is
higher along thicker neural fibers.
Some fibers are covered with a special
substance, myelin, which increases the
speed of conduction of action
potentials. Information exchange among
cells occurs at special sites called
synapses. The mechanism of synaptic
transmission involves special chemical
substances, mediators that can
depolarize or hyperpolarize the
postsynaptic membrane. Neural cells
integrate the incoming information and
generate action potentials when the
effects of several synapses or several
action potentials coming at a high
rate are summed up.
Chapter 5
Skeletal Muscle
5.2 Myofilaments
Myofilaments are major force-producing elements of muscle cells,
consisting of two major molecules, myosin and actin (figure 5.2).
Thick filaments contain mostly myosin, while thin filaments contain
mostly actin. Actually, thin fibers contain two actin molecules that
form a structure that looks like a double helix (resembling the famous
DNA structure). In order to develop force, the two sets of filaments,
actin and myosin, must make connections to each other. These
connections are called cross-bridges. Actin and myosin filaments
are organized so that regularly arrayed myosin filaments are
surrounded by six actin filaments (like a kitchen floor mosaic), while
each actin molecule is in contact with three molecules of myosin. At
one end, actin filaments are attached to a structure called a Z-line.
Two sets of actin filaments and one set of myosin filaments between
two Z-lines constitute a sarcomere. Sarcomeres are the most
important functional units of myofibrils in the production of muscle
force. In order to characterize the state of a myofibril, two more
terms are used: The length of the myosin filaments within a
sarcomere is called the A-band, while the length of actin filaments
that do not overlap with myosin filaments is called the I-band. These
bands are clearly seen under a strong microscope as alternating
dark (A-band) and light (I-band) zones (figure 5.2).
Figure 5.2 The structure of a myofibril. The lower figure shows the sequence of
dark and light bands (A-bands and I-bands). The upper drawing shows the typical
configuration of actin and myosin molecules within a myofibril.
PROBLEM 5.1
Imagine that you have a muscle without acetylcholinesterase.
What can you expect in response to a single presynaptic action
potential?
PROBLEM 5.2
A presynaptic action potential induces contraction of 20 myofibrils.
Can a sequence of action potentials induce the contraction of
more than 20, less than 20, or exactly 20 myofibrils?
Figure 5.6 Muscle action potential travels along the sarcolemma, enters T-
tubules, and leads to a release of Ca++ ions from the sarcoplasmic reticulum.
Figure 5.7 The sliding filament theory. Ca++ ions remove troponin and free a site
for myosin to bind to actin (this process uses the energy from ATP). A ratchet
motion occurs, moving the filaments with respect to each other.
PROBLEM 5.3
What would happen if the troponin–tropomyosin complex were
permanently inactivated?
PROBLEM 5.5
Why does the twitch contraction outlast the action potential?
PROBLEM 5.6
Smooth tetanus is rarely observed in individual fibers in real life.
Why are our muscle contractions normally smooth?
Figure 5.11 A simple mechanical model of a muscle. It contains a force
generator (F), a damping element (B), and two elastic elements, a parallel spring
(K1) and a series spring (K2).
PROBLEM 5.7
The last statement is not 100% correct. Why?
PROBLEM 5.8
In which direction will the curve shift if you increase either the
frequency or the amplitude of the stimulation?
One can measure the force–length relationship in a single
sarcomere (i.e., inside the contractile element shown as F in figure
5.11). The active force developed by the sarcomere will show a
dependence on the sarcomere length that is somewhat similar to the
one for the whole muscle: At low values of sarcomere length, cross-
bridges cannot develop force because of the lack of space for new
attachments; at intermediate values of length, the force is maximal;
while at high values of sarcomere length, there are only a few cross-
bridges that can generate force, and it drops again.
Another important relation for a whole muscle is the force–velocity
curve. Such curves are usually studied in experiments when a
muscle performs a twitch contraction under different loads and the
peak velocity of muscle shortening is measured. The curve typically
looks hyperbolic (figure 5.13) and can be well approximated with the
famous Hill equation:
Figure 5.13 A typical force–velocity curve for a whole muscle. The x-axis
represents velocity of muscle stretch (frequently, this curve is drawn using velocity
of muscle shortening). Note that the muscle develops higher forces when it is
lengthening (positive velocity) than when it is shortening (negative velocity).
Compare this figure with the Hill equation.
Figure 5.14 A muscle always works against a load. Three types of loads are
illustrated. Isometric load prevents changes in the length of the “muscle plus
tendon” complex; an isotonic load does not change; an elastic load acts like a
spring. A typical muscle characteristic is shown for comparison (the thin curve).
PROBLEM 5.9
What will happen with the relative length of muscle fibers and with
the tendon under activation in isometric conditions?
CHAPTER 5 IN A NUTSHELL
Muscle contractions are produced by an
interaction of two types of molecules,
actin and myosin, inside muscle cells.
Muscle cells are excited through
neuromuscular synapses with the help
of a mediator, acetylcholine. Action
potentials lead to the release of
calcium ions, which make cross-bridge
formation between actin and myosin
molecules possible. In response to a
single stimulus, muscle fibers
generate a single twitch contraction.
A number of stimuli coming at a high
frequency lead to the summation of
individual twitches and the generation
of a tetanic contraction. Muscle force
increases with muscle length and
decreases with the velocity of muscle
shortening. Muscles always act against
external loads, typical examples being
isometric, isotonic, and elastic
loads.
Chapter 6
Peripheral Receptors
PROBLEM 6.1
Give an example of a receptor system that does not obey the
Weber-Fechner law.
Figure 6.1 The body of a sensory neuron is located in a ganglion near the spinal
cord. One branch of its T-shaped axon goes to the peripheral sensory ending,
while the other branch goes through the dorsal roots into the spinal cord.
PROBLEM 6.2
What can you say about the state of the muscle (its length and
velocity) if you know the frequency of firing of a primary ending in
this muscle?
PROBLEM 6.3
Draw a graph of time changes in the firing frequency for a primary
ending and for a secondary ending if muscle length changes as a
sine function.
PROBLEM 6.4
Draw a graph of length changes for a muscle whose typical
primary spindle ending shows a smooth, ramp-like increase in the
firing frequency with time from a certain steady level to another,
higher steady level followed by a ramp-like decline in the firing
frequency to a level somewhat higher than the original one.
Endings in muscle spindles have very high sensitivity to low-
amplitude changes in muscle length, particularly if these changes
occur at a high frequency. This is particularly true for primary spindle
endings, which can be forced to generate an action potential or even
several action potentials in response to every cycle of high-
frequency vibration (on the order of 100 Hz) when the amplitude of
the vibration is 1 mm and the vibration is applied to the skin over the
muscle belly or over the tendon. If a vibrator is attached directly to
muscle fibers, a few μm of vibration amplitude are enough to drive
primary spindle endings at the frequency of the vibration.
PROBLEM 6.5
How will a secondary ending react to an increase in the activity of
dynamic γ-motoneurons innervating the spindle?
PROBLEM 6.6
When we voluntarily contract a muscle, its length decreases.
However, the frequency of firing of the spindle endings in the
muscle may remain constant. How can this happen?
Figure 6.7 Golgi tendon organs are located in series with extrafusal muscle
fibers at their junction with the tendon. They are innervated with fast-conducting Ib-
group axons of sensory neurons in spinal ganglia.
PROBLEM 6.7
Imagine that you prevent movement of your right elbow joint and
then quickly activate the biceps to a rather large force (isometric
conditions). Draw a graph of the change in the firing frequency for
a primary and secondary muscle spindle ending and for a Golgi
tendon organ. Assume that all endings showed a steady-state
firing level before the force increase.
PROBLEM 6.8
Now, do the same for a very fast elbow flexion movement against
a constant external force.
CHAPTER 6 IN A NUTSHELL
Receptors are specialized cells or
parts of cells that can respond to
external stimuli of certain types.
Signals from receptors lead to a
perception of stimuli that is related
to the strength of the stimuli by a
nonlinear law. Proprioceptors produce
information about the relative
configuration of body segments. Muscle
spindles contain sensory endings of
two types, sensitive to muscle length
and sensitive to both muscle length
and velocity. The sensitivity of
spindle endings is modulated by a
special system of neurons, the
fusimotor or gamma-system. Golgi
tendon organs are sensitive to muscle
force. Articular receptors are
sensitive to both joint angle
(typically, close to the anatomical
limits of joint rotation) and joint
capsule tension. Skin and subcutaneous
receptors measure pressure on the
skin. Sensory endings send signals
along the peripheral branch of the T-
shaped axon to spinal ganglia, where
the bodies of sensory neurons are
located. Then, signals travel along
the central branch of the T-shaped
axon into the central nervous system.
Chapter 7
PROBLEM 7.1
Can one motor unit produce different levels of muscle force?
Why?
PROBLEM 7.2
You want a large motor unit and a small motor unit to contract
simultaneously. How would you time neural commands to the two
motoneurons?
There are two basic tests that are used to describe the functional
properties of motor units. One is twitch contraction, while the other
is fatigue. Figure 7.2a shows twitch contractions of three motor
units. Note that these motor units generate different levels of peak
force and take different times to reach the peak force level. In
particular, motor unit 3 takes the longest time for its twitch
contraction and generates the lowest peak force, while motor unit 1
is the first to reach the peak force level and has the highest peak
force. If the same motor units are stimulated at a rather high
frequency, they produce tetanic contractions. This can be achieved
by stimulating the axons of the motor units with an external electrical
stimulator. If the frequency of the stimulation is the same for all three
motor units, the peak force will once again be the highest for motor
unit 1 and the lowest for motor unit 3 (figure 7.2b). If the stimulation
continues for a rather long time, changes in the level of the
contraction force will be observed (figure 7.2b), induced by fatigue.
The mechanisms of fatigue will be discussed in chapter 30. Here, it
is important to note that the changes in the force levels of motor unit
2 and motor unit 3 are small, while the force level of motor unit 1
drops significantly.
Thus, there seem to be three types of motor units. Motor units of
the motor unit 1 type are typically called fast-twitch fatigable, motor
units of the motor unit 2 type are called fast-twitch fatigue resistant,
and motor units of the motor unit 3 type are called slow-twitch fatigue
resistant. These groups are sometimes referred to as FF, FR, and S
motor units. Slow motor units (S) typically have fewer muscle fibers,
smaller motoneurons, and thinner axons. Correspondingly, the
speed of conduction of action potentials along the axons of slow
motor units is the lowest (although it is rather high since, as you
remember, the axons of α-motoneurons are thick, myelinated, group
I neural fibers). FF motor units are characterized by the highest
conduction velocity. The difference in conduction velocity may be
more than twofold (from 40 m/s in some S motor units to 100 m/s in
some FF motor units).
The differences in the physical properties of motor units correlate
with their different biochemical and morphological characteristics.
Consider three major sources of energy used for muscle contraction.
The first is ATP contained in myofibrils; the importance of this source
may be assessed by the level of activity of the enzyme ATPase that
participates in metabolizing ATP. The second source is oxidative
metabolism occurring in mitochondria. Its rate may be assessed by
the activity of a couple of enzymes, succinic dehydrogenase and
NADH dehydrogenase. The third source is glycogen, whose
metabolism is anaerobic. Table 7.1 shows the three major types of
motor units and the relative representation of different physiological
and biochemical characteristics of their muscle fibers. The table
shows that S motor units contain mostly slow, oxidative fibers
characterized by a high level of mitochondrial oxidative processes
and a well-developed blood supply network. Fast motor units use
more energy from ATP and glycogen metabolism. FR motor units
have a rich capillary supply comparable to that of S motor units,
while FF motor units have a sparse capillary supply reflected in the
color of muscles with a high percentage of FF motor units: these are
pale muscles.
Figure 7.4 Muscle force is kept constant. A change in the number of recruited
motor units correlates (negatively!) with their mean frequency of firing.
PROBLEM 7.3
Formulate the size principle for the order of motor unit
involvement when the contraction is induced by progressively
increasing the strength of electrical stimulation of the muscle
nerve.
There are situations when the Henneman principle does not work
perfectly, although these are relatively rare. In particular, if a muscle
participates in a task where it is not the primary mover (for example,
it participates in a postural task component), the order of motor unit
recruitment within this muscle may change, leading to a violation of
the size principle for certain pairs of motor units: A larger motor unit
may be recruited before a smaller one. A reversal of the size
principle can also be seen in certain reflex responses, in particular in
responses to cutaneous stimulation.
PROBLEM 7.4
Which motor units would you expect to find in abundance in a
marathon runner, in a weightlifter, and in a swimmer?
PROBLEM 7.5
You have invented a way to induce abrupt synchronization of
motor units in human muscles. What groups of athletes would you
recommend this method to, and what groups of athletes would
you suggest not even try it?
7.5 Electromyography
There are two basic methods of recording muscle activity:
Intramuscular or needle electromyography and surface or
interferential electromyography.
In the first method, a thin needle (with a diameter of less than 1
mm) is inserted into a muscle (figure 7.6). Inside the needle, there is
a very thin wire that is electrically isolated from the needle. The tip of
the wire is not isolated. An amplifier picks up the difference of
potentials between the tip of the wire and the needle. Since the
dimensions of each electrode and the distance between the two
electrodes are very small, the electrodes selectively pick up signals
(action potentials) in the closest proximity to the tip of the wire. Such
electrodes are designed to record the patterns of activity of individual
motor units. Note that each motor unit contains many muscle fibers,
but they all generate action potentials synchronously, so that the
electrode picks up the compound action potential of the whole motor
unit. Typically, a needle electrode can record the electrical activity of
a few motor units whose muscle fibers happen to be in close
proximity to the electrode. However, because each motor unit has a
somewhat different number of muscle fibers and also because the
location and orientation of these fibers with respect to the electrode
are different, each motor unit has a different, unique pattern of
voltage changes, a unique signature (figure 7.7). These differences
make it possible to record several motor units with one electrode and
identify their compound action potentials with a high degree of
certainty. Needle electromyography is frequently used in clinical
tests.
Figure 7.6 Intramuscular electromyography uses thin needle electrodes. Inside
the needle, there is a very thin wire that is electrically isolated from the needle. The
difference of potentials (Δφ) between the tip of the wire and the tip of the needle is
amplified and recorded.
7.6 Processing
Electromyographic Signals
It is impossible to recommend a universal method to record and
process electromyograms. There are several standard types of
procedures; however, each researcher selects his or her own
methods of data processing based on the actual goals of the study
and the researcher’s own imagination. Three operations are
frequently used in processing a surface electromyogram.
The first is filtering. Note that action potentials are very fast events
with typical times of potential changes of about a few milliseconds.
So a high-pass filter is frequently used, which cuts off all the
frequencies equal to or below 60 Hz. As a result, the 60 Hz noise is
reduced as well as possible reactions of the electrodes to purely
mechanical factors that are usually much slower than changes in
biological potentials. On the other hand, the upper limit of filtering
frequency is chosen based on the characteristic times of events that
are of interest for the experimenter. If the experimenter is not
interested in the microstructure of the electromyographic signals
such as the shapes of individual action potentials, a low-pass cut-off
frequency is commonly chosen on the order of a few hundred hertz.
The second operation is rectification. The purpose of this
procedure is to be able to get a quantitative estimate of an
electromyographic signal. If an action potential runs along a muscle
fiber under a pair of recording electrodes (figure 7.10), the difference
of potentials at the electrodes will change gradually, leading to a
reversal in its sign. Actually, many biopotentials show a nearly
symmetrical picture with respect to zero level. Integrating an
unrectified signal over a reasonably long time will yield a very small
number (close to zero) because the signal is composed of an
approximately equal number of positive and negative values.
Integrating a rectified electromyogram will result in a value reflecting
the average magnitude of the activity over the time of integration.
Figure 7.10 An action potential runs under a pair of electrodes. The difference of
potentials recorded by the electrodes will change its sign (the upper record).
Rectification means making all the values of the difference of potentials positive
(the lower record).
There are two types of rectification. Full-wave rectification consists
of turning all the negative values of the difference of potentials into
positive values of equal magnitude (figure 7.10). Half-wave
rectification cuts off all the negative values and substitutes them with
zeros. Full-wave rectification is used more frequently. However,
recent analyses have shown that using this method may lead to
misleading outcomes, in particular in the identification of the timing of
bursts of muscle activation (Farina et al. 2004).
PROBLEM 7.6
Imagine that you have an EMG record. You can filter it and then
rectify it or, alternatively, you can rectify it and then filter. Which
method is better? Why?
PROBLEM 7.7
Suggest methods to normalize EMG signals during very fast
movements and during very small changes in the level of muscle
activity.
CHAPTER 7 IN A NUTSHELL
A motor unit is a motoneuron and all
the muscle fibers innervated by its
axon. There are three major types of
motor units: slow, fatigue resistant;
fast, fatigue resistant; and fast,
fatigable. Slow motor units contain
neurons with a smaller body, thinner
axon, slower conduction velocity of
action potentials, and fewer
innervated muscle fibers. During
natural muscle contractions, motor
units are recruited in a fixed order,
from the smallest to the largest (the
size or Henneman principle).
Derecruitment follows the opposite
order, from the largest to the
smallest. Electromyography is a method
of studying muscle activation levels
and patterns that requires learning
many nuances to achieve accurate
results.
Problems for Part I
Self-Test Problems
1. You have an atypical neural cell in which the concentration of
K+ ions inside the cell is not 150 mmol/L, but only 50 mmol/L.
Everything else is exactly as in “regular” cells. Calculate the
equilibrium potential for K+. How will the equilibrium potential
on the membrane and the action potential in this cell differ
from those in “regular” neurons?
2. You observe two fibers under changes in temperature. A
stimulator is placed at one end of each fiber, and you record
the response at the other end. One fiber decreased the speed
of transmission of action potentials with a decrease in
temperature and eventually stopped transmitting them. The
other fiber did not transmit action potentials at higher
temperatures, started to transmit them at lower temperatures,
and stopped transmitting them at very low temperatures. What
can you conclude about these two fibers? Explain the
differences in their behavior.
3. You have a neural cell with one excitatory and one inhibitory
synapse. At a certain frequency of stimulation of both
presynaptic fibers, the neuron does not generate an action
potential. You increase the frequency of stimulation of the
inhibitory input without changing the frequency of stimulation
of the excitatory input. After some time, the neuron starts to
generate action potentials. Why? In another experiment, you
increase the frequency of stimulation of the excitatory input.
The neuron generates several action potentials and then
becomes silent. Why?
4. You study the response of a neural cell to a single excitatory
input. The cell generates action potentials at a certain
frequency. You add an excitatory neurotransmitter to the
extracellular space and the cell stops firing. What happened?
5. You induce a twitch contraction of a muscle by a direct
electrical stimulus. The external load is zero. Draw time
changes in the frequency of firing of a primary spindle ending,
of a secondary spindle ending, and of a Golgi tendon organ.
Prior to the contraction, each receptor showed steady firing at
a constant frequency. Solve the same problem for isometric
conditions, that is, when the “muscle + tendon” complex
cannot change its length.
6. A person generates 5% of the maximal voluntary contraction
force of a muscle. Then muscle force increases slightly so that
only one new motor unit is recruited. What can you say about
the properties of this motor unit? The same person generates
95% of the maximal voluntary contraction force. Again, muscle
force increases slightly so that only one new motor unit is
recruited. What can you say about the properties of this motor
unit?
Neuroanatomical Foundations of
Motor Control
Chapter 8
Cerebral Cortex
Figure 8.2 The dorsal visual stream facilitates visuomotor function. This stream
is further divided into the dorsomedial and dorsolateral streams. The dorsomedial
stream facilitates reaching movements, and the dorsolateral stream facilitates
grasping movements. Both of these streams project to different parts of the
premotor cortex. The ventral visual stream subserves visual perceptual functions.
The dorsal stream is further divided into the dorsomedial and
dorsolateral streams, where the former facilitates reaching
movements and the latter grasping movements. The dorsomedial
and dorsolateral streams project to the dorsal premotor (PMd) and
the ventral premotor (PMv) cortices, respectively. These premotor
areas then project to different areas in the primary motor cortex (M1).
Some motor neurons in M1 project directly to the spinal cord, and
others project indirectly through the brainstem to activate the
interneurons and α-motoneurons in the spinal cord that activate the
limb muscles to move the arm and fingers. As stated earlier, we will
focus only on the structure and functions of M1, PMd, PMv, and SMA
in this chapter. We will discuss the dorsal and ventral visual streams
in more detail in chapter 14.
PROBLEM 8.1
During a reach-to-grasp movement, the grip aperture gradually
scales as a function of the remaining distance between the hand
and the object. What does that tell us about how visual sensory
information is integrated during a limb movement?
Figure 8.3 The numbers shown in the map indicate Brodmann’s areas.
Brodmann assigned these numbers to the brain regions based on their
cytoarchitectonic structure.
Figure 8.5 The frontal cortex areas involved in motor planning. The premotor
cortex (dorsal and ventral) and SMA receive extensive projections from the parietal
cortex and other brain areas. The pre-SMA receives strong projections from the
prefrontal cortex and is considered to be involved in cognitive motor control. The
motor planning areas are connected to the primary motor cortex (M1).
PROBLEM 8.2
While driving a car in the countryside, you decide to increase the
speed of the car by 5 mph. Which frontal premotor area will be
primarily responsible for facilitating the change in the car’s speed?
At another time during the drive, you see that cars ahead are
moving slowly, and many of them have their emergency flashers
on. You then gradually slow down the car. Which premotor area
will be primarily responsible for this action?
PROBLEM 8.3
How can you interpret the fact that stimulation of the premotor
areas requires higher currents to induce visible muscle
contractions than does stimulation of the primary motor cortex?
Figure 8.7 About 90% of the axons of the corticospinal tract decussate and form
the lateral corticospinal tract. They innervate α-motoneurons of the distal muscles.
The remaining 10% of the axons form the ventral corticospinal tract and innervate
trunk and proximal limb muscles.
PROBLEM 8.4
Does the last finding prove that pyramidal neurons control muscle
force? Why?
Figure 8.8 The projections from the cerebellum and the basal ganglia to the
frontal motor areas, premotor cortex (PM), and primary motor cortex (M1) go
through the thalamus.
The parietal cortex and cortical inputs are a major source of input
into the cortical motor areas. The parietal cortex integrates
multisensory information from the somatosensory system
(proprioception and tactile) and the visual and vestibular systems
and provides this information to the cortical motor areas for motor
planning and execution.
The premotor and supplementary areas also receive input from
the prefrontal cortex. The prefrontal cortex is involved in higher-order
executive functions, such as representation of abstract and complex
rules, short-term working memory, and decision-making. The
connections between the prefrontal and premotor areas indicate that
the output of the prefrontal cortex targets specific areas of the
premotor cortex involved in motor control (Lu et al. 1994). The
prefrontal cortex also shares strong connections with the pre-SMA,
which suggests that the pre-SMA plays an important role in cognitive
motor control.
PROBLEM 8.5
Suggest an explanation for the difference between the typically
short reaction times (commonly under 200 ms) and the rather long
readiness potential.
Figure 8.10 (a) The figure shows a raster plot of a single neuron during arm
movements in eight directions. Here 0 on the x-axis indicates movement onset.
Each vertical line on the raster plot indicates an action potential. The activity of this
particular neuron increased when a movement was made between 90º and 225º
and decreased below baseline levels when the movement was made between 45º
and 315º. (b) The firing frequency of a cortical neuron demonstrates a cosine-like
dependence on the direction of voluntary movement. For this particular neuron, the
preferred direction is 135º.
Adapted from Georgopoulous (1982).
Figure 8.12 A brain-machine for upper limb motor function. The patient is asked
to imagine making movements of the paralyzed arm. The neural activity is
recorded with electrodes (often implanted in the brain). Then those signals are
processed and decoded. The decoded information is then used to generate control
signals to drive the motors of the prosthetic device.
Reprinted by permission from E. López-Larraz, A. Sarasola-Sanz, N. Irastorza-Landa, N.
Birbaumer, and A. Ramos-Murguialday, “Brain-Machine Interfaces for Rehabilitation in
Stroke: A Review,” NeuroRehabilitation 43, no. 1 (2018), with permission from IOS Press.
The publication is available at IOS Press through 10.3233/NRE-172394.
CHAPTER 8 IN A NUTSHELL
The motor areas of the brain are
involved in both the planning and
execution of actions. The primary
motor cortex (M1) provides the most
input to the corticospinal tract that
is involved in the control of fine
upper-limb arm and hand functions. M1
receives input from the premotor areas
of the brain. The premotor areas are
strongly interconnected with the
parietal and prefrontal cortices. The
premotor areas are involved in the
transformation of sensory input to
parameters of limb motor control.
These areas also project to the
corticospinal tract. The premotor
cortex is divided into the dorsal and
ventral areas, which are both involved
in movements initiated by sensory
signals. The main role of the dorsal
premotor cortex is integration of
sensory information into motor
commands and specification of reaching
movement amplitude, direction, and
speed. In contrast, the ventral
premotor cortex is involved in
preshaping of fingers for grasping
movements. It also contains mirror
neurons that respond to action
observation. The supplementary motor
area is also an important region that
abuts the premotor cortex. It receives
extensive input from the prefrontal
regions and is involved in intentional
movements.
Chapter 9
Basal Ganglia
For many years, the basal ganglia have been viewed as a very
important set of brain nuclei for motor function. In particular, the
multilevel hierarchical scheme for the neural control of movement
introduced by Nikolai Bernstein (1947) involved two levels with
crucially important contributions from structures within the basal
ganglia. These were the level of synergies and the level of spatial
field. Bernstein postulated an important role for the basal ganglia in
forming task-specific groups of motor elements and ensuring the
dynamic stability of natural movement (done at the level of
synergies) and in the production of movements to spatial targets
(controlled at the level of spatial field).
The basal ganglia consist of several large subcortical nuclei: the
putamen and caudate nucleus—which are addressed together as
the striatum—the globus pallidus, the substantia nigra, and the
subthalamic nucleus. These nuclei neither receive direct inputs from
nor send direct outputs to the α-motoneurons in the spinal cord. The
importance of the basal ganglia for control of voluntary movements
has been assumed based mostly on clinical observations. Basal
ganglia disorders bring about quite different clinical pictures ranging
from excessive involuntary movements to slowness and a lack of
movement. Because of these clinical observations, it was supposed
that the basal ganglia were major components of an extrapyramidal
system that was thought to participate in the control of movements in
parallel with and largely independent of the pyramidal (corticospinal)
system. However, we now know that the pyramidal and
extrapyramidal systems are not independent but dynamically
cooperate for movement control. Basal ganglia also play an
important role in motor learning and habit formation. The function of
the basal ganglia is not limited to controlling movements; these
structures have a role in cognitive and emotional functions as well.
Figure 9.1 A sagittal and frontal view of the brain showing the basal ganglia and
the two input nuclei of the basal ganglia, the caudate nucleus and the putamen.
These two structures constitute the striatum. The striatum is the major input
structure of the basal ganglia. One of the major output structures of the basal
ganglia is the globus pallidus.
Figure 9.2 (left) The main cortical structures that project to the basal ganglia and
the input nuclei of the basal ganglia, the caudate and the putamen. (right) The
internal projections within the basal ganglia.
PROBLEM 9.1
Would putamen neurons fire as a person is reaching for an ice
cream cone, or in anticipation before the reaching movement is
initiated? Why?
PROBLEM 9.2
If a baseball batter decides not to swing for a curveball, how
would the direct and indirect pathway contribute to that?
PROBLEM 9.3
A person is learning a skill that requires the initiation of rapid eye
movement (a saccade) at the onset of a visual stimulus. If the
basal ganglia played a role in the acquisition of this skill, would
you expect the reaction time to increase or decrease with
practice?
Figure 9.8 The motor and oculomotor loops of the basal ganglia.
CHAPTER 9 IN A NUTSHELL
The basal ganglia are a group of
subcortical nuclei that are involved
in motor control, motor learning,
executive functions, and regulation of
emotions. The basal ganglia have input
nuclei, output nuclei, and intrinsic
nuclei. Input nuclei structures are
the caudate nucleus and the putamen,
and they receive incoming information
from cortical and other sources. The
output nuclei consist of the internal
segment of the globus pallidus (GPi)
and the substantia nigra pars
reticulata (SNr). These structures
send basal ganglia information to the
thalamus. The intrinsic nuclei consist
of the external segment of the globus
pallidus (GPe), the subthalamic
nucleus, and the substantia nigra pars
compacta (SNc), and they are located
between the input and output nuclei.
The basal ganglia need dopamine as an
input for proper functioning. Dopamine
is critical for reinforcement-based
motor learning, and dopamine
dysfunction is associated with several
movement disorders, such as
Parkinson’s disease and Huntington’s
disease.
Chapter 10
Cerebellum
PROBLEM 10.1
Which part of the cerebellum is involved in posture stabilization?
Figure 10.1 (a) The flattened cerebellar surface with the three major
subdivisions. (b) The cerebellar peduncles in the flattened view. (c) The sagittal
view of the cerebellum.
Figure 10.2 The frontal and parietal areas project to the cerebellum through the
ipsilateral pontine nuclei. Axons from the pontine nuclei decussate and enter the
contralateral cerebellum. These fibers are the largest source of input to the
cerebellum.
PROBLEM 10.2
If there is focal damage to the central part of the cerebellar cortex,
which types of movements will be affected?
Figure 10.4 Mossy and climbing fibers provide strong excitatory input to the
Purkinje cells. Purkinje cells are GABAergic and project to the deep cerebellar
nuclei; they are the only output cells of the cerebellar cortex. The output of the
Purkinje cells is completely inhibitory.
PROBLEM 10.3
GABAergic Purkinje cells provide the inhibitory output of the
cerebellar cortex. So all the output of the cerebellum is inhibitory.
But the deep cerebellar nuclei receive excitatory input from the
mossy and climbing fibers. What might be the role of such an
arrangement?
PROBLEM 10.4
A person is having difficulty reading road signs while walking.
Which part of the cerebellum might be damaged?
Figure 10.6 The red nucleus, inferior olive, and cerebellum form a closed-loop
feedback system that facilitates motor learning through error correction.
PROBLEM 10.5
If a person sustains an injury to the inferior olive only, which
aspect of motor behavior might be affected the most?
CHAPTER 10 IN A NUTSHELL
The cerebellum is located at the back
of the brain, underlying the occipital
and temporal lobes of the cerebral
cortex. The cerebellum is considered a
motor region because most of its
output goes to the motor areas, and
cerebellar damage causes impairments
in motor control and postural balance.
The main roles of the cerebellum are
coordination of voluntary movements,
maintenance of balance, motor
learning, and cognitive functions. The
cerebellum can be divided into two
parts, the cerebellar cortex and the
cerebellar nuclei. The cerebellar
cortex consists of the
cerebrocerebellum, the
spinocerebellum, and the
vestibulocerebellum. The cerebellar
nuclei are the output structures of
the cerebellum. Purkinje cells make
inhibitory connections onto the
cerebellar nuclei and are the sole
source of output from the cerebellar
cortex. Three cerebellar peduncles
carry the input and output of the
cerebellum. Damage to the cerebellum
causes cerebellar ataxia and deficits
in motor learning.
Chapter 11
Figure 11.1 The brainstem is made up of the midbrain, pons, and medulla
oblongata. It connects the cerebrum of the brain to the spinal cord and cerebellum.
It is involved in both sensory and motor functions.
Figure 11.2 Neurons in the reticular formation are scattered in different parts of
the brainstem. The reticular formation at the level of the midbrain is called the
mesencephalic reticular formation. At the level of the lower pons, it is called the
pontine reticular formation. At the level of the medulla, it is called the medullary
reticular formation. The superior colliculus is in the posterior segment of the
midbrain.
Figure 11.3 EMG responses in lower limb muscles to auditory stimulation during
locomotion. The EMG data were averaged and subtracted from both individual and
averaged stimulus trials. Figure shows two facilitatory startle responses at short
(~80-100 ms) and long latencies (~150-200 ms).
Adapted from Nieuwenhuijzen et al. (2000).
PROBLEM 11.1
The startle response shows a phenomenon called prepulse
inhibition. When a weak auditory stimulus (that does not elicit a
startle response) is followed by a strong stimulus within 0.5 s that
would normally elicit a response, the weak stimulus suppresses
the response of the strong stimulus. In Huntington’s disease
patients, the prepulse inhibition is reduced. What would this
suggest about the interactions between the basal ganglia and the
reticular formation?
PROBLEM 11.2
You are talking to a friend at an outdoor party when suddenly
through your peripheral vision you see a large object (an inflated
ball) rapidly approach your head. What kind of motor response
would you expect to make, and which brainstem structure would
make that possible?
Figure 11.4 The four components of the vestibular nuclei serve important
functions for controlling eye movements and postural equilibrium. The efferent
projections from the vestibular nuclei project to the nuclei of other cranial nerves
(III, IV, and VI). Lat = lateral; Sup = superior; Med = medial; Inf = inferior.
PROBLEM 11.3
Which brainstem structure would play an important role in posture
stabilization if we were unexpectedly pushed from behind?
Figure 11.5 The 12 cranial nerves of the nervous system shown along a
transverse view of the brainstem and cortex.
Figure 11.6 The approximate origins of the four main brainstem tracts involved in
motor control.
CHAPTER 11 IN A NUTSHELL
The reticular formation, superior
colliculus, red nucleus, and
vestibular nuclei are four subcortical
regions involved in motor control. The
descending tracts originating from
these areas are called the
reticulospinal, tectospinal,
rubrospinal, and vestibulospinal,
respectively. The reticular formation
is located in the rostral midbrain to
the caudal medulla. It is involved in
the coordination of axial and proximal
limb muscles, in particular during
standing and walking. The superior
colliculus is a major hub for
processing sensory information and for
reflexive actions that involve
orienting movements toward or away
from salient visual and auditory
stimuli. The main role of the red
nucleus is in facilitating
communication between motor nuclei in
the cortex and the cerebellum. The
vestibular nuclei are the major
destination of the axons that form the
vestibular division of the eighth
cranial nerve. The projections from
the vestibular nuclei control axial
muscles and proximal limb muscles.
They play an important role in gaze
stabilization. There are a total of 12
cranial nerves that serve sensory and
motor functions, and most of them
arise from the brainstem.
Problems for Part II
Self-Test Problems
1. Under the action of a new drug, the outputs of all the
cerebellar nuclei are increased 10-fold. What changes in
movements can be expected during standing, walking, and
reaching?
2. The neuronal population vector in a brain structure points
consistently in the direction of hand movement when a
monkey performs a reaching task from a standard position to
targets distributed along a circle (center-out task). The
researcher draws a conclusion that the brain structure
encodes a hand velocity vector in space. Offer alternative
interpretations of this finding.
3. A subject is asked to imagine performing a motor task, such
as a fast arm movement. No actual movement or change in
the muscle activation levels occurred. In what structures of the
central nervous system would you expect to see changes in
the background activity of the neurons? In what structures
involved in the generation of actual movement would you
expect to see no changes in the neuronal activity?
4. The coils of two transcranial magnetic stimulators have been
placed over the right and left primary motor cortices. When
one of them generates a single stimulus, a response is seen
in the contralateral arm muscles at a latency of about 20 ms.
When the other stimulator produces a stimulus a few
milliseconds prior to the first one, the response decreases.
What neurophysiological mechanisms are likely to be involved
in the modulation of the response?
5. A drug has been discovered that increases the efficacy of the
projections from the putamen to the external part of the globus
pallidus by a factor of 10. A healthy person takes a single
dose of the drug. What changes would you expect to see in
their motor performance?
6. If a patient with Parkinson’s disease makes a saccade to a
target, where would you expect the saccade to land? On the
target, short of the target, or beyond the target? Why?
Central Processing of
Somatosensory Information
The axons entering the dorsal roots divide into ascending and
descending systems. It is important that some neurons entering the
dorsal roots synapse directly on motoneurons in the ventral horn.
This circuit forms the basis of the monosynaptic stretch reflex. The
stretch reflex or the myotatic reflex (an obsolete term) refers to a
muscle contraction in response to stretching of muscle fibers. The
proprioceptive afferent axons in the ascending branches travel along
with axons of tactile afferents in the dorsal column.
PROBLEM 12.1
Why would the nervous system only send a copy of the
proprioceptive information, and not tactile information, to the
cerebellum?
Figure 12.3 Frontal view of the sensory homunculus is a map of the body in the
somatosensory cortex. Adjacent regions of the body are represented next to each
on the map. Body parts with smaller receptive fields of tactile neurons have a
disproportionately larger representation in the somatosensory cortex.
12.6 Integration of
Somatosensory Input With Other
Sensory Modalities
Neurons in S1 also project to areas 5 and 7 in the parietal cortex.
The projections to the parietal cortex relay the current muscle state
(muscle length, velocity, and force) to the visual areas in the parietal
cortex. The parietal cortex is also the site for the integration of
somatosensory and visual information with information from the
vestibular system (chapter 13). By virtue of these projections,
parietal areas integrate and process multisensory information. These
parietal areas then project the processed and integrated signals from
multiple modalities to the premotor and motor cortex in the frontal
cortex for preparation and execution of goal-directed movements.
PROBLEM 12.2
Tabes dorsalis is a disorder in which there is demyelination of the
axons in the dorsal column of the spinal cord. If an individual with
this disorder has suffered unilateral damage to axons, would
somatosensory function be altered on the ipsilateral side of the
body or the contralateral side?
CHAPTER 12 IN A NUTSHELL
Somatosensory information originates
in the peripheral receptors in the
skin, muscle spindles, and tendons and
makes its way to the central nervous
system through a series of neurons.
The first-order neurons carry
information from the ipsilateral side
of the body to the spinal cord and the
lower medulla. This bundle of axons is
called the dorsal column. The first-
order neurons in this column synapse
on the second-order neurons that
decussate and carry somatosensory
information to the contralateral
nuclei of the thalamus. The bundle of
axons that carry the second-order
neurons are called the medial
lemniscus. The third-order neurons
carry information from the thalamus
through the internal capsule to the
primary somatosensory cortex. The
primary somatosensory cortex (S1) is
topographically organized. The trunk
and the legs are represented more
medially, and the hand and the face
areas are represented more laterally.
Information from S1 is projected to
higher-order sensory areas such as S2
and areas 5 and 7 in the parietal
cortex for further processing and
integration with other sensory
modalities.
Chapter 13
Figure 13.1 (a) The vestibular organs in the inner ear consist of otolith organs
(i.e., utricle and saccule, which sense linear acceleration) and three almost
orthogonal semicircular canals, which sense rotational acceleration. The vestibular
nerve projects signals from otoliths and semicircular canals to the central nervous
system. (b) The vestibular system encodes linear movement in three-dimensional
space denoted as front, back, left, right, up, and down directions (by otolith organs)
and rotational movements (i.e., yaw by the horizontal canal and pitch and roll by
both anterior and posterior canals).
Linear motion of the head is detected by the two otolith organs,
the saccule and the utricle. These two similar organs lie against the
walls of the inner ear. The receptors for these two organs, maculae,
are hair cells. Overlying the hair cells is a gelatinous layer called
otolithic membrane that consists of calcium carbonate crystals called
otoconia. The relatively heavier otoconia add inertia to the otolithic
membrane. So, when the head tilts, the gravitational pull causes the
membrane to move relative to the macula, and the resulting parallel
force displaces the hair bundles and produces a change in the
electric potential on the receptor membrane. Stereocilia are the key
mechanosensors of hair cells, and the orientation of the stereocilia
toward the kinocilium changes the receptor potential. Kinocilium is
the longest cilium located on the hair cell next to many stereocilia.
When the stereocilia lean toward the kinocilium, the cell is
depolarized, and there is an increase in the vestibular nerve activity.
In contrast, when the stereocilia tilt away from the kinocilium, the cell
is hyperpolarized, and there is a decrease in afferent nerve activity.
In the utricle, the kinocilia are oriented toward the middle line called
striola, and in the saccule they are oriented away from it. The
different orientations of the saccule and utricle allow the vestibular
system to detect head tilts in different directions.
Semicircular canals detect head rotations arising from either
voluntary movements or from angular accelerations caused by
external forces, such as on a roller coaster ride. The three
semicircular canals, almost at right angles to each other, encode
head rotations along three perpendicular axes. Each of the three
canals has at its base an expansion called an ampulla that houses
the crista. The crista contains hair cells that extend out into a
gelatinous mass called the cupula, which extends along the entire
length of the ampulla. In contrast to the saccular and macular hair
cells, all hair cells in the crista are organized with their kinocilia
pointing in the same direction (see figure 13.2).
Figure 13.2 The ampulla of the semicircular canal and a cross-sectional view of
the vestibular hair cells. Head rotation causes the membranous canal to distort the
cupula.
PROBLEM 13.1
If an animal has a damaged otoconia, which type of head
movement would it have difficulty detecting?
13.2 Vestibular Afferents
Respond to Head Motion
Axons of the otolith organs and semicircular canals exhibit
spontaneous levels of high activity. Thus, they can convey
information by both increasing and decreasing their firing rates. For
otolith organs, the firing rates increase or decrease based on the
direction of the head tilt. The response rate of the neurons remains
elevated (or depressed) as long as the head remains tilted—
meaning that the tonic activity of the neurons encodes the static
force on the head. In contrast, when there is a sudden translational
movement of the head, such as what the head experiences when a
vehicle suddenly accelerates, there is a transient increase or
decrease in the firing rate of the otolith organs.
The firing rate of the receptors in the semicircular canals
increases when the head rotation accelerates and the cupula is
deflected in one direction, and the firing activity decreases when the
head decelerates and the cupula is rotated in the opposite direction.
This receptor apparatus is very sensitive and can detect angular
acceleration as small as 0.1°/s2. Note that physical displacements of
the cupula are less than 10 nm, comparable to those produced by
low-amplitude sound in the auditory system. During rotation at
constant speed, the firing rate returns to baseline levels. Head
rotations at constant speed can be encountered during flights and
slow rides in amusement parks.
PROBLEM 13.2
A woman finds her shoelaces undone, and she bends down and
tilts her head downward to tie her shoes. She then straightens her
head back up again. How would the firing rate of the vestibular
nerve change during this activity?
13.3 Central Projections From
the Otolith Organs and
Semicircular Canals
The vestibular organs communicate via cranial nerve VIII
(vestibulocochlear nerve), which synapses in the brainstem and the
cerebellum. The cell bodies of the neurons of the vestibular section
of the vestibulocochlear nerve are in the vestibular nerve ganglion
(also called Scarpa’s ganglion). These neurons are bipolar. The
peripheral processes innervate the otolith organs and the
semicircular canals, and the central processes project to the
vestibular nuclei and the cerebellum. The vestibular nuclei are
located in the medulla and pons of the hindbrain. Afferent signals
from the otolith organs and semicircular canals converge in the
vestibular nuclei.
An important thing to note is that the receptor potentials
generated by the hair bundles in the otolith organs do not distinguish
between static head tilts and translational motions of the head.
Otolith organs are excited tonically by static head tilts and phasically
by translational movements of the head, whereas the semicircular
canals are only excited by head rotations that accompany tilts.
Otolith afferents respond to inertial motion (e.g., head motion of a
passenger when a bus suddenly accelerates) and a change in head
orientation relative to gravity. To disambiguate head motion arising
from different sources, the vestibular system also uses signals from
the semicircular canals and combines those signals more centrally in
the vestibular nuclei and the cerebellum. In fact, the mixing of the
signals from otolith organs and the semicircular canals in the
vestibular nuclei makes it possible for the nervous system to
disambiguate translational and rotational head movements. Thus,
when there is a head tilt, integration of information from both the
otolith organs and the semicircular canals in the vestibular nuclei and
the cerebellum can be used to disambiguate head tilts from
translational movements of the head.
The cerebellum, specifically the vestibulocerebellum and
spinocerebellum (see chapter 10), is a major destination of vestibular
afferent information, and the cerebellum in turn projects back to the
vestibular nuclei. The major nodes in the cerebellum include the
flocculus, nodulus, uvula, and rostral fastigial nuclei. The nodulus
and uvula integrate information from otolith organs and semicircular
canals to disambiguate head tilts from translation. Another important
role played by the cerebellum is that it differentiates vestibular
signals that arise from self-motion and external forces. For example,
if a ballet dancer slips and their head rotates backward, the
vestibular system will trigger protective reflexes, but if the same
posture is acquired by the dancer during a performance, then the
cerebellum will suppress any protective reflexes. It has been
hypothesized that predictive signals in the cerebellum cancel out the
ascending vestibular information in the rostral fastigial nucleus
(reviewed in Cullen 2019) during self-motion (see figure 13.3).
An alternative interpretation is that voluntary movements are
produced by changes in referent body configurations (see chapter
21), which are used as the origins of the coordinate systems to
measure and interpret information from relevant sensory receptors
(see chapter 28). When signals from vestibular receptors change
without a change in the referent body configuration, they are
interpreted as deviations from that configuration caused by external
forces (perturbations) and lead to quick corrective changes in muscle
activation. If the referent configuration changes to produce a
voluntary movement, similar changes from the vestibular system do
not produce such corrections.
In addition to the cerebellum, the information from the vestibular
nuclei also reaches the thalamus on its way to the cerebral cortex.
The regions in the cerebral cortex include Brodmann’s area 3a, 2v,
and the parietoinsular vestibular cortex. Activation in these regions is
important for perceptions arising from vestibular sensations. The
parietoinsular vestibular cortex is also a multisensory area that
responds to proprioceptive and visual information. This suggests that
most cortical processing of vestibular information is multisensory in
nature.
PROBLEM 13.3
A figure skater has sustained a concussion that has temporarily
disrupted communication in the ascending vestibular pathways to
the cerebellum. How could the skater’s performance be affected
on skills such as the layback spin?
Figure 13.3 The vestibular projections to the cortex and the cerebellum. The
inputs from the vestibular labyrinth project to the vestibular nuclei. From there the
projections extend to both the cerebellum and the cortex.
PROBLEM 13.4
A person reports feeling dizzy while nodding their head in
agreement while looking at someone’s face. What would be the
most likely cause of this disorder?
PROBLEM 13.5
A patient reports intact hearing in both ears, but difficulties with
spatial localization of sound sources. This is called the central
auditory processing disorder (CAPD). Which are the likely neural
substrates involved in CAPD?
CHAPTER 13 IN A NUTSHELL
The vestibular system is involved with
the maintenance of equilibrium and the
orientation of the body in space. The
auditory system is concerned with the
perception of sound, its spatial
location, and the perception of
language. The vestibular system
encodes both translational and
rotational motion of the head. Axons
of the otolith organs and semicircular
canals exhibit spontaneous levels of
high activity and convey information
both by increasing and decreasing
their firing rates. The vestibular
system plays an important role in
stabilizing gaze during head
movements. The cerebellum and
parietoinsular vestibular cortex are
important areas in the brain for
processing vestibular signals. The
cochlea, inferior colliculus, and
primary auditory cortex play an
important role in processing auditory
signals.
Chapter 14
Visual System
PROBLEM 14.1
A person with dichromatic vision will have difficulties performing
which common activities of daily living (and why)?
14.4 Optic Nerve, Tracts, and
Radiations
Retinal ganglion cells exit the retina through a region called the optic
disk in a bundle called the optic nerve. The optic nerve runs to the
optic chiasm, where more than half (~60%) of the fibers of the optic
nerve decussate and enter the contralateral cortex. The remaining
axons continue toward the lateral geniculate nucleus (LGN) of the
thalamus and other structures in the midbrain, such as the superior
colliculus. Once the axons go past the optic chiasm, they form the
optic tract. Because of the decussation, the optic tract contains nerve
fibers from both eyes. The thalamic neurons send axons to the
primary visual cortex through the optic radiations in the internal
capsule (see figure 14.4).
Figure 14.4 The visual pathway from the retina to the primary visual cortex.
PROBLEM 14.2
If a person sustains an injury that leaves a complete lesion of the
left optic tract, how would that affect their vision? How could this
person compensate for the lesion of the left optic tract?
PROBLEM 14.3
A person was born with congenital damage to the right retina.
Which normal visual function would this person have a difficulty
performing?
Figure 14.7 The spatial and perceptual aspects of vision are processed by
different but interdependent streams. The figure shows the interconnections
between the different nodes of the dorsal stream (solid black) and the ventral
stream (dashed black). SPL = superior parietal lobule; IPL = inferior parietal lobule;
MT = middle temporal area; MST = medial superior temporal area; LGN = lateral
geniculate nucleus; FFA = fusiform face area; IT = inferotemporal cortex.
PROBLEM 14.4
A healthy adult slipped on ice and suffered a concussion. The
patient later tells the family doctor about experiencing difficulties
at four-way intersections with two-way stop signs. They cannot
gauge how fast a car is approaching to determine if it is safe to
pull through the intersection. Which area of the extrastriate cortex
most likely was affected by the concussion?
14.8 Neurons of the Two Visual
Streams
Layers in the LGN receive input from individual retinas from each
eye. The layers also differ based on the cell size of the neurons. In
each LGN, the two ventral layers are primarily composed of large
magnocellular neurons (M cells), and the four dorsal layers are
composed of the relatively smaller parvocellular neurons (P cells).
These neurons receive inputs from different retinal ganglion cells
that differ in morphology. The axons of M ganglion cells have larger
diameters, and the cells have more extensive dendritic trees and
larger cell bodies. M ganglion cells also have larger receptive fields
and faster conduction velocities. Besides these morphological
differences, M cells generate phasic responses to presentations of
visual stimuli, whereas P cells produce tonic sustained responses. P
ganglion cells are sensitive to differences in wavelengths of light in
their receptive fields, and consequently they respond to color. In
contrast, M cells do not respond to the color of the stimulus.
The M and P cells give rise to the magnocellular and parvocellular
pathways, respectively. The magnocellular pathway forms the dorsal
ventral stream, and the parvocellular pathway forms the ventral
visual stream (see figure 14.8). The magnocellular pathway carries
information about large and fast-moving objects. It has low spatial
resolution but high temporal resolution, and it is agnostic to the color
of the stimulus. The parvocellular pathway carries information about
small objects and also encodes their color and has high spatial
resolution and low temporal resolution.
Figure 14.8 The neuronal connections between the retina and the parvocellular
and magnocellular layers of the lateral geniculate nucleus (LGN). The
parvocellular and magnocellular neurons are the bases of the ventral and dorsal
visual streams, respectively.
Adapted by permission from C. Meissirel, K.C. Wikler, L.M. Chalupa, and P. Rakic, “Early
Divergence of Magnocellular and Parvocellular Functional Subsystems in the Embryonic
Primate Visual System,” Proceedings of the National Academy of Sciences of the United
States of America 94, no. 11 (1997): 5900-5. Copyright (1997) National Academy of
Sciences, U.S.A.
PROBLEM 14.5
A monkey sustained an injury to layers 4 through 6 of the
thalamus. Which visual function or functions are most likely to be
affected by this injury?
14.9 Visual Deficits Due to
Area-Specific Visual System
Damage
Visual deficits can occur due to damage to the retina, optic nerve,
tracts, radiations, or the magnocellular or parvocellular pathways.
Damage to the retina causes macular degeneration, which results in
loss of central vision. Age-related macular degeneration is the
leading cause of severe, permanent vision loss in people over age
60. Glaucoma is caused by high pressure in the eye, and it results in
damage to the optic nerve. It is also one of the leading causes of
blindness for people over age 60.
Damage to the primary visual pathway can cause specific visual
field deficits, called anopsias. Damage to one of the optic tracts of
one eye causes a complete loss of sight in the contralateral visual
field. This is referred to as homonymous hemianopsia. In contrast,
damage to the optic chiasm causes loss of bilateral vision in the
lateral visual fields. This is because about half of the optic nerve
fibers from each eye that map the lateral visual field decussate at the
optic chiasm. This bilateral visual field deficit is called heteronymous
hemianopsia. Similarly, damage to the optic radiations causes visual
deficits in one half of the contralateral visual hemifield (see figure
14.4). For example, damage to optic radiations will result in loss of
vision in the inferior contralateral visual field. This is called
homonymous quadrantanopsia.
Cerebral akinetopsia is a rare disorder that occurs because of
damage to the motion-processing area MT/MST along the
magnocellular pathway. Patients suffering from this disease have
difficulty crossing the street because they are unable to perceive the
motion of an oncoming vehicle. Cerebral achromatopsia occurs
because of damage to areas V4 and V8 along the parvocellular
pathway. Patients suffering from this disease are unable to see the
world in color, though most other aspects of their vision remain
intact.
PROBLEM 14.6
Damage to the dorsal visual stream in the parietal lobe will affect
which type of eye movement?
CHAPTER 14 IN A NUTSHELL
Photoreceptors transform light falling
onto them into action potentials.
There are two types of photoreceptors,
rods and cones. Rods are responsible
for vision at low light levels. Cones
are densely packed in the fovea, and
they are active at higher light
levels. There are three types of
cones: short (blue), medium (green),
and long wavelength (red) cones. These
cones mediate trichromatic color
vision. A loss of green or red cones
contributes to color blindness. The
retinal ganglion cells carry action
potentials to the lateral geniculate
nucleus in the thalamus. From the
thalamus, visual information reaches
the primary visual cortex (V1).
Extrastriate cortical areas receive
information from V1 and are organized
into two independent systems, the
dorsal and ventral visual streams,
that specialize in visuomotor and
visuoperceptual processes,
respectively. The eyes make different
types of movements to gather
information from the visual scene and
to keep the image of an object on the
retina. Saccades are ballistic eye
movements used to gather information.
Smooth-pursuit eye movements are
slower eye movements that track moving
objects to keep their image on the
retina. Vergence eye movements involve
rotation of the two eyes in opposite
directions to view objects in depth.
Problems for Part III
Self-Test Problems
1. High-frequency muscle vibration typically leads to illusions of
motion in a joint spanned by the muscle. Most frequently, the
illusory movement corresponds to muscle elongation. Under
some conditions, however, the direction of the illusory
movement reverses. Offer an interpretation for these
observations.
2. If a person sustains injury to the left medial lemniscus, they
will lose tactile sensations on which side of the body?
3. A person reports “unstable bouncing vision” and an inability to
read road signs during walking. What is this person suffering
from?
4. A person demonstrates normal ocular reflexes and no
changes in the peripheral eye structures. What pathology can
you suspect if the person reports no visual experiences
(blindness)?
5. A person has suffered a lesion in the left middle temporal
areas of the brain. What ability have they most likely lost?
Reflexes
PROBLEM 15.1
Suggest examples of actions, commonly addressed as reflexes,
that violate the suggested criteria.
PROBLEM 15.2
What about plants? Their parts (e.g., roots, branches, and fruits)
grow and change location in space (move!). Can these
movements be classified into reflex and voluntary?
PROBLEM 15.3
What is missing in the suggested definition of reflex latency as the
sum of three components?
Figure 15.6 A quick increase in a stimulus can produce both phasic and tonic
reflexes. Phasic reflexes are transient. Tonic reflexes are steady state.
PROBLEM 15.4
What would you expect to see in a person lacking inhibitory
neuromediators in the spinal cord?
Figure 15.7 Conditioned reflexes emerge when a natural stimulus for a particular
response is replaced with another stimulus as a result of multiple presentations of
the second stimulus. This can happen for the salivation reflex, which is seen
naturally in response to seeing or smelling food, if food presentation is preceded
by a stimulus of a different modality (e.g., a bell ring).
PROBLEM 15.5
Imagine that you developed a typical Pavlovian conditioned reflex
and then rang the bell a few times without bringing food to the
dog. And then, once again, ring the bell and bring food at about
the same time. What will happen with salivation?
Pavlov’s views dominated the field for many years and were
considered ultimate truth in the Soviet Union. The only person who
dared to criticize Pavlov openly was Nikolai Bernstein (2003).
According to Bernstein, reflexes played the very important role of a
foundation for functional movements, but behaviors were initiated
primarily within the central nervous system, not in response to
sensory stimuli. Based on this “principle of activity,” Bernstein
created his multilevel control scheme for biological movements
described in chapter 1. During the later years of his life, Bernstein
(1966) started to sketch a new field in physiology that he termed
physiology of activity. Unfortunately, this field remains barely
developed.
CHAPTER 15 IN A NUTSHELL
Reflex is a loosely defined notion
meaning an involuntary motor reaction
to an external stimulus. The simplest
reflex consists of a receptor (sensory
ending), an afferent nerve, at least
one synapse on a neuron within the
central nervous system, an efferent
nerve, and an effector. Three main
characteristics of reflexes are time
delay (latency), gain, and
habituation. Monosynaptic reflexes
contain only one central synapse; they
always induce excitatory effects on α-
motoneurons. Oligosynaptic reflexes
contain two or three synapses within
the central nervous system. The neural
loop of polysynaptic reflexes is
typically unknown. Reflexes can induce
short-lasting reactions (phasic
reflexes) or steady-state changes in
muscle activation (tonic reflexes).
They can induce excitation or
inhibition of baseline muscle
activity. Some reflexes are seen at
birth. Conditioned reflexes represent
responses to an unusual stimulus that
replaces the natural one as a result
of repetitive presentation of both
stimuli. According to Pavlov’s theory,
conditioned reflexes form the basis of
all animal behaviors. This theory was
criticized by Bernstein, who suggested
that voluntary actions could be
initiated without any external
stimulus by the central nervous system
(the principle of activity).
Chapter 16
PROBLEM 16.1
What is the antonym of “laminar”? What could be the origin of the
laminar structure of the spinal cord?
Figure 16.3 Each vertebra has a body and a spinous process. Peripheral
information gets into the spinal cord through the dorsal roots, while efferent signals
are sent from the spinal cord through the ventral roots.
PROBLEM 16.2
Can an inhibitory neuron be used to increase the excitability of
another neuron?
PROBLEM 16.3
A muscle is actively contracting against a constant load. What
effects do you expect from the action of Renshaw cells upon γ-
motoneurons? Is this a negative or a positive feedback?
Figure 16.8 Ia interneurons receive excitatory inputs from Ia afferents and make
inhibitory synapses on α-motoneurons innervating the antagonist muscle. Ia INs
are inhibited by Renshaw cells and also receive descending inputs.
PROBLEM 16.4
You stimulate a pool of Ia interneurons. How will muscle force
change for an agonist and for an antagonist muscle? What kind of
change do you expect to see in the firing rate of Renshaw cells?
PROBLEM 16.5
Is postsynaptic inhibition equally effective in decreasing the
response of the cell to all the presynaptic excitatory synapses
(e.g., those on remote dendrites and those close to the axonal
hillock)?
PROBLEM 16.6
What will happen if an inhibitory synapse is acting on the
presynaptic membrane close to the synaptic cleft?
CHAPTER 16 IN A NUTSHELL
The spinal cord is protected by the
spine; it has a laminar structure and
is divided into segments that
innervate certain areas of the body.
The spinal cord contains numerous
motoneurons, interneurons, and
conduction pathways that carry both
descending and ascending information.
Afferent information enters the spinal
cord through the dorsal roots, while
efferent nerves that carry motor
signals exit through the ventral
roots. Inhibition within the central
nervous system is vital for its proper
functioning. Postsynaptic inhibition
hyperpolarizes the postsynaptic
membrane and makes the neuron less
sensitive to all excitatory inputs.
Presynaptic inhibition works through
depolarization of presynaptic fibers
and selectively decreases the
effectiveness of only some of the
inputs. Renshaw cells are interneurons
that are excited by axons of α-
motoneurons and inhibit motoneurons of
the same pool (recurrent inhibition).
Ia interneurons are excited by Ia
afferents from primary spindle sensory
endings; they inhibit α-motoneurons of
the antagonist pool (reciprocal
inhibition). Persistent inward
currents through dendritic membranes
play a major role in modulating the
excitability of spinal motoneurons and
defining their recruitment patterns.
Chapter 17
Monosynaptic Reflexes
Monosynaptic reflexes are the only ones with the afferent source and
reflex pathway relatively well defined. They originate from primary
spindle endings and make only one intraspinal excitatory connection
(synapse) with α-motoneurons of the muscle housing the spindle or,
sometimes, its agonist (i.e., a muscle causing joint movement in the
same direction). Not all human muscles show monosynaptic
reflexes. This depends on the density of synapses of Ia afferents on
the α-motoneurons and a number of other factors, such as the
strength of presynaptic inhibition of the Ia afferents.
PROBLEM 17.1
Why does the last phrase contain the expression “nearly
synchronously”? What factors can lead to violations of
synchronicity?
Figure 17.1 A scheme of an experiment with electrical stimulation of a muscle
nerve. Note that the stimulus is applied to both afferent and efferent fibers,
resulting potentially in four volleys of action potentials (AP1, AP2, AP3, and AP4)
traveling orthodromically and antidromically along the afferent and efferent fibers.
PROBLEM 17.2
What can happen with a motoneuron if an antidromic action
potential arrives at the axonal hillock? Consider as many
scenarios as possible.
PROBLEM 17.3
What will happen if two action potentials (one orthodromic and
one antidromic) are moving toward each other and meet at a
certain point on the axon?
PROBLEM 17.4
What is the frequency of stimulation at which you may expect to
see the M-response become smaller?
Figure 17.6 A tendon tap excites spindle endings and may induce a
monosynaptic reflex contraction (T-reflex). Its reflex pathway is the same as for the
H-reflex.
17.3 Effects of Voluntary
Muscle Activation on
Monosynaptic Reflexes
Monosynaptic reflexes are rather poorly controlled by the subject’s
will, although prolonged training of monkeys has been shown to lead
to their ability to modulate the amplitude of a monosynaptic reflex
without changing the activation level of any muscle (Wolpaw 1987;
Wolpaw and Carp 1993). These studies used the operant
conditioning technique, which rewards desired responses by the
animal. Later studies have shown that humans can also show such
plastic changes in monosynaptic reflexes after a relatively short
training session, which can be useful in clinical practice (Thompson
and Wolpaw 2015).
Human subjects can modulate the amplitude of monosynaptic
reflexes indirectly, for example, by activating certain muscle groups.
Voluntary activation of a muscle leads to depolarization of the
membranes of many α-motoneurons innervating the muscle. Some
of the motoneurons generate action potentials, while others are
depolarized under the threshold for action potential generation.
These motoneurons are more excitable and can be more easily
recruited by a standard afferent discharge induced by an electrical
stimulus. As a result, more motoneurons respond to a standard
stimulus, leading to an increase in the overall reflex muscle
response. For example, voluntary activation of the calf muscle group
increases the H-reflex in this muscle induced by an electrical
stimulation of the tibial nerve (figure 17.7).
Figure 17.7 Voluntary muscle activation increases the amplitude of the H-reflex
in the activated muscle through an overall excitation of the motoneuronal pool.
Figure 17.8 Voluntary activation of an antagonist muscle leads to suppression of
the H-reflex amplitude via activation on Ia interneurons (Ia-IN).
PROBLEM 17.5
What will happen with the H-reflex amplitude if you strongly
coactivate the agonist–antagonist pair of muscles?
Besides that, activation of distant large muscle groups can lead to
a modulation of the H-reflex in the calf muscles. This can be
achieved, for example, with the Jendrassik maneuver, which consists
of a steady, strong contraction of remote muscle groups. For
example, the Jendrassik maneuver may involve clenching the hands
in front of the chest and trying to separate them by a strong
contraction of the shoulder and back muscles. Alternatively,
clenching teeth strongly can lead to similar effects. After several
seconds, the Jendrassik maneuver may lead to a change (usually,
an increase) in the amplitude of the H-reflexes in the calf muscles.
The mechanism of this effect is not known, but it likely involves the
conduction of action potentials along propriospinal pathways, which
connect spinal segments at different levels of the spinal cord.
17.4 F-Wave
As mentioned earlier, the H-reflex can be readily seen only in some
human muscles. This may be due to a number of factors, in
particular to differences in the density and effectiveness of
monosynaptic connections of Ia afferents on α-motoneurons, and to
different relations between the diameters of Ia afferents and efferent
axons.
PROBLEM 17.6
How can a synapse on an α-motoneuron be more or less effective
than another synapse?
PROBLEM 17.7
For a given set of Ia afferents, is their stimulation more likely or
less likely to induce the H-reflex in a muscle innervated by very
thin and very thick axons of α-motoneurons?
CHAPTER 17 IN A NUTSHELL
Reflex is a loosely defined notion
meaning an involuntary motor reaction
to an external stimulus. The simplest
reflex consists of a receptor, an
afferent nerve, at least one synapse
on a central neuron, an efferent
nerve, and an effector. Monosynaptic
reflexes contain only one central
synapse; they always induce excitatory
effects on α-motoneurons. They can be
induced by an abrupt change in the
activity of primary muscle spindle
afferents, as during tendon tap (T-
reflex), or in response to an
electrical stimulation of a muscle
nerve (H-reflex). In the latter case,
motor axons are also stimulated,
leading to a direct muscle response
(M-response). Voluntary activation of
a muscle leads to an increase in its
monosynaptic reflexes. In some
muscles, an F-wave can be observed,
which is a direct response of α-
motoneurons to an antidromic volley
traveling up their axons, induced by
an electrical stimulus.
Chapter 18
Figure 18.2 Golgi tendon organs send their axons (Ib afferents) to Ib
interneurons, which exert an inhibitory action on the agonist α-motoneurons and
disinhibit (excite) antagonist α-motoneurons via another interneuron.
PROBLEM 18.1
What kinds of reflex effects can be expected from the action of Ib
afferents during a fast flexion movement against a constant
external load?
PROBLEM 18.2
What can be expected from the action of Ib afferents during a fast
development of extension force against a stop?
PROBLEM 18.3
Reflex eye movements make it easy to maintain gaze direction
during voluntary head movements. Is this a tonic or a phasic
reflex?
Figure 18.3 Tonic and phasic components of the muscle reflex to fast stretch
(thick black lines). Stretching the muscle over the same range slowly leads to the
tonic reflex only (gray dotted lines).
Figure 18.6 A muscle is slowly stretched by an external force. First, it resists the
stretching only due to its passive elasticity (passive characteristic, dashed). At a
certain threshold, recruitment of α-motoneurons begins, leading to active force
development (tonic stretch reflex). The force–length dependence is called the tonic
stretch reflex characteristic (thick, solid line). Muscle activation increases with
further stretch (illustrated with different sizes of the EMG label). The slope of the
characteristic (its apparent stiffness; see the dotted line) increases with length.
PROBLEM 18.5
If all the Ia afferents are driven by vibration, why don’t they induce
monosynaptic reflexes in response to each vibration cycle?
Figure 18.7 High-frequency muscle vibration leads to a slow reflex increase in
muscle force (tonic vibration reflex). It starts at a substantial delay (commonly, a
few seconds) and lasts for some time after the stimulus is turned off.
PROBLEM 18.6
Suggest at least two explanations for the suppression of H-
reflexes during the tonic vibration reflex.
Figure 18.11 Stimulation of the flexor reflex afferents in one limb induces a reflex
response in flexor muscles of this limb and a crossed extensor reflex in extensor
muscles of the contralateral limb.
CHAPTER 18 IN A NUTSHELL
Oligosynaptic reflexes contain a few
central synapses, while polysynaptic
reflexes contain many central
synapses. Reflexes of these two groups
can be excitatory or inhibitory.
Phasic stimuli or responses are those
that change in time quickly; those
that are steady-state are called
tonic. Muscle spindles and Golgi
tendon organs are sources of
oligosynaptic reflexes that can be
viewed as negative feedback loops.
Polysynaptic reflexes can involve
muscles of a whole limb, as during the
flexor reflex, which is induced by the
activity of a variety of receptors
with relatively thin axons. The tonic
stretch reflex is an increase in the
activity of motoneurons innervating a
muscle that is being stretched by an
external force. The tonic vibration
reflex is a steady increase in the
level of activation of a muscle during
vibration at a high frequency and low
amplitude. This reflex can be seen in
other muscles, suggesting that its
neural loop involves a spinal central
pattern generator for locomotion. The
tonic vibration reflex is accompanied
by suppression of monosynaptic
reflexes induced by an increase in
presynaptic inhibition.
Chapter 19
PROBLEM 19.1
The difference between the latency of the monosynaptic response
and that of M2 is the same in arm muscles and in leg muscles.
What does this finding tell us about the possible transcortical
nature of M2?
Figure 19.2 Preprogrammed reactions (M2 and M3) demonstrate strong
dependence on the instruction. If the subject is instructed to resist perturbations,
the preprogrammed reactions are large (solid lines); if the subject is asked to let
the limb move (let go), the preprogrammed reactions are much smaller (dashed
lines). Note that the M1 response is the same.
PROBLEM 19.2
What would you expect to happen if, in the previous example, the
glass suddenly started to fall (watched by the subject at all times)?
PROBLEM 19.3
You are carrying a bundle of firewood with extended arms.
Suddenly, all the firewood drops. What kinds of reactions can you
expect in your biceps and triceps?
Figure 19.5 Muscle vibration has different effects on different components of the
responses to an external perturbation. The early response (M1) is suppressed, just
as the H-reflex is, while the preprogrammed response (M2-M3) is unchanged.
PROBLEM 19.4
You apply a sequence of perturbations in the same direction but of
different magnitudes and measure an integral of a
preprogrammed reaction in consecutive trials. Do you expect this
integral to correlate with any characteristic of the perturbations?
Vibration-induced changes in muscle afferent activity may not be
directly related to preprogramming, and thus, they influence only the
amplitude of a signal for the preprogrammed response playback. On
the other hand, the amplitude of the preprogrammed response does
not depend upon the magnitude of the triggering signal giving rise to
the response, which explains the lack of vibration influence upon
preprogrammed reactions. Note that muscle vibration suppresses
monosynaptic reactions through the mechanism of presynaptic
inhibition, which is a selective inhibition mechanism. Apparently, it
acts on the terminals of Ia afferents on α-motoneurons but not on the
terminals of interneurons that participate in the hypothetical loop
bringing about preprogrammed reactions.
PROBLEM 19.5
You are picking up a small object that turns out to be very heavy
(unexpectedly for you). What kinds of reactions do you expect to
see in the arm muscles?
19.7 Corrective Stumbling
Reactions
Locomotion is another very commonly used movement in everyday
animal and human activity (chapter 26). Therefore, it is reasonable to
expect the neural mechanism of locomotor movement generation to
always be associated with preprogramming of corrective actions to
perturbations. This is particularly true for bipeds, such as humans,
who have to combine the stability of locomotion with the stability of
the vertical posture.
A particular pattern of long-latency reactive responses has been
observed during cat (and human) locomotion associated with
overcoming an unexpected obstacle: the corrective stumbling
reaction (Forssberg et al. 1975, 1977; Duysens and Pearson, 1976;
Duysens et al. 1990; Prochazka et al. 2002). This pattern can be
observed in cats in response to weak mechanical stimulation of skin
areas of the paw or of the leg (even with an air puff) or during short
electrical stimulation of skin nerves.
The application of any of these stimuli during the swing phase
gave rise to a flexor reaction with the hindlimb transferring over a
hypothetical obstacle (figure 19.7). The same stimulation applied
during the stance phase could give rise to an extensor reaction
accelerating the step. The coordinated, functionally appropriate
pattern of this reaction and the relative independence of the stimulus
suggest that it is a preprogrammed response of a mechanism
responsible for the compensatory reactions needed during everyday
life. For example, imagine that one hits a stone or a branch with the
tip of the foot during the swing phase. The corrective stumbling
reaction will contribute to lifting the foot and stepping over the
obstacle. In another situation, imagine stepping on a sharp object
(e.g., a tack). Given that the other foot is in the swing phase, lifting
the foot off the sharp object is impossible because the body would
collapse. The only feasible strategy is to accelerate the step to
minimize the time the object is in contact with the foot.
Figure 19.7 A mechanical or electrical stimulation of the paw during locomotion
induces different reactions in the swing and the stance phases. In the swing phase
(A), there is a flexor reaction, so that the leg “steps over” a fictitious obstacle. In
the stance phase (B), there is an extensor reaction, leading to shortening of the
stance phase for this limb.
PROBLEM 19.6
Can you define the afferent source of the corrective stumbling
reaction with the method of successive limb denervation—that is,
eliminating afferent inputs from areas of the leg?
PROBLEM 20.1
Suggest examples of negative and positive feedback from the
material we have already considered in this book and from
everyday life.
PROBLEM 20.2
Suggest an area of sports in which having a servo mechanism
would be vital for good performance. Suggest another area in
which such a mechanism would be disastrous.
Figure 20.7 illustrates how the servo hypothesis functions with the
help of force–length muscle characteristics. The central command
specifies the location of a characteristic corresponding to a certain
value of muscle length. In order for the servo mechanism to ensure
perfect compensation of possible changes in external load, the
characteristic must be vertical; then muscle length will not depend on
muscle (and external) force. Voluntary movements are performed by
shifting the characteristics along the x-axis so that the independently
controlled variable may be associated with the signal to γ-
motoneurons (γ1, γ2, γ3 in figure 20.7).
PROBLEM 20.3
What will happen, according to the servo hypothesis, if you try to
activate a muscle to produce a movement but the movement is
unexpectedly blocked (isometric conditions)?
20.5 α-γ Coactivation
The servo model made a number of predictions that could be tested
in experiments. Unfortunately for the servo model, the findings in
those experiments did not follow the predictions. In particular, the
servo model predicts that voluntary movements are initiated by a
change in the activity of γ-motoneurons, while changes in the activity
of α-motoneurons follow at a delay characteristic of the tonic stretch
reflex arc. The development of new methods, in particular direct
recordings from human peripheral nerves pioneered by Swedish
scientist Vallbo (1971, 1981), allowed direct assessment of the
relative timing of changes in the activity of α- and γ-motoneurons
during voluntary muscle contractions. These observations showed
that during virtually all voluntary movements, changes in the activity
of α- and γ-motoneurons happen simultaneously. This phenomenon
is termed α-γ coactivation, and it is illustrated in figure 20.8.
Figure 20.8 The modified version of the servo hypothesis assumes signals are
sent by the brain simultaneously to α-motoneurons and γ-motoneurons (α-γ
coactivation).
In the mid-1960s, researchers were very reluctant to abandon the
servo hypothesis, and they suggested that the servo mechanism
worked as postulated by Merton while voluntary movements were
initiated by a combination of a feedforward command signal (to α-
motoneurons) and a signal to the length-controlling servo (to γ-
motoneurons) (Matthews 1970, 1972).
PROBLEM 20.4
What will happen with the activity of spindle afferents in a flexor
muscle during a voluntary increase in flexion force against a stop?
PROBLEM 20.5
What will happen with the activity of spindle afferents in a flexor
muscle during a fast flexion movement against a constant external
load?
20.7 Equilibrium-Point
Hypothesis
According to the second view, central commands use the
mechanisms of muscle reflexes to induce changes in the levels of
muscle activity and to specify parameters of these reflexes. This
view is compatible with all the observations on the reflex effects on
voluntary muscle activation. On the other hand, it does not go to an
extreme of implying an infinite gain in any of the reflex arcs, and thus
it avoids the problems of the servo hypothesis. This view emerged
as a formal language for describing a body of experimental data on
observations of single-muscle force–length characteristic curves in
animal preparations and single-joint torque-angle characteristic
curves in human subjects (Matthews 1959; Asatryan and Feldman
1965; Feldman 1966; Feldman and Orlovsky 1972; reviewed in
Latash 1993). Animal experiments were typically performed on cats
with a lesion of the central nervous system, so that the cats lacked
the ability to make voluntary movements. Then, an electrical
stimulator was placed on the residual part of the brain. This
stimulator simulated different descending commands.
PROBLEM 20.6
How can one change the velocity of a voluntary movement
according to the equilibrium-point hypothesis?
CHAPTER 20 IN A NUTSHELL
Control signals can be generated in a
feedforward fashion or based on
feedback signals. Positive feedback
tends to amplify the error, while
negative feedback tends to eliminate
errors; however, the exact effects
depend on gains and time delays of
feedback loops. A servo is a
particular case of feedback loop that
keeps output at a preset value.
Merton’s theory considers muscle
spindles to be the sensor in the
perfect servo controlling muscle
length; the theory has been proven to
be wrong. Voluntary activation of a
muscle is accompanied by simultaneous
activation of the α- and γ-motoneurons.
The equilibrium-point hypothesis
considers control of voluntary
movements as a process of central
modulation of the threshold of the
tonic stretch reflex for the
participating muscles. The tonic
stretch reflex and peripheral muscle
and tendon elasticity provide for the
springlike behavior of muscle. Reflex
effects play an important role in
defining levels of muscle activation.
Chapter 21
PROBLEM 21.1
Suggest one or two more examples of the apparently suboptimal
design of the body with respect to movement production.
An alternative philosophical attitude to the design of the human
body is that this design is perfect given the whole variety of tasks
humans face in everyday life. This design allows motor performance
to display both stability and flexibility when external conditions
require a switch in motor patterns. According to this view, we should
not be ashamed of our bodies but proud of their design. And the
central nervous system does not need computational “crutches”;
instead, motor control researchers need better theories.
The differences that have been mentioned between the designs of
the human body and those of artificial systems engineered by
humans should also make one very cautious when trying to apply
schemes developed for the control of artificial systems to the neural
control of movements in humans and animals. Nevertheless,
knowledge of the fundamentals of the control theory is essential for
the first steps in the area of motor control. The exact formulations
and rigorous mathematical apparatus of the control theory are
stimulating examples for motor control researchers who should strive
to develop a comparably rigorous approach for the control of
biological movements.
For now, let us assume that this problem has somehow been
solved, and a target vector ΦT has been found. The starting vector of
joint angles, ΦS, is supposed to be known. The next step of the
problem is to find patterns of joint torques that would move the arm
from ΦS to ΦT. This problem is known as the problem of inverse
dynamics (Hollerbach and Atkeson 1987; An et al. 1988; Atkeson
1989). Potentially, this is another ill-posed problem, partly because
the original and final arm configurations do not define arm trajectory:
There is an infinite number of time functions Φ(t) that can move the
fingertip to the target. Once again, let us, for now, assume that this
problem has also been solved, and appropriate patterns of joint
torques, T(t), have been computed.
However, the controller (the central nervous system) can only
produce active changes in joint torque by sending signals to the
physiological motors, the muscles. So the next step is to compute
patterns of muscle forces, FM(t), that are adequate to produce the
required function T(t). Now, we need to take into consideration that
typically joints are spanned by many muscles. For example, the
seemingly simple elbow joint has three flexors (biceps, brachialis,
and brachioradialis) and three extensors (the heads of the triceps
muscle). Some of these muscles span more than one joint, while
activation of a muscle frequently induces torque production along
several possible axes of joint rotation (Hogan 1985, 1990; Zajac and
Gordon 1989). Besides, torques in individual joints are defined not
only by the forces produced by muscles crossing the joint but by
other sources such as external forces (in particular, gravity) and
torques related to the motion of other joints of the limb. The latter are
commonly addressed as interaction torques (Zatsiorsky 2002).
Once again, assume that this problem is somehow solved by the
very powerful computing device somewhere in the central nervous
system.
The central nervous system produces active changes in muscle
forces by changing the levels of muscle activation A(t). These
functions need to be computed and realized to make sure that the
vector of muscle forces is FM(t). Note that muscle force depends on
both neural signals to the muscle and actual movement kinematics,
which is defined by many factors, including the external forces and
their time changes. That is why this stage requires predicting the
exact changes in muscle length and velocity over the time course of
the movement.
The last stage of the problem is to compute physiological control
signals, U(t), that are going to be sent by the brain to neurons in the
spinal cord to produce required levels of activation of α-motoneurons
innervating the involved muscles. These patterns should take into
account the time-varying excitation received by motoneurons from
peripheral receptors via reflex pathways. So the whole process of
computing descending signals from the brain to the spinal cord can
be represented (in a very much simplified way!) as:
PROBLEM 21.2
Speeding up a motor action has been shown to lead in some
cases to abrupt changes in the relative timing of the
subcomponents of the action. Do these findings disprove the
generalized motor program theory?
PROBLEM 21.3
Is it possible to consider the driving example within the
generalized motor program theory? Is there a “driving program”?
21.4 Equilibrium-Point
Hypothesis: Main Ideas
The ideas of the equilibrium-point hypothesis can be traced back to
studies performed in the beginning of the 20th century. A famous
German physiologist, Kurt Wachholder (reviewed in Sternad 2002)
asked a question that may seem simplistic: How can humans relax
their muscles at different joint positions? Indeed, muscles are known
to have springlike properties (chapter 5). This means that if an
agonist-antagonist pair of muscles acting at a joint are relaxed at a
certain joint position (position A in figure 21.4), one of them will be
stretched and the other shortened when the joint moves to another
position (position B in figure 21.4). The balance of moments of force
will be violated (ΔM ≠ 0), and active muscle force production will be
required to prevent the joint from moving back to position A.
Wachholder measured muscle activity at different joint positions and
found out that humans could indeed relax completely at different
positions. This observation allowed him to conclude that the
muscle’s springlike properties had to be modified by neural
structures.
Later, in the 1950s, this insight was generalized in the form of a
posture–movement paradox (Von Holst and Mittelstaedt 1950). By
that time, it was known that joint postures were stabilized by
neurophysiological mechanisms, some of which have been
discussed in chapters 18 and 19. If a joint is at an equilibrium at a
combination of external load and angle (figure 21.5a, filled circle,
EP0), any deviation from this point (open circle, EP1) leads to the
generation of muscle forces trying to bring the joint back to its
equilibrium position. The question is: How can an active movement
occur without triggering the resistance of these posture-stabilizing
mechanisms?
PROBLEM 21.4
What physiological mechanisms contribute to posture stabilization
of a joint if it is moved away from an equilibrium?
Figure 21.5 (a) An equilibrium position of a joint (EP0) is defined by the joint
moment-angle characteristic (dashed line) and external load. If the joint is moved
to another position (P1) and released, it will tend to return to EP0. (b) If the
commands to joint muscles are changed to correspond to a new equilibrium joint
position, the moment-angle characteristic shifts (from the dotted to the dashed
line), and posture-stabilizing mechanisms are re-addressed to the new position,
the original EP0 becomes a non-EP, and movement occurs to the new equilibrium
point (EP1).
To use the earlier example, when the driver turns the steering
wheel, the driver’s effort is only one of the factors that define how the
steering wheel turns. Another factor is, for example, the availability of
a power-assisted steering system whose gain can be different in
different makes and models.
The main control variable within the equilibrium-point hypothesis,
the threshold of the tonic stretch reflex, is assumed to be expressed
in units of muscle length—that is, in positional units. This assumption
fits well with the general notion that most of our motor tasks are
expressed in units of location in space.
Now that the reader is hopefully accustomed to the idea that
muscle elastic properties are an essential feature of the design of the
human body, return to the seemingly obvious earlier statement that
the central nervous system has to make sure that adequate force
patterns are produced for a required motion. Since force and
displacements are coupled in any elastic system, this statement
implies that movement kinematics is known perfectly in advance.
This is obviously unrealistic, as emphasized by Bernstein many
years ago (Bernstein 1947), for any system equipped with position-
dependent force generators.
PROBLEM 21.5
Some motor control hypotheses suggest that the controller
precomputes patterns of muscle activation or their direct
precursors, such as total presynaptic input to a pool of α-
motoneurons. Do these hypotheses avoid typical problems that
face force-control models?
PROBLEM 21.6
Imagine that finger movements produce an opposite pattern of
movement on a computer screen—that is, when the fingers move
in-phase, the cursors on the screen move out-of-phase, and vice
versa. Which of the two patterns of finger movement would you
expect to switch to the other pattern with an increase in the
movement frequency?
Motor Synergies
where FFL,i are forces of individual flexor muscles, FEX,j are forces of
individual extensor muscles, and Li,j are corresponding lever arms (i
and j stand for the individual flexor and extensor muscles,
respectively). Obviously, this equation cannot be solved. That is why
such problems are sometimes called ill-posed.
A similar problem can be formulated at the level of a single
muscle: Define what motor units should be recruited and at what
frequencies to produce a required level of muscle activation. The
size principle (chapter 7) imposes constraints on possible solutions,
but it does not prescribe a unique solution to the problem.
PROBLEM 22.1
If motor units are always recruited in a fixed order, how can the
same muscle force be produced by different combinations of
motor units?
PROBLEM 22.2
Suggest a couple of examples of the Bernstein problem from
everyday life. Does the controller find unique solutions in your
examples?
PROBLEM 22.3
Another optimization approach has been proposed based on
minimization of a torque-change function similar to jerk
minimization. For what kinds of mechanical systems do the two
criteria lead to identical results?
Figure 22.2 A person producing a sequence of movements will start from slightly
varying initial states (black dots). Trajectories in unstable directions are expected
to diverge (top). Trajectories in stable directions are expected to converge
(bottom).
In addition, body segments are linked mechanically. So, any
unexpected deviation in the motion of a joint produces deviations of
motion in other joints, leading to even more unpredictable
mechanics. Based on this analysis (much more detailed in the
original works!), Bernstein concluded that dynamical stability was
essential for any functional movement and, as a result of evolution
and practice, neural control patterns converged on those that
resulted in dynamically stable trajectories of salient performance
variables such as the trajectory of the fingertip during pointing and
the body’s trajectory during walking. Finding such neural control
patterns and ensuring the stability of movements with respect to
important performance variables with the help of feedback loops was
the second major function of the Level of Synergies.
PROBLEM 22.4
Suggest the smallest and the largest structural unit you can
imagine.
PROBLEM 22.5
Do you like the term “good variability”? It does not hurt
performance, but why is it good, not irrelevant?
PROBLEM 22.6
Can the two methods, PCA and UCM, be combined within one
study? What kinds of questions can be answered by such a
combined approach?
CHAPTER 22 IN A NUTSHELL
The human neuromotor system is
characterized by redundancy at many
levels of analysis. The traditional
approach to the problem of motor
redundancy has been based on the idea
of elimination of redundant degrees of
freedom. Optimization approaches have
commonly been used to solve the
problem. An alternative approach views
the system as abundant, not redundant.
Within that approach, the controller
does not look for a unique solution to
the problem but facilitates families
of equivalent, good enough solutions.
Motor synergies may be characterized
by two features, organizing numerous
variables produced by elements at the
selected level of analysis (elemental
variables) into groups and ensuring
the stability of important performance
variables. The first feature is
commonly explored and quantified using
principal component analysis and
related methods. A quantitative method
to study the stability of performance
has been developed within the
uncontrolled manifold (UCM)
hypothesis. According to this
hypothesis, the controller acts in a
space of elemental variables, allows
large deviations in a subspace (UCM)
corresponding to a desired value of a
performance variable of the whole
system, and limits deviations
orthogonal to that subspace. The
method is based on partitioning motor
variability into “good” (which does
not affect important performance
variables) and “bad” (which does).
Chapter 23
Patterns of Single-Joint
Movements
PROBLEM 23.1
What will happen with the activity of Golgi tendon organs in the
agonist muscle and tendon during a fast movement in isotonic
conditions?
PROBLEM 23.2
What will happen with the level of activity of muscle spindles
(primary and secondary endings) during a muscle contraction in
isometric conditions?
where CJERK stands for the integral jerk index, MT is movement time,
and α is joint angle trajectory (see also chapter 22). This criterion
has been shown to generate trajectories that show all the typical
features of the trajectories of natural movements.
PROBLEM 23.4
Can one and the same muscle be both an agonist and an
antagonist for the same movement (or contraction)?
PROBLEM 23.5
What will happen with the first agonist burst and the antagonist
burst if the subject is performing movements of different
amplitudes within the same movement time?
PROBLEM 23.6
What kind of an EMG pattern do you expect to see if a subject is
performing a fast isotonic movement to a target and quickly back
to the initial position?
23.5 EMG Patterns During
Single-Joint Isometric
Contractions
Two types of isometric contractions have usually been studied: step
and pulse contractions. Step contractions require the subject to
increase joint torque up to a certain level, while pulse contractions
require also returning quickly back to the initial level of joint torque
(commonly, zero net torque). Other instructions have been used to
specify the time of the torque increase, accuracy constraints, and the
like—similar to what has been done in studies of single-joint isotonic
movements. Fast isometric contractions are accompanied by
triphasic EMG patterns similar to those observed during fast isotonic
movements (Ghez and Gordon 1987; Gordon and Ghez 1987;
Corcos et al. 1990). However, the second delayed agonist burst is
more frequently absent. Bursts of the EMG activity become smaller
and poorly defined with a decrease in the rate of torque increase,
and slow increases in joint torque are accompanied by a tonic
increase in the agonist EMG and a smaller increase in the antagonist
EMG.
Figure 23.7 Typical changes in the triphasic electromyographic (EMG) pattern
during isometric contractions with different rates of torque rise.
Adapted by permission from D.M. Corcos, G.L. Gottieb, G.C. Agarwal, and B.P. Flaherty,
“Organizing Principles for Single-Joint Movements: IV. Implications for Isometric
Contractions,” Journal of Neurophysiology 64, (1990): 1033-1042. ©1989 The American
Physiological Society.
Figure 23.11 Schematic illustration of kinematic and EMG patterns for two
movements, both performed over the same amplitude and under the same
instruction to be “as fast as possible.” The inertial load was, however, four times as
high during the second movement. Note the difference in movement kinematics,
which will obviously be reflected in different reflex effects on both agonist and
antagonist α-motoneurons.
PROBLEM 23.7
Can one consider EMG patterns during isometric contractions as
pure reflections of hypothetical central commands? Why?
Multijoint Movement
PROBLEM 24.1
What can you conclude from the fact that the variability of
individual joints is higher than the variability of the tip of a tool
moved by the hand?
PROBLEM 24.2
Give examples from everyday life of the different roles of the
dominant and nondominant arms.
PROBLEM 24.3
You want to perform a single-joint movement in a joint that is
crossed by a uniarticular muscle and a biarticular muscle. What
kind of reflex effects would be helpful for such an action between
peripheral receptors in the uniarticular muscle and α-motoneurons
innervating the biarticular muscle?
Figure 24.5 A spinal frog can remove an irritating stimulus from its back with a
coordinated movement of the ipsilateral hindlimb. The wiping is performed as a
sequence of movements in different directions.
Adapted by permission from M.L. Latash and M.T. Turvey, Dexterity and Its Development
(Mahwah, NJ: Springer and Lawrence Erlbaum, 1996), 286. Reproduced with permission of
The Licensor through PLSclear.
A series of experiments was performed investigating the effects of
unexpected perturbations on the wiping (reviewed in Latash 1993).
In one series, a loose thread loop was placed on the hindlimb,
preventing movements in the knee joint beyond a certain limit so that
the maximal knee joint excursion was about 5°. The frog was able to
remove the stimulus from its back during the first attempt! Then, the
knee was released, and a cast was placed preventing movements in
the next (more distal) joint. The frog once again wiped the stimulus
at the first trial! Next, a lead bracelet was placed on the distal part of
the hindlimb; the weight of the bracelet was similar to the weight of
the hindlimb itself. The frog still was able to wipe the stimulus
accurately.
These experiments show that control signals, even at the spinal
level, are not formulated in terms of contractions of individual
muscles or even movements in individual joints. Otherwise, loading a
segment or blocking a joint would have led to grossly inaccurate
movements. Second, they imply the existence of very fast
corrections of movement patterns that are presumably built into the
sequence of control signals for wiping generated by the spinal cord.
Taken together, the observations point to the neural control being
organized at the level of the working point of the hindlimb (its toes)
directed toward the target. The ability of the frog to wipe the stimulus
off in the presence of major perturbations is an example of the
dynamical stability of the toe trajectory, suggesting that the spinal
cord is able to organize kinematic synergies stabilizing salient
variables (see chapter 22 and later, in section 24.7).
Further investigations of the organization of hindlimb movements
in the spinal frog used electrical stimulation of spinal cord structures
(Giszter et al. 1993; Bizzi et al. 1995). These studies have shown
that the mechanical effects of such stimulation could be described
with a relatively small number of basic force fields known as motor
primitives. Some of the primitives were associated with converging
force fields to a particular equilibrium of the hindlimb’s endpoint in
the body-centered space. Other fields moved the endpoint to an
anatomical limit, which can be a viewed as a consequence of
specifying an equilibrium point beyond the anatomical reach of the
hindlimb. One can question the physiological adequacy of the
electrical stimulation used because it likely activated numerous
groups of interneurons with various functions. However, in general,
these findings point at two important conclusions. First, control of the
hindlimb by the spinal cord uses the equilibrium-point principle (see
chapter 21). Second, control may be simplified by manipulating with
a small number of primitives rather than with a larger number of
muscles (see the discussion of synergies in chapter 22).
PROBLEM 24.5
Where is the referent spatial coordinate for the endpoint of a limb
when pressing with constant force against a rigid object?
24.6 Equilibrium-Trajectory
Hypothesis
The equilibrium-trajectory hypothesis, originally suggested by
Tamar Flash and Neville Hogan (1985), is a natural generalization of
the single-joint equilibrium-point hypothesis (see chapters 20 and 21)
to multijoint movements. It implies that the central nervous system
can shift an image of the working point along a desired trajectory
expressed in external Cartesian coordinates. During a movement,
this virtual trajectory is always ahead of the current actual position of
the working point, and this disparity produces active forces driving
the working point (figure 24.9). Note that the actual trajectory of a
working point may be different from the virtual trajectory due to a
number of factors, including the dynamic properties of the limb and
changes in the external forces (for example, changes in the effects
of gravity during movements that change the orientation of limb
segments). The equilibrium-trajectory approach avoids the
computational problem of inverse dynamics and inverse kinematics,
since muscle forces are not calculated by the central nervous system
but emerge as results of shifts of centrally defined equilibrium
coordinates of the working point, which is done in the external
Cartesian space.
PROBLEM 24.6
What kind of a multijoint synergy can be expected during pressing
by the endpoint against a stationary object with a required force
vector?
Kinematic multijoint synergies stabilizing important variables have
been documented for a range of limb movements, including
reaching, pointing, aiming, and moving an object to be thrown at a
target (Scholz et al. 2000; Reisman and Scholz 2003; Tseng et al.
2003; Yang and Scholz 2005; Mattos et al. 2011; Hasanbarani and
Latash 2020). Multi-muscle synergies stabilizing the salient endpoint
variable during actions by a multijoint limb have also been
documented (Krishnamoorthy et al. 2007).
Figure 24.11 illustrates the time profiles of the two variance
components quantified in the joint configuration space, affecting and
not affecting the orientation of the barrel of the handheld infrared
pistol with respect to the infrared-sensitive target. In other words,
one of the components reflected joint variance within the
uncontrolled manifold (VUCM; see chapter 22), and the other one
reflected joint variance orthogonal to the UCM (VORT). Note that VUCM
> VORT from the very beginning of the movement (i.e., the central
nervous system of the subject stabilized the pistol barrel orientation
even when it pointed at a large angle from the target). Such
strategies—stabilizing salient variables during the whole movement
duration—have been reported in other studies, such as those
involving throwing a Frisbee and a basketball (Yang and Scholz
2005; Hasanbarani and Latash 2020).
Figure 24.11 also implies the existence of RC-based synergies—
for example, stabilization of the task-related variables by covarying
involvement of the r-command and c-command at the level of the
endpoint or at the level of individual joints, or of λs at the level of
muscles. Such studies have not yet been performed during reaching
movements, but RC-based synergies have been documented for
finger-pressing tasks (Ambike et al. 2016; Reschechtko and Latash
2017).
Figure 24.11 Multijoint synergies stabilizing important variables during reaching,
pointing, and shooting tasks are reflected in higher inter-trial variance along the
corresponding uncontrolled manifold (VUCM) as compared to variance along the
orthogonal to the UCM direction (VORT).
CHAPTER 24 IN A NUTSHELL
Reaching movements to a target are
characterized by nearly straight
endpoint trajectories, nearly
symmetrical bell-shaped velocity
profiles, and two-peak acceleration
profiles. Movement trajectories vary
across trials at the same motor task.
Trajectory of the working point
(endpoint) demonstrates lower
variability than individual joint
trajectories, suggesting the existence
of multijoint synergies. Interaction
torques play an important role during
multijoint movements. Interjoint and
intermuscle reflexes provide patterns
of joint interaction that may benefit
everyday movements. The spinal cord
can organize multijoint targeted
movements and associated synergies
stabilizing endpoint trajectory.
Generalization of the equilibrium-
trajectory hypothesis views the
control of a multijoint limb as a
process of shifting of a referent
coordinate for the endpoint. Exact
control variables for multijoint
movements are unknown; they are not
muscle force patterns, patterns of
muscle activation, or joint
trajectories. The motor cortex, the
cerebellum, and the red nucleus are
likely to be very important for the
control of voluntary reaching
movements.
Chapter 25
Postural Control
PROBLEM 25.1
Can one maintain postural equilibrium if the projection of the COM
falls beyond the limits of the area of support?
Figure 25.3During quiet standing, both the COM and the COP show
spontaneous migration known as postural sway.
PROBLEM 25.2
Based on the two views on postural sway, predict what will
happen with sway if a person balances on a board with a narrow
beam glued to its bottom. Will the sway be bigger or smaller than
during standing on the floor? Why?
Figure 25.6 Vestibular nuclei include the lateral vestibular nucleus (also known
as Deiters’ nucleus), the medial vestibular nucleus, the superior vestibular nucleus,
and the inferior vestibular nucleus.
PROBLEM 25.3
Which cerebellar cells project onto the Deiters’ nucleus? Can
these projections be excitatory or inhibitory?
Usually, there are limits to the extent to which the human brain
“believes” any one of the sources of information used for postural
control. Thus, an illusion created by a flow of information along one
of the channels is kept down or may even be eliminated by
information coming from other sources. In order to create a really
strong illusion, one needs to manipulate synchronously all three
major sources of information related to body position and orientation:
vestibular, visual, and proprioceptive. Disney World uses such
methods in their simulation rides during which the viewers are
seated in comfortable chairs that can be tilted synchronously with
changes of an image projected on the large screen.
The second factor is less trivial: The neural controller scales the
magnitude of APAs with the magnitude of an action that the person
uses to trigger a perturbation even if the magnitude of the
perturbation does not depend on that action. Studies have
demonstrated that, if an unusually minor action triggers a large
perturbation, the amplitude of the adjustment may be scaled with
respect to the magnitude of the action, even if the perturbation is
standard and fully predictable (Aruin and Latash 1995). For example,
if a person shoots a rifle, the rebound commonly leads to a strong
postural perturbation. One needs to be an experienced marksman to
be able to generate APAs that would compensate for this predictable
and relatively standard perturbation. This dependence of APAs on
action characteristics may result from everyday experience, which
suggests that more vigorous actions are typically associated with
more vigorous perturbations.
The third factor reflects the dependence of APAs on postural
stability and on background whole-body activity. APAs are reduced in
conditions of both very stable and unstable vertical posture (Nardone
and Schieppati 1988; Nouillot et al. 1992; Aruin et al. 1998). If a
person performs an action associated with a postural perturbation
while walking or swaying, APAs in anticipation of a standard
perturbation are modulated within the cycle of the whole-body action
and may even show reversals of their direction (Hirschfeld and
Forssberg 1991; Krishnamoorthy and Latash 2005). This happens
when APAs themselves may potentially destabilize posture—for
example, if their action is associated with COP shift toward a
dangerously close edge of stability.
PROBLEM 25.4
Suggest an example from everyday life when postural
adjustments are scaled with respect to an action rather than with
respect to an expected perturbation.
The second line of defense is the stretch reflex with its phasic
(mostly monosynaptic) and tonic (mostly polysynaptic) components
(chapters 17 and 18). The stretch reflex also demonstrates
viscoelastic properties and contributes to damping external
perturbations, although at a certain, relatively short, reflex delay.
However, these two mechanisms are not enough to ensure
equilibrium of the body in the field of gravity. The next defensive
mechanism has a longer delay and belongs to the group of
preprogrammed reactions (see chapter 19). It is more powerful and
more flexible than the first two mechanisms. In particular, in certain
situations, preprogrammed responses can be seen in muscles
whose length is unaffected or even decreased by an external
perturbation (Marsden et al. 1979; Nashner et al. 1979; Nashner and
Cordo 1981). Such posture-stabilizing reactions can be seen, for
example, in arm muscles when a person stands in a bus and the bus
unexpectedly starts moving.
In laboratories, unexpected platform rotations or translations are
commonly used as sources of controlled postural perturbations.
Preprogrammed reactions to such perturbations are frequently
described as combinations of muscle activation patterns specific for
a given perturbation. The earliest of these reactions appear at
relatively short latencies, less than 80 ms, suggesting their
preprogrammed rather than voluntary nature. They are supposed to
be triggered by multimodal sensory inputs with an important
contribution from proprioceptive, visual, and vestibular receptors.
Some of these reactions seem to be rather general—for example,
coactivation of agonist–antagonist muscle pairs stabilizing a postural
joint irrespective of the direction of a perturbation. Other reactions
are specific to the type and direction of a perturbation.
There are certain preferred patterns of corrective postural
reactions. In young, healthy subjects standing on the force platform,
a relatively slow forward translation of the platform induces a body
sway backward, leading to an increase in the background activity of
the ventral muscles (such as the tibialis anterior, rectus femoris, and
rectus abdominis). On the other hand, a backward translation of the
platform results in a body sway forward and an increase in the
background activity of the dorsal muscles (such as soleus, biceps
femoris, and erector spinae). These responses occur at a delay of
about 80 ms in a distal-to-proximal order. This pattern of muscle
activation and the accompanying kinematics have been called the
ankle strategy (Horak and Nashner 1986).
When healthy subjects stand on a surface fitted with support that
is narrow in the anterior–posterior direction or when the platform is
translated very quickly, the order of muscle recruitment is reversed to
proximal-to-distal, known as the hip strategy (Horak and Nashner
1986; Woollacott and Shumway-Cook 1990). The hip strategy is also
seen in elderly persons in conditions when young persons typically
show the ankle strategy (Woollacott et al. 1988).
Because of the geometry of the human body, an ankle movement
leads to a larger horizontal displacement of the COM as compared to
a hip movement of the same amplitude. On the other hand, if a
larger or a smaller-than-optimal joint rotation is generated, the ankle
strategy is likely to lead to a bigger error in the displacement of the
COM and consequently to a higher possibility of losing balance.
Thus, the ankle strategy seems to be more effective and challenging,
while the hip strategy trades efficacy for safety. There are also
intermediate patterns of corrective postural adjustments that may
involve the knee joint (Allum et al. 1989).
PROBLEM 25.5
Can one see anticipatory and corrective postural adjustments in
arm muscles? Give examples.
PROBLEM 25.6
What changes in anticipatory and corrective postural reactions do
you expect to observe if one and the same perturbation is applied
many times to a standing person?
PROBLEM 25.7
When a car makes a turn at a high speed, the upper bodies of the
driver and the front seat passenger deviate in opposite directions,
toward each other or from each other. Why? If a car makes a left
turn, in what directions will the upper bodies of the driver and the
passenger tilt?
Figure 25.10 According to the traditional view, two motor commands are
generated by the central nervous system based on the task and external force
fields. One is related to the “focal” movement, while the other provides postural
stability. They target different subsets of joints.
Figure 25.11 A motor command is generated by the central nervous system
based on the task and external force fields. It is later transformed to commands to
individual joints. Some of them may have an apparently postural function, while
others may be “focal.” This classification, however, is made by the researchers, not
by the brain.
CHAPTER 25 IN A NUTSHELL
Human vertical posture is inherently
unstable. Quiet stance is associated
with postural sway. Sway depends
strongly on sensory signals; it
increases in the absence of vision and
decreases when a person touches a
stable point in space. Voluntary limb
movements generate postural
perturbations because of joint
coupling and changes in body geometry.
Postural stabilization is helped by a
number of mechanisms, including APAs,
peripheral elasticity of muscles and
tendons, muscle reflexes,
preprogrammed postural corrections,
and voluntary corrections. Postural
synergy is a relation among elemental
variables, such as joint rotations or
recruitment of muscle groups, with the
purpose to stabilize salient
performance variables such as body
orientation in space. The vestibular
apparatus plays a major role in
postural equilibrium; it consists of
peripheral receptors sensitive to
acceleration and vestibular nuclei.
Other sensory systems, such as vision
and proprioception, also contribute to
postural control. Changes in visual or
proprioceptive information can lead to
postural disturbances.
Chapter 26
Locomotion
PROBLEM 26.1
Some animals can participate in different kinds of locomotion,
such as swimming, walking, hopping, crawling, and flying. How
many CPGs do these creatures have?
PROBLEM 26.2
Imagine that the input from the brain in figure 26.1 is completely
absent. Which output will the network produce?
Figure 26.2 A CPG can be driven both by descending signals from “higher
centers” and by afferent information. Ultimately, the CPG leads to changes in the
patterns of activation of α-motoneuronal pools.
PROBLEM 26.3
Are there spinal CPGs for vertical posture? Give reasons for your
answer.
PROBLEM 26.4
What are, from your point of view, the most important limiting
factors for the maximal speed of locomotion?
Note that figure 26.6a lacks coordination or, more precisely, all the
details of coordination are delegated to the ultimate controller; in
other words, they are assumed to be preplanned by the supreme
omnipotent homunculus. This is not very attractive for a number of
reasons, in particular because, as we know, similar coordination
patterns can be observed in spinal animals that apparently lack any
homunculus. Besides that, assigning all the details of coordination to
a smart “black box” does not solve the problem but rather
emphasizes our inability to deal with it.
Figure 26.6b illustrates a much more appealing approach to
coordination, which can emerge without any supreme problem
solver, but it lacks the element of control. This fish will never be able
to change its behavior based on its own will, only in response to
signals from the environment. Remember that in chapter 21 we
discussed the differences between the physiology of initiative and
the physiology of reflex-type movements and came to the conclusion
that behavior cannot be based exclusively on reactions to external
stimuli. Thus, this image is also unsatisfactory.
Figure 26.6c shows a “hybrid fish” that retains all the coordinative
links among its elements but also has an independent descending
signal, generated by its upper neural structures, that can be used for
movement initiation or modification even if the environment does not
dictate it. This descending signal represents one of the important
parameters in the equations that describe the coordination.
Bernstein’s principle of initiative states that this input cannot be
reduced to reactions to external stimuli. At this time, its nature
remains mysterious. This may not sound very scientific, but
unfortunately, there seem to be no viable alternatives to the scheme
shown in figure 26.6c. Coordination and control can and must
coexist to allow both active central generation of meaningful
movements and adjustment of coordination to changes within the
body and the environment.
PROBLEM 26.5
In classic experiments by Scott Kelso, when the subjects tapped a
rhythm with two index fingers, an increase in the tapping
frequency could lead to an involuntary switch from an out-of-
phase regime into an in-phase regime. Try to interpret these
observations based on the three approaches illustrated by the
three parts of figure 26.6.
PROBLEM 26.6
Imagine that a person stands on a board placed on a very narrow
support that does not allow the COP to shift in the anterior–
posterior direction. How can a person initiate a step in such
conditions?
26.9 Corrective Stumbling
Reaction
A particular pattern of automatic, reflex-like responses was observed
during cat locomotion associated with overcoming an unexpected
obstacle, the corrective stumbling reaction (see chapter 19). This
pattern could be observed during a weak mechanical stimulation of
skin areas of the paw or of the leg (even with an air puff) or during
short episodes of electrical stimulation of skin nerves or dermatomes
(Duysens and Pearson 1976; Forssberg et al. 1975, 1977). Similar
patterns have also been seen in humans (Lisin et al. 1973).
The application of any of these stimuli during the swing phase
gave rise to a flexor reaction with the hindlimb transfer over a
hypothetical obstacle (figure 26.9b). The same stimulation applied
during the stance phase could give rise to an extensor reaction
(figure 26.9a). The latency of these reactions was higher than the
latency of monosynaptic reflexes and lower than the voluntary
reaction time. The functional appropriateness of these reactions and
the relative independence of the stimulus suggest that they are in
fact preprogrammed responses of a mechanism responsible in
everyday life for the compensatory reactions during stumbling (see
chapter 19).
Figure 26.9 The corrective stumbling reaction involves a coordinated adjustment
of the walking pattern in response to an external stimulus (mechanical or electrical)
applied to one of the feet (or paws). It involves different changes in the muscle
activation patterns and different kinematic consequences, depending on the phase
of the step cycle when the stimulus is delivered.
PROBLEM 26.7
When you suddenly step on a nail while walking, preprogrammed
reactions are generated (e.g., jumping). What kind of reaction
would you expect to see in a person stepping on a nail while
crossing an abyss on a narrow plank?
PROBLEM 26.8
Give a couple of examples of other tasks that are associated with
a system of preprogrammed corrections.
CHAPTER 26 IN A NUTSHELL
Locomotion is an activity during which
the location of the whole body of the
animal in the environment changes.
Electrical stimulation of an area in
the reticular formation of the
midbrain and also of areas in the
cervical spinal cord can induce
locomotion and lead to changes in
gait. Peripheral stimulation of the
legs as well as certain drugs can lead
to gait initiation and changes in a
spinal animal preparation. Locomotion
can be viewed as a result of activity
of a CPG that is under control of a
higher executive structure and can
change its activity in response to
changes in peripheral information.
Alternatively, locomotion can be
viewed as an emerging pattern within a
complex system that involves central
neural structures, peripheral organs,
and interactions with the environment.
There is evidence for the existence of
spinal CPGs for locomotion in humans.
To initiate a first step, the COP
shifts show a characteristic pattern
that unloads the stepping leg and
creates a moment of reactive force
that rotates the body forward. Reflex
responses of limb muscles to
mechanical stimulation of a paw or to
an electrical stimulation of an
afferent nerve in a limb depend on the
phase of locomotion. This pattern is
known as the corrective stumbling
reaction.
Chapter 27
Prehension
Each of the extrinsic hand muscles has four distal tendons that
attach to individual fingers. In particular, the flexor digitorum
superficialis (FDS) sends four tendons to the four fingers of the
hand. These tendons attach at the intermediate phalanges (figure
27.2). The other major flexor, flexor digitorum profundis (FDP), has
four tendons that attach to the distal phalanges of the four fingers.
This design of the muscular hand apparatus may be viewed as
contributing to the interdependence of forces and movements
produced by individual fingers. For example, if you try to flex the ring
finger, other fingers will also show some flexion movement. However,
this interdependence is not absolutely obligatory; professional
musicians can learn to override it and produce much more
independent finger actions (Slobounov et al. 2002).
Extrinsic hand muscles are commonly viewed not as single
physiological structures but as combinations of several muscle
compartments (Jeneson et al. 1990; Serlin and Schieber 1993).
The idea of compartments implies the existence of subgroups of
motor units that produce forces transmitted primarily to one of the
four tendons and lead to motor effects in one of the four fingers. This
idea allows for both independent and combined finger action. It has
received support in a study that showed that force response by a
finger to a single transcranial magnetic stimulus (TMS) applied over
the primary motor area of the cortex depends strongly on the
background force produced by this particular finger and shows only
weak dependence on the forces produced by other fingers of the
hand (Danion et al. 2003a).
PROBLEM 27.1
How will the force response of a finger to a single TMS pulse
applied over the primary motor cortex depend on the background
force produced by that finger?
PROBLEM 27.2
What kinds of changes in responses of a muscle to a standard
TMS stimulus applied over the primary motor cortex can be
expected after a prolonged practice of this muscle or of its
antagonist?
PROBLEM 27.3
What changes of enslaving and force deficit would you expect to
see in the right and left hands of a trained right-handed cellist?
Figure 27.6 To take a sip from a glass, the grip force (FGRIP) has to be within a
rather wide range, over the slipping level and below the crushing level. The
moment of forces (M) should be controlled precisely to avoid spilling the contents
of the glass.
PROBLEM 27.4
Suggest a natural task in which the production of accurate finger
forces would be more important than the production of accurate
moments of forces.
27.5 Grasping
During object manipulation, the hand can be used to produce
different types of grasps (Schlesinger 1919; Napier 1956), from a
precision grip (holding a small object between the tips of the index
finger and the thumb) to a power grip (using a wrench). Kinematic
analysis of grasping commonly uses the notion of grip aperture, the
distance between the tips of the thumb and the index finger. When a
hand reaches for an object, grip aperture increases up to about 60%
to 70% of the duration of the reaching movement, and then the digits
start to close to land on the object. The maximum grip aperture
depends on the properties of the object to be grasped, including its
size, weight, and texture. Grasping actions strongly depend on
sensory information, including visual information and signals from
somatosensory receptors in the muscles, tendons, joints, and skin.
There is an alternative view on grasping that views it as a
combination of two pointing movements of the opposing digits
directed at certain landing points on the object (Smeets and Brenner
1999).
Brain imaging studies (as well as more direct studies of the brain
activity in monkeys) have suggested the importance of three cortical
areas for grasping: the primary motor cortex, the premotor cortex,
and the parietal cortex. Studies of the neural activity in the ventral
premotor cortex of the macaque monkey have shown the existence
of neurons that discharge both during the execution of hand actions
and during the observation of the same actions made by others:
mirror neurons (Rizzolatti et al. 2001; Rizzolatti and Craighero
2004). Of course, one has to keep in mind that the actions were “the
same” at a very high, abstract, topological level: At the level of
mechanics, such as joint rotations and muscle forces, the actions
could not be the same because of the obvious differences in the
anatomy and physiology of the monkey and the experimenter.
Indirect evidence suggests that such neurons exist in humans as
well; in particular, it has been shown that grasp observation is
accompanied by significant activation of the middle temporal gyrus of
the cortex, including that of the adjacent superior temporal sulcus
(Brodmann’s area 21) and the caudal part of the left inferior frontal
gyrus (Brodmann’s area 45) (Fadiga et al. 2005).
PROBLEM 27.5
Suggest a functional explanation for the existence of the mirror
neurons.
Figure 27.9 Hierarchical control of the hand involves two levels. At the first level,
mechanical actions of the thumb and the virtual finger (VF) are defined. At the
second level, the mechanical action of the VF is distributed among the actual
fingers.
Reprinted by permission from J.K. Shim, Rotational Equilbrium Control in Multi-Digit Human
Prehension. (Ph.D. Dissertation, 2005). © 2005, J.K. Skim.
Experimental studies of variations in the individual finger forces
across repetitions of the same task have shown that humans indeed
satisfy the three equations (Shim et al. 2003). However, they do so in
a particular way. This is possible because the number of variables in
the equations is more than three; that is, the system is abundant.
Note that the third equation contains variables that are also parts of
the first two equations. So, one could expect that a change in the
rotational action (equation 27.4) would lead to changes in
components that produce adjustments in grasping action and net
force generation (force production, equations 27.2 and 27.3).
Nevertheless, experiments show that humans organize the
elemental variables into two groups. Within one group, the variables
covary to stabilize the total force acting on the object. The other
group unites variables that covary to stabilize the rotational effect
produced by the hand on the object. So there seem to be two
prehension synergies, force-stabilizing and moment-stabilizing. The
two synergies seem to be relatively independent, so a change in one
of the variables (for example, the total moment) requires adjustment
only in the corresponding synergy without affecting the other (force-
stabilizing) synergy.
These observations fit the principle of superposition suggested
in robotics for the control of artificial grippers (Arimoto et al. 2001).
The principle states that skilled actions can be decomposed into
several elemental actions that are controlled independently by
separate controllers. Such a decoupled control in robotics has been
demonstrated to reduce the total computation time. Human
experiments suggest that the central nervous system may also use
the principle of superposition to organize the control of two elemental
actions: (1) grasping a handheld object with an adequate force and
(2) producing a rotational action on the object.
PROBLEM 27.6
A change in the force of a digit always produces a change in the
moment of force about any axis with respect to which the force
vector has a nonzero lever arm. How can control of the grip force
be decoupled from control of the moment of forces?
CHAPTER 27 IN A NUTSHELL
The human hand combines both parallel
and serial kinematic chains. Digit
action is controlled by extrinsic,
multidigit muscles and intrinsic, more
digit-specific muscles. The extrinsic
muscles are composed of compartments
that show a high degree of
physiological independence. Cortical
hand representations are mosaic
without clear borders between
representations of anatomically close
areas. Finger interactions are
reflected in unintended movements and
force production by fingers
(enslaving) and lower forces produced
by fingers in multifinger tasks as
compared to single-finger tasks (force
deficit). These interactions have a
strong neural component. They show
similar patterns among fingers of a
hand and between the thumb and the
fingers. Multifinger synergies in
pressing tasks commonly show patterns
stabilizing the pronation and
supination moment. These patterns may
be conditioned by everyday experience
that places strict accuracy
constraints on the rotational hand
action. During manipulation of a
handheld object, multidigit synergies
show patterns that stabilize both the
grip force and the total moment of
force applied to the object. The
existence of these two synergies
conforms to the principle of
superposition suggested in robotics.
Problems for Part V
Self-Test Problems
1. A subject is performing a series of very fast elbow flexions
over 40° against a constant external load. Movement time is
200 ms. Unexpectedly, in one trial, the movement is
completely blocked. Draw time patterns of the biceps and
triceps EMGs based on central programming of EMG patterns
and on equilibrium-point control ideas.
2. A person sits with the right forearm, hand, and index finger
vertical and pointing up. The hand is fully supinated. The
person makes a very fast elbow extension movement over
about 90° in response to a visual signal. Movement time is
250 ms. Describe (or draw) EMG time profiles for the following
muscle pairs: elbow flexor-extensor, shoulder flexor-extensor,
and wrist flexor-extensor (assume that all muscles are
uniarticular).
3. A student is standing quietly. Unexpectedly, a friend pushes
them from behind. Describe all the mechanisms that help the
student not to fall down.
4. While walking barefoot, a person steps unexpectedly on a
sharp object. Describe muscle activation patterns that help
this person to avoid falling down. What differences in muscle
activation patterns would you expect in the other leg?
5. High-frequency, low-amplitude muscle vibration applied to the
back of the neck of a person, who is standing with his or her
eyes closed and the arms hanging by the sides, can lead to a
deviation of the body posture from the vertical. Why? In what
direction would you expect the body to move?
6. A person makes a series of quick reaching movements to a
target while holding a 1 kg load in one hand. In one trial, the
load unexpectedly becomes 3 kg. The person is instructed not
to correct actions voluntarily. Describe changes in the
grasping and reaching components of the action and their
typical time delays. Where will the movement terminate?
For Those Addicted to Multiple-Choice Tests
You have 20 minutes. Circle only one answer (statement) for each
question. Write a short phrase explaining why you chose this
answer.
1. During a very fast voluntary isotonic movement
a. the length of the muscle fibers does not change
b. antagonist burst occurs a few tens of milliseconds prior to
the agonist burst
c. damping properties of muscles and tendons help
accelerate the movement
d. muscle fiber stiffness is higher than in the relaxed state
e. none of the above
Why?
2. You saw a huge bag and expected it to be very heavy. You
lifted it quickly with the right arm, but the bag happened to be
very light (empty), and you nearly lost your balance. What
phenomena caused the problem with the balance?
a. elastic properties of muscles and tendons
b. monosynaptic reflexes from muscle spindles
c. monosynaptic reflexes from Golgi tendon organs
d. anticipatory postural adjustments
e. preprogrammed responses
Why?
3. A person moves a handheld object rhythmically in a vertical
direction at a very high speed (accelerations over 1 g). Which
of the following can be expected?
a. Grasping force will show rhythmic changes at the
frequency of the motion.
b. Grasping force will remain constant.
c. Grasping force will change at twice the motion frequency.
d. Grasping force will be modulated based on proprioceptive
feedback.
e. All of the above are correct.
Why?
4. When a person performs a series of very fast voluntary
reaching movements
a. joint trajectories are ideally reproduced
b. many muscle pairs show triphasic activation patterns
c. muscle reflexes do not contribute to muscle activation
d. muscle force patterns are ideally reproduced
e. all of the above
Why?
5. The locomotor central pattern generator in an animal with the
spinal cord surgically separated from the brain at a cervical
level
a. helps the animal to maintain the posture needed for
locomotion
b. can be driven by stimulation of certain brain structures
c. can lead to changes in locomotion patterns in response to
sensory information
d. is unable to produce rhythmic patterns of activity
e. all of the above
Why?
Part VI
Kinesthetic Perception
PROBLEM 28.1
Are muscle reflexes, including monosynaptic ones, examples of
perception–action coupling?
where k is a constant.
For example, you may be able to distinguish a weight of 2 N from
another weight when it differs from the first one by 0.2 N. If the
original weight is now 4 N, you would need a difference of 0.4 N to
be able to make such a distinction. This law works across sensory
modalities (e.g., for brightness of light, loudness of sound, intensity
of pain, etc.). It fails at very low and very high magnitudes of the
physical variables but is valid within a broad midrange of
magnitudes.
Fechner offered a generalization of Weber’s law and stated that
the intensity of our sensation (S) increases as the logarithm of an
increase in the relevant physical variable:
Figure 28.2 In isometric conditions, muscle fiber length changes (shortens with
force increase) while the “muscle plus tendon” length remains unchanged.
PROBLEM 28.2
Does muscle and tendon elasticity affect signals from Golgi
tendon organs? Do these receptors always accurately monitor
tendon force?
Figure 28.5 The scheme for the combined perception of muscle length and
force. Setting a value of RC (λ for a muscle) defines the force–length
characteristic. The efferent signal (EFF) defines a referent point, while signals from
afferents (AFF) define deviation of the muscle from RC. Deviations from λ along
the characteristic lead to a parallel increase in signals from a number of sources,
including muscle spindles, Golgi tendon organs, and α-motoneurons.
Figure 28.6 A schematic of the afferent signal transmission along the dorsal
columns of the spinal cord, via the lemniscus, to the ventroposteriorlateral (VPL)
nucleus of the thalamus.
28.5 Vibration-Induced
Kinesthetic Illusions
High-frequency muscle vibration (e.g., at 100 Hz) creates a large
number of motor and sensory effects mediated primarily by the very
high induced activity of the primary spindle endings. These sensory
endings are highly sensitive to velocity and can be driven by
relatively low-amplitude vibration (e.g., with the peak-to-peak
amplitude of 1 mm applied to the belly of a muscle or to its tendon);
this means that the primary endings generate one or several action
potentials in response to each cycle of vibration (Matthews and Stein
1969). Motor effects, including the tonic vibration reflex, were
discussed in chapter 18. In this section, we will discuss the effects of
vibration on kinesthetic perception.
Most commonly, muscle vibration leads to kinesthetic illusions
corresponding to perceiving joint positions associated with longer
length of the muscle subjected to vibration (Goodwin et al. 1972; Roll
and Vedel 1982; see figure 28.7). Such illusions can be very strong
and can even lead to a perception of anatomically impossible joint
configurations (Craske 1977). Interpretations of such vibration-
induced kinesthetic illusions (VIKIs) have linked it to the unusually
high activity of primary spindle endings. The velocity sensitivity of
these endings probably plays an especially important role. Indeed,
sometimes humans report very high velocities of illusory movements
produced by muscle vibration while illusions of joint displacement are
relatively modest, much smaller than what could be expected from
the reported joint velocity (Sittig et al. 1985).
Figure 28.7 High-frequency muscle vibration can lead to misperception of the
position in a joint crossed by the muscle. In this illustration, the person is trying to
match the elbow joint positions under the vibration applied to the left elbow flexor.
Trying to match the joint position of the left elbow (αL) commonly leads to showing
a joint position in the right elbow (αR) that corresponds to longer elbow flexors.
PROBLEM 28.3
Imagine that two vibrators are placed over the triceps surae and
tibialis anterior. What effects would you expect to observe in a
standing person?
Figure 28.9 (a) Vibrating a flexor muscle can lead to misperception of the length
of this muscle without a change in perception of the extensor muscle. (b) This
should lead to perception of an anatomically impossible situation. (c) Subjects
typically report an intermediate percept corresponding to length changes in both
muscles, although only one of them is subjected to vibration.
PROBLEM 28.4
How would you interpret the reports of somewhat decreased force
under muscle coactivation in figure 28.10?
Figure 28.11 Two patterns of dots are alternating at a regular pace (black circles
and gray circles; imagine that both sets are of the same color). When a person
watches this screen, he or she typically perceives that the dot pattern rotates
clockwise or counterclockwise, commonly with changes in the direction of dot
rotation. If the person imagines holding a steering wheel, rotating this imaginary
wheel drives the illusion: The dots start to move in the direction of wheel rotation.
CHAPTER 28 IN A NUTSHELL
Sensation of physical variables
follows the Weber-Fechner law and the
Stevens law. Perception of body
configuration and forces at the
interface with the environment is
known as kinesthesia. Kinesthetic
perception is based on an interaction
of efferent (motor) and afferent
(sensory) signals. Signals from
peripheral sensory endings deliver
ambiguous information that requires
further processing to produce
kinesthetic percepts. These signals
are transmitted via the dorsal spinal
column to nuclei in the medulla and
then, via the lemniscus, to the
thalamus and further to somatosensory
areas in the parietal cortex.
According to one theory, efferent
signals participating in perception
(efference copy) define a referent
spatial coordinate, and sensory
signals define deviations from this
coordinate. This theory links the
perception of forces and coordinates
into a single scheme. Artificial
changes in sensory signals (e.g., by
muscle vibration) lead to complex
patterns of kinesthetic illusions as
associated phenomena, such as postural
deviations. The stability of percepts
suggests the existence of a stable
subspace (iso-perceptual manifold,
IPM) in the combined afferent–efferent
space. Coupling between perceptual and
motor processes takes place at
different levels, from muscle reflexes
to complex actions.
Chapter 29
Multisensory Integration
PROBLEM 29.1
In the mammalian nervous system, many types of multisensory
neurons have been found that respond to a combination of visual,
auditory, vestibular, and somatosensory inputs. This allows many
biological events to be registered by more than one sense, and
animals can use these senses synergistically. Suggest one or two
good hypotheses as to why the nervous system would develop
both unisensory and multisensory neurons.
PROBLEM 29.2
The theory of Bayesian integration suggests that the nervous
system integrates multisensory information based on its relative
reliability. What does optimal integration mean to you? What
would suboptimal integration look like? Could you relate this to
any real-world experience or example?
Figure 29.2 The visual and proprioceptive systems process sensory information
with different delays. The proprioceptive system is faster but has higher levels of
sensory noise.
So how does the nervous system reconcile temporal delays
across vision and proprioception that can differ by tens of
milliseconds (see figure 29.2)? This question was addressed by
Crevecoeur and colleagues in an elegant study (Crevecoeur et al.
2016). The authors applied either mechanical perturbations to the
hand or visual perturbations to the hand position feedback. In some
conditions, they applied both perturbations. Their main outcome
variables of interest were the long-latency reflexes (M3; also see
chapter 19), saccadic reaction times, and limb kinematics. They
used saccadic eye movements as proxies for estimating participants’
hand motion and showed that participants relied more on
proprioceptive feedback than on visual feedback to estimate hand
motion. They also showed that vision mostly influenced fine motor
control near the end of movement, whereas limb motor corrections in
the early stages of the movement appeared to be primarily driven by
proprioceptive feedback. These results suggest that the nervous
system accounts for the differential temporal conduction delays in
different sensory modalities for limb motor control.
Figure 29.3 Participants make reaching movement from the solid dark circle to
the solid light circle. During the movement, a new target is presented (dashed light
circle), but the hand is constrained to be within the “force channel” (shown as a
dashed rectangle). Within this channel, the robot applies a strong elastic and
viscous force to prevent the hand from escaping the channel. Inset shows the
experimental system used in these studies.
PROBLEM 29.3
The flash-lag effect occurs when a nonmoving object is quickly
flashed directly next to a moving object. Though they are
presented in the same location at the same time, the two stimuli
are perceived to be displaced from one another. Specifically, in
this visual illusion the nonmoving object is perceived as “lagging”
the moving object. What do you think would cause this illusion?
Figure 29.5 The difference vector between the target and the hand (XTarget-Hand)
is first transformed from a retinotopic coordinate frame to a joint (shoulder-based)
coordinate frame. The crosshairs show the current location of the gaze. The
dashed lines show the locations of the targets in an eye-based coordinate frame.
The dotted lines show the locations of the targets in a shoulder-based coordinate
frame.
PROBLEM 29.4
The parietal cortex stores object locations for reaching in a
retinotopic reference frame. So if a saccade is made during an
ongoing reaching movement, the location of the object would be
remapped in a retinotopic reference frame. Would that affect the
sensorimotor transformations associated with the planning of
reaching movements? Would that affect the trajectory of the
reaching movement?
PROBLEM 29.5
Humans can walk on flat surfaces with their eyes closed for short
periods of time, but it is almost impossible to do tightrope walking
with eyes closed. Why is that, and what does that tell us about
multisensory integration and conduction in the visual and
vestibular sensory systems?
Figure 29.6 Multiple sensory modalities contribute to maintaining postural
equilibrium. At any given instance, self-motion signals from the visual system,
proprioceptive information from the neck and lower limb, and vestibular information
are simultaneously integrated. This helps us resolve perceptual ambiguities in real
time, especially when the sensory receptors receive conflicting information from
different sensory modalities. Resolving these conflicts is necessary to be able to
walk in a stable fashion on uneven terrain.
PROBLEM 30.1
How is cerebral achromatopsia different from the color blindness
discussed in chapter 14?
Figure 30.1 Large-field visual motion elicits strong neural activity in neural areas
MT and MST. (left) Large-field visual motion of random dots where all dots are
moving in the same direction to the left (100% coherence). (right) All dots are
moving randomly in different directions. In typical experiments, 3% to 50% of the
dots move in the same direction and the rest move randomly.
MST is also located in the superior temporal sulcus. MST neurons
receive input directly from MT. There are two types of neurons in
MST. First, neurons with smaller receptive fields are important for the
execution of smooth-pursuit eye movements. Recall that smooth-
pursuit eye movements are slow eye movements (maximum speed
of 80º/s to 100º/s) that are involved in tracking the motion of moving
stimuli. Second, MST neurons with larger receptive fields are
involved in the perception of optic flow and self-motion. In addition,
MST also receives input from parietal areas that are involved in
processing vestibular signals. This helps with distinguishing motions
that arise from external sources versus those that arise from self-
motion. Thus, MST plays an important role in distinguishing different
sources of motion, self-generated and externally generated motion.
Damage to MT/MST (through injury) or temporary inactivation
(through experimental manipulation) causes deficits in visual motion
perception. This can range from mild deficits, such as reduction in
the accuracy of direction discrimination at small spatial
displacements (Rudolph and Paternak 1999) to more pronounced
disorders, such as akinetopsia. A quarterback who had this deficit
would see an opposing linebacker moving toward them as “jumping”
from one stationary position to another, instead of running toward
them in one continuous motion.
PROBLEM 30.2
Which other activities of daily living are critically dependent on
simultaneous neural processing in both the dorsal and ventral
visual streams?
PROBLEM 30.3
If your friend threw a ball toward you from a distance to catch,
which eye movements would you be making to track the ball?
30.7 Roles of Two Visual
Streams in Eye–Hand
Coordination
Before limb movements are initiated, the visual system must select
an action goal from among many viable alternatives. For example,
the quarterback could throw to any of the teammates who might be
free to receive the ball. But a good quarterback would also assess
which of those teammates have the maximum chance of carrying the
ball farthest downfield. Thus, the expectation of reward or outcomes
also determines which action we select in a situation. The cortical
areas represent targets for action (dorsal visual stream) and
perception (ventral visual stream) and act with the reward centers
(basal ganglia) to execute complex movements. The frontoparietal
cortical areas along the dorsal stream facilitate automatic control of
action, and the inferotemporal areas along the ventral stream also
contribute to conscious judgments of those actions.
The PPC is a critical node in the dorsal visual stream responsible
for the automatic control of eye and hand movements. The main role
of the PPC is target localization for static targets (reaching
movements) and rapid spatial updating of target locations for
dynamic targets (interception movements). Specific areas within the
intraparietal sulcus (IPS) serve as interfaces between the sensory
and motor systems for controlling arm and eye movements
(reviewed in Grefkes and Fink 2005; Vesia 2012). Motor commands
for reaching movements are prepared based on input from both
reach-related regions (such as the parietal reach region) and eye
movement–related regions (such as the LIP) in the parietal cortex.
These regions communicate with frontal areas involved in limb motor
planning (premotor cortex) and eye movements (FEF). Once
movement is initiated, the frontoparietal circuits are simultaneously
activated (Naranjo et al. 2007) to minimize visuomotor errors through
continuous comparisons between target locations and hand
trajectory (Franklin et al. 2016; Binsted et al. 2010).
The role of the ventral visual stream in eye-hand coordination is
not entirely clear. But it is likely that it plays a role in grasping during
reach-to-grasp movements. If we are trying to grasp a cup versus a
pen with a precision grip, the grasp component of the movement
would be different in the two cases. The precision grip is facilitated
by canonical neurons in the ventral premotor cortex. These neurons
discharge during motor execution and in response to the
presentation of three-dimensional objects (Rizolatti et al. 1988).
These neurons receive extensive input from the lateral parietal areas
and are considered to encode important object dimensions that allow
us to calibrate our digits to the dimensions of the objects. The
canonical neurons most likely receive input from the ventral visual
stream through the ventrolateral prefrontal cortex (reviewed in
Kravitz et al. 2013; Thorpe and Fabre-Thorpe 2001). This pathway
from the ventral visual stream to the frontal cortex and ventral
premotor cortex would act in a feedforward fashion to shape the
grasp aperture in preparation for contact with the object of interest
(see figure 30.3).
Figure 30.3 The neural network involved in judging object shape to plan and
execute the finger movements for the grasp component of the reach-to-grasp
movement. IT = inferotemporal cortex; PFC = prefrontal cortex; PMd = dorsal
premotor cortex; M1 = primary motor cortex.
Adapted from S.J. Thorpe and M. Fabre-Thorpe, “Neuroscience - Seeking Categories in the
Brain,” Science, 291, no. 5502 (2001), https://doi.org/10.1126/science.1058249. PMID:
11253215.
Figure 30.5 Typical task design for online feedback control experiments.
Participants start from rest. Then a peripheral target (shown on the right) comes
on, and the participant initiates a movement toward the target. While the
movement is underway, a secondary target comes on and the participant aborts
the previous motor plan and moves toward the new target. Inset shows the entire
experimental setup.
The role of the PPC for online feedback correction was first
studied by Mountcastle and colleagues. They showed that certain
neurons in the PPC only fired during reaching movements to a target
(Mountcastle et al., 1975). These neurons did not fire either when
the target was presented, or if the arm moved toward a nonvisual
target. The neurons fired only when the arm moved toward a visual
target and stopped firing once the arm reached the target. This
suggests that the neurons encode the error signal between the arm
and the target. Correction of a hand trajectory toward a displaced
target during reaching occurs without conscious control (Goodale et
al. 1986) and is often referred to as the “automatic pilot.” These
automatic corrections during target movements occur at a short
latency (within 150 ms) (Day and Lyon 2000; Kadota and Gomi
2010) compared to the typical reaction times observed for reaching
movements when the movement is initiated from rest (~200-250 ms).
The involvement of the PPC in online corrections has been further
corroborated by a study done by Pisella and colleagues (Pisella et
al. 2000). They tested a stroke patient with bilateral lesions to the
parietal cortex. The patient had optic ataxia and accurately pointed
to stationary targets, but if the target was displaced at movement
onset, her ability to correct an ongoing movement was severely
impaired. This study suggests that the parietal cortex plays a central
role for online feedback control to regulate reaching movements in
response to sudden and unexpected changes in target location.
PROBLEM 30.4
Rapid online control is quantified in many studies that focus on
developmental coordination disorders (DCDs). Many children with
DCDs can make simple reaching movements with relative ease,
but they exhibit impairments in the ability to make rapid online
corrections. Even healthy adults sometimes experience difficulties
executing movements that involve rapid online control. Why would
movements that involve rapid online control be “harder” to learn
and execute than simple reaching movements?
30.10 Eye and Hand Coordination
While Intercepting Moving
Targets
Intercepting and catching a moving projectile are fundamental motor
skills that require precise timing and coordination between the eye
and the hand. The moving projectile is tracked with smooth-pursuit
eye movements, and then the limb is directed toward the point of
impending collision. The main difference between reaching and
interception movements is that in contrast to saccadic eye
movements that are made during reaching, interception movements
primarily engage smooth-pursuit eye movements. Tracking moving
objects enables continuous high-acuity vision, but smooth-pursuit
eye movements are limited to a maximum speed of 80º/s to 100º/s
(Meyer et al., 1985). Thus, for stimuli that move faster than that
speed, the oculomotor system launches catch-up saccades to
compensate for position and velocity errors.
During catching or interception action in sports, players track the
moving object with a combination of eye and head movements until
the hand is aligned with the moving object at the point of interception
(reviewed in Fooken et al. 2021). Often there is a short phase of
smooth tracking observed right before the ball is intercepted,
suggesting that keeping the eye on the object is important for
accurate prediction of object motion (Hayhoe et al. 2012; Land and
McLeod 2000). If the eye movements are constrained (e.g., when
participants are asked to fixate at a point instead of freely pursuing
the moving projectile), their performance is degraded. The most
likely explanation for this is that object speed during visual fixation is
overestimated, leading to interception movements ahead of the
moving object.
It is currently also believed that the nervous system uses
efference copies of oculomotor commands associated with smooth-
pursuit eye movements to make predictions about object motion and
coordinating eye and hand movements for interception. The reliance
on both retinal projections of the object image and the extraretinal
oculomotor signals might be more important for smooth eye–hand
coordination when the object motion is more unpredictable (e.g.,
intercepting a curving soccer kick or catching a fly) (see figure 30.6).
The studies on manual interception with human participants and
primates have highlighted the important role played by the MT/MST
and FEF areas and the parietal cortex in the precise spatiotemporal
coordination of both eye and limb movements (Merchant et al. 2004,
2006; Zago et al. 2009). Lesions in the parietal cortex, FEF, and
MT/MST cause deficits in smooth-pursuit eye movements (Dursteler
and Wurtz 1988; Heide et al. 1996) and consequently in eye–hand
coordination.
MT and MST project to the premotor cortex, especially the FEF
and the dorsal premotor cortex. These projections likely are involved
in continuous transformation of motion signals from a projectile into
motor commands to move the limb to the right location in space to
make contact with the projectile. Simultaneously, the nervous system
also prepares a postural response to absorb the projectile’s
momentum.
Evidence for the transformation of motion signals to continuous
modification of limb posture has come from a study by Selen and
colleagues (Selen et al. 2012). In their experiment, participants
viewed a dynamic random dot display and indicated their decision
about the direction of the dot movement by moving a handle to one
of two targets. The random dot (figure 30.1) display engages neural
area MT/MST, which is also involved in tracking moving objects. The
experimenters perturbed the arm at random times during decision
formation. The long-latency reflex gains of the M3 component (also
see chapter 18) were modulated by the strength and duration of
motion of the random dots, reflecting the accumulated evidence in
support of the evolving decision. Their results support the existence
of a sensorimotor process that continuously transforms motion
signals for limb motor control. Furthermore, this suggests that the
motor system might continuously set the referent joint configurations
for catching movements based on the sensory output from area
MT/MST.
Figure 30.6 It has been hypothesized that oculomotor extraretinal signals are
used as a feedforward input to form predictions and guide eye and hand
movements. Humans rely predominantly on retinal signals when the object moves
predictably. Unpredictable motion of a visual stimulus causes stronger reliance on
extraretinal motor commands associated with eye movements.
Part a adapted from J. Fooken, P. Kreyenmeier, and M. Spering, “The Role of Eye
Movements in Manual Interception: A Mini-Review,” Vision Research, 183 (2021): 81-90.
CHAPTER 30 IN A NUTSHELL
The visual and oculomotor systems must
work in synchrony for normal visual
perception. The visual system has
dedicated streams and specialized
neurons for visuomotor and perceptual
functions. The dorsal and ventral
visual streams facilitate visuomotor
and perceptual function, respectively.
The oculomotor system governs eye
movement kinematics and directs the
fovea to static target locations or to
follow moving objects. For reaching
movements, target foveation allows the
visual system to extract information
about finer details of the object and
localize targets in space with high
spatial resolution. The parietal
cortex, specifically the dorsal visual
stream, plays a critical role in
spatial planning of movements, as well
as online correction of movement
trajectories. For interception
movements, target foveation is
important for accurate prediction of
object motion.
Problems for Part VI
Self-Test Problems
1. Suggest examples of kinesthetic illusions induced by distorted
or inappropriate efferent copy signals.
2. You participate in an experiment where a researcher draws
one of four different weights of 1, 2, and 4 lb from a box and
places it on your left hand. She then takes another weight
drawn from a different box that could weigh anywhere from
0.5 to 4 lb in steps of 0.1 lb and places it on your right hand.
She then asks you to judge if the weight on the right hand is
heavier or lighter than the weight on the left hand and by how
much (approximately). She keeps drawing the weights till you
report that both weights are approximately the same. The
same process is repeated multiple times with different weights
on the left hand. She then compiles a score of how off you
were (in pounds) for each of the weights placed on the left
hand. Are you likely to be more accurate for the 1 lb weight or
the 4 lb weight? Why?
3. Tasks such as throwing a basketball free throw require
dynamic multisensory integration. First, the visual system
must localize the hoop with respect to the body. Then the
proprioceptive system judges where the joints are with respect
to the body and the force applied by the limb. If a person has
suffered damage to the proprioceptive system at the cervical
level, how would they be able to perform a free throw?
4. The reference frames in which sensory stimuli from different
modalities are encoded are different from each other and from
those of motor effectors. For example, visual information is
encoded in a retinotopic reference frame, but proprioceptive
information is encoded in a joint-based reference frame.
Neurons in the parietal cortex transform sensory signals into a
common reference frame to guide movements. If neurons that
transform visual and vestibular signals into a common
reference frame were to be inhibited selectively with a drug,
what kind of impact would we expect to see on motor
performance during the execution of complex motor skills that
require posture stabilization and fine motor control?
5. The sensation of illusory self-motion is called vection. We
sometimes experience an instance of this sensation when a
car parked right next to ours suddenly moves. In contrast, this
sensation is almost never experienced when we ride a bicycle
and the bicycle stopped next to ours suddenly moves. What
might explain this discrepancy?
6. How would the absence of canonical neurons affect reach-to-
grasp actions? What would we expect to see with the
movement of the digits?
Fatigue
PROBLEM 31.1
Why does the slowing of conduction lead to a decrease in the
amplitude and an increase in the duration of a potential recorded
from the muscle surface?
PROBLEM 31.2
Prolonged fatiguing contraction is commonly accompanied by
tremor at a frequency of 4 to 6 Hz. Can you suggest a mechanism
for this phenomenon?
PROBLEM 31.3
A decrease in apparent joint stiffness has been reported during
fatigue. Suggest a mechanism for this finding.
Figure 31.4 The two lines illustrate two relations between the level of muscle
activity (normalized EMG) and isometric muscle torque (also normalized). Under
fatigue (dashed line), the muscle shows much higher activation levels for the same
torque levels as compared to the prefatigue state (solid line).
PROBLEM 31.4
Why does the EMG spectrum shift to lower frequencies in
conditions of excessive motor unit synchronization? Can you
suggest conditions under which a spectral shift toward high
frequencies might be expected?
CHAPTER 31 IN A NUTSHELL
Fatigue is a complex phenomenon that
receives contributions from peripheral
and central neural and psychological
factors. Fatigued muscles show a
slower conduction velocity of action
potentials and prolonged twitch
contractions, primarily due to the
prolongation of the relaxation phase.
Fatigue-related reflex changes, in
particular the suppression of the H-
reflex, are likely to be mediated by
changes in the activity of small
receptors of groups III and IV,
leading to increased presynaptic
inhibition. Medium-latency
preprogrammed reactions show a modest
decline, but long-latency reactions
can increase under fatigue, suggesting
their transcortical nature. During
voluntary contractions, there is an
increase in the EMG level required to
keep force unchanged. Motor units show
an increase in the synchronization of
their firing patterns with fatigue.
Prolonged, fatiguing contractions in
humans are accompanied by a gradual
increase in the activity of cortical
neurons. There are adaptive changes to
fatigue at different levels, including
adaptive changes in motor synergies.
An unusual sense of fatigue (e.g.,
chronic fatigue and fatigue in
multiple sclerosis) is likely to be
mostly of a central nature.
Chapter 32
Effects of Aging
PROBLEM 32.1
Increased muscle cocontraction apparently does not contribute to
the net torque acting at a joint and may be seen as wasteful. Can
you suggest benefits from using increased muscle cocontraction
levels?
PROBLEM 32.2
Suggest reasons why, in elderly persons, motor unit size and
fatigability stop showing the typical relationship in which larger
motor units are more fatigable.
The lack of smaller motor units may be blamed for the poor
control of low forces, high force variability, and poor smoothness of
movements because the recruitment and derecruitment of larger
motor units leads to larger changes in the total muscle force. Figure
32.4 illustrates the high variability in the average frequency of action
potentials in motor units of the first dorsal interosseous muscle when
the subjects were producing 5% of their maximal voluntary force;
note also the larger fluctuations in the force level in the elderly
(Laidlaw et al. 2000). Despite the described changes in motor unit
properties, changes in fatigability with age are ambiguous when
quantified at the level of motor performance. There are reports of no
changes, an increase, and a decrease in fatigability in the elderly
(Chan et al. 2000; Bilodeau et al. 2001; Allman and Rice 2002).
PROBLEM 32.3
Suggest reasons for the different effects of aging on proximal and
distal muscles.
PROBLEM 32.4
How can the conduction speed of action potentials change with
age? Suggest a mechanism. How can the speed of reaction of a
sensory ending change with age?
PROBLEM 32.5
Reduced APAs are seen in young persons when they stand on a
board with a reduced support area. Suggest a common
explanation for this observation and the reduced APAs in the
elderly.
CHAPTER 32 IN A NUTSHELL
Aging leads to losses in muscle mass,
strength, and the number of α-
motoneurons. Processes of denervation,
sprouting, and reinnervation lead to a
smaller number of motor units that
are, on average, slower and larger
than in younger persons. There is an
impairment of the sensory functions
with age. Reactions to sensory
stimuli, from monosynaptic reflexes to
voluntary reactions, are typically
slower and smaller in magnitude.
Control of vertical posture and gait
declines with age, leading to more
frequent falls. This decline is
reflected in the larger postural sway,
smaller and delayed anticipatory
postural adjustments, and smaller
preprogrammed reactions to
perturbations. Movement patterns in
elderly persons are characterized by
slowness and excessive cocontraction
of agonist–antagonist muscle pairs.
Movements are less smooth and more
variable. Prehensile tasks are
associated with excessive grip forces
and weaker multidigit synergies
stabilizing the hand action. Some of
the features of movements in elderly
persons may be viewed as reflecting
adaptive strategies by the central
nervous system. Training can increase
the strength and endurance of elderly
persons as well as accuracy in force-
producing tasks and the associated
synergies.
Chapter 33
PROBLEM 33.1
What would you expect from incompletely myelinated fibers, an
inability to conduct action potentials or slow conduction of action
potentials?
PROBLEM 33.2
How can conduction speed change with age? Suggest a
mechanism. How can the speed of reaction of a sensory ending
change with age?
PROBLEM 33.3
Suggest other examples in which motor variability may be viewed
as exploratory and functionally useful.
PROBLEM 33.4
Can the increased motor variability in young adults with Down
syndrome be viewed as exploratory and functionally useful? Why?
Figure 33.6 If a person tries to produce a constant force level by pressing with
four fingers, two strategies of finger force covariation are possible. (a) A “fork
strategy” when finger forces covary positively. (b) A flexible strategy when the
finger forces covary predominantly negatively. Deviations of the total force are
smaller in the right graph.
PROBLEM 33.5
Suggest a benefit of the “fork strategy” as compared to more
typical patterns seen in persons without Down syndrome.
Figure 33.7 Prior to practice, persons with Down syndrome typically showed a
“fork strategy” (negative values of the synergy index). After 2 d of practice, the fork
strategy is replaced by a more typical negative covariation of finger forces (positive
values of the synergy index) that help stabilize the total force. Persons who used
variable practice (thin solid line) showed greater improvement than those who
used blocked practice (thin dashed line).
Reprinted by permissions from M.L. Latash, N. Kang, and D. Patterson, “Finger
Coordination in Persons with Down Syndrome: Atypical Patterns of Coordination and the
Effects of Practice,” Experimental Brain Research 146 (2002): 345-355. © 2002 Springer.
33.7 Autism
Autism was first described in 1943 by Dr. Leo Kanner. It is estimated
that babies with autism are born at a rate of 1 out of 250 live births.
One and a half million Americans are estimated to have autism.
Autism is a spectrum disorder. This means that it is characterized
by a spectrum of signs that may not necessarily be linked to a single
neurophysiological or other mechanism. Persons with autism
commonly show resistance to change, distress for unclear reasons,
difficulty in mixing with others, lack of responsiveness to words, and
difficulty in verbal expression, sometimes reflected in repeating
words. Their motor behavior may be characterized by such dissimilar
features as physical overactivity or extreme underactivity. They may
show stereotypical, repeated movements and sustained odd play.
Their gross and fine motor skills may show very uneven
development.
Asperger’s disorder is a neurobiological disorder named after
the Austrian physician Hans Asperger, who in 1944 described a
pattern of behaviors in young boys who had normal intelligence and
language development, but who also exhibited autistic-like behaviors
and marked deficiencies in social and communication skills. Persons
with Asperger’s disorder are similar to persons with autism in their
inability to regulate social interaction by using nonverbal behaviors
such as body posture and gestures, eye contact and facial
expression. They show a preoccupation with stereotypical behaviors,
routines, and rituals and have problems making social contacts and
developing peer relationships. Unlike persons with autism, those with
Asperger’s disorder show no significant delay in language, in
cognitive development, or in the development of self-help skills,
adaptive behaviors (other than in social interaction), and curiosity
about the environment in childhood.
Babies with both autism and Asperger’s disorder often show
abnormal features of motor development, including lack of protective
reflexes when falling and delayed development of walking
(Teitelbaum et al. 1998, 2004). There is increasing evidence that
both disorders are associated with cerebellar abnormalities, more
pronounced in autism, including reduced cerebellar gray matter and
neuronal loss in the cerebellar nuclei (Courchesne 1997; Palmen et
al. 2004; McAlonan et al. 2005). As discussed earlier in the section
on Down syndrome, this may be related to problems in assembling
multi-element synergies. Here, under synergies, we imply not only
acts of motor coordination but a more general notion of elements
that are brought to work together. The role of elements can be
played by words—then, one observes difficulties with such things as
verbal expression and agrammatisms. The role of elements can also
be played by persons—then, problems with communication and
interacting with others may be expected.
Autism is characterized by a specific feature that suggests that
persons with autism have difficulties perceiving others as “other
beings just like me” who can have their own views and ideas. This
feature is revealed in the false belief test (Muris et al. 1998; Pilowsky
et al. 2000; Grant et al. 2001). In this test, a person is shown a
sequence of pictures with two kids and a cake. One of the kids puts
the cake into the refrigerator to make sure that it does not get
spoiled. Then, this kid leaves the room. The other kid thinks that the
cake is to be eaten in a few hours and it makes sense to warm it up.
So, the second kid takes the cake from the refrigerator and puts it
into a cabinet. Then, the second kid leaves the room, and, after
some time, the first kid enters. The question is: “Where will the first
kid look for the cake?” Giving the right answer (“In the refrigerator”)
implies that the person taking the test understands that these
personalities may have wrong views of the world (false beliefs)
based on outdated information. Children with autism have been
shown to fail this test at ages when typically developing children
pass it.
CHAPTER 33 IN A NUTSHELL
Human babies are born with an
incompletely developed central nervous
system. The process of myelinization
continues over the first 6 mo of life.
Motor development typically follows a
progression along a set of milestones.
The idea of a fixed timetable has been
challenged by proponents of a view
that motor development results from
changing interactions between
constraints imposed by the organism
and by the environment. Persons with
Down syndrome have an extra chromosome
in the 21st pair. This leads to both
motor and nonmotor consequences. Motor
consequences can be summarized as
“clumsiness,” which gets contributions
from longer reaction times, slower
movements, and higher variability.
Persons with Down syndrome show
preference for cocontraction patterns
of muscle activation in a variety of
motor tasks. Practice leads to
dramatic changes in motor patterns of
these persons. They become faster in
simple single-joint motor tests. Their
multifinger synergies show shifts
toward patterns observed in persons
without Down syndrome. Autism and
Asperger’s disorder are characterized
by features that may be partly
summarized as a poor ability to form
synergies; this impairment may be
causally related to cerebellar changes
in these states. Development
coordination disorder is a delay in
motor coordination that is also
potentially linked to a cerebellar
dysfunction.
Chapter 34
Motor Learning
PROBLEM 34.1
When you look at a great painting, you feel emotions reflecting the
ideas of the artist who died a long time ago. Can we view the
painting as a memory of those ideas?
PROBLEM 34.3
Pavlov built “towers of silence” where his dogs were deprived of
any stimuli except those used for conditioned reflexes. Predict the
behavior of the dogs based on the theory of conditioned reflexes
and Bernstein’s idea that all actions are initiated from within the
central nervous system.
PROBLEM 34.4
What physiological mechanisms could be involved in changes of
the amplitude of the monosynaptic responses in Wolpaw’s
experiments?
34.6 Short-Term and Long-Term
Memory
Two overlapping categories of memory are identified in humans. The
first one is called short-term memory. Its effectiveness usually lasts
a few minutes or hours. Special, memory consolidation processes
are invoked to explain the process of conversion of short-term
memories into long-term memories, which can last for years, up to
a lifetime (figure 34.2). There are many clinical cases, mainly
different types of brain trauma, leading to a memory loss or
deficiency. In particular, some patients lack the ability to consolidate
short-term memory. They demonstrate normal ability to maintain
short-term memories, but lose them when their attention is
distracted. There are also cases when a trauma leads to a loss of
memories related to events that happened for some time period prior
to the trauma (retrograde amnesia) or to a loss of memories related
to events that happened for some time period immediately following
the trauma (anterograde amnesia).
Figure 34.2 Processing of a sensory stimulus may lead to the creation of a short-
term memory trace in parallel with the production of an effector (motor) output.
Short-term memory can be consolidated into a long-term memory. Both short-term
and long-term memories can participate in processing future incoming sensory
stimuli and forming an effector response.
PROBLEM 34.5
An alternative explanation of returning to straight trajectories with
practice is using high muscle coactivation, which reduces the
kinematic effects of the force field. Present arguments for or
against this explanation.
Within the equilibrium-point hypothesis (chapters 20-21), changes
in neural commands with practice are described in terms of control
trajectories, such as the time changes in the tonic stretch reflex
threshold, λ(t) for a single muscle or in the referent coordinates of
the involved effector, RC(t). Currently, rules that result in adjustments
of control trajectories during practice are unknown. An elegant
solution has been suggested (Gribble and Ostry 2000) that takes
advantage of the fact that RC is expressed in actual kinematic
variables. If one wants to move an effector along a certain trajectory
and observes a deviation from the desired path (figure 34.4a), the
RC trajectory (dashed lines in figure 34.4) is adjusted in the next trial
following a simple rule (figure 34.4): The difference between the
actual and desired trajectories (34.4b) is added to the previous RC(t)
function (figure 34.4c). Using this simple rule has been shown to
lead to a very quick adaptation to novel conditions taking 2 to 3 trials
without any computations related to the complex transformations of
neural and mechanical variables involved in producing the task
mechanics.
Figure 34.4 (a) Referent coordinate trajectory (RC) is similar to the actual
trajectory (TR) in shape in the absence of force field. (b) When an unusual force
field is on (central image, see figure 34.3), TR becomes curved. (c) The deviation
from the desired trajectory is perceived by the subject and used to correct RC.
This results in a straight TR after just one trial. (d) After the force field has been
turned off, the trajectories show an aftereffect: They become curved in the
opposite direction.
Figure 34.5 Samples of writing by two subjects, who used (a) separate-letter
writing and (b) fused-cursive writing. Both subjects improved their timing index (TI)
and error index (EI) with practice. Note the improved handwriting by the
nondominant hand that had not practiced and also in the writing of a new,
nonpracticed phrase.
Reprinted by permission from M.L. Latash, “Mirror Writing: Learning, Transfer, and
Implications for Internal Inverse Models,” Journal of Motor Behavior 31, (1999): 107-112. ©
1999 HELDREF Publications. https://www.tandfonline.com/.
CHAPTER 34 IN A NUTSHELL
Practice leads to both transient,
easily reversible, changes in
performance (adaptation) and longer-
lasting changes (learning) based on
neuronal memory. The effects of memory
can be seen at different levels,
including habituation of reflexes,
classical or operant conditioning, and
skill acquisition. Pavlov’s theory
considers animal behavior to be a
combination of inborn and conditioned
reflexes. Bernstein’s physiology of
initiative emphasizes the principle of
activity and views motor skill
acquisition as an active search
process. Neuronal mechanisms of memory
are unknown. Long-term potentiation
and depression are frequently viewed
as the neuronal mechanism of memory
implying synapses as the storage site.
The classic theory of three stages in
learning a motor skill involves
freezing and releasing degrees of
freedom followed by optimized use of
external forces. Even short practice
can lead to plastic changes in the
brain motor and sensory maps. Within
the concept of abundance, there are
two stages: (1) developing and
strengthening relevant synergies and
(2) selecting families of solutions
that satisfy additional explicit or
self-imposed requirements.
Problems for Part VII
Self-Test Problems
1. A person practices a fast reaching movement to a target while
sitting in the center of a rotating centrifuge. Vision is removed
as soon as movement is initiated. What differences in
movement trajectories would you expect? What will happen
with the trajectories after prolonged practice? What will
happen after the practice when the same movements are
performed in a motionless centrifuge?
2. A person tries to keep maximal ankle plantarflexion force
against a stop for about 2 min. The force drops by about 50%.
A brief electrical stimulus is applied to the muscle nerve (n.
tibialis) at the end of this period on the background of
continuing flexion effort. The stimulus induces an M-response
and an H-reflex. What can you say about motor units that
contributed to the M- and H-responses? Would you expect the
responses (M-, H-, and force twitch) to be higher or lower than
those seen in response to the same stimulus in the same
muscle tested in nonfatigued, relaxed conditions? Why?
3. A toddler stands in a bus, and the bus suddenly starts to
move. What mechanisms would help the toddler not to fall
down? How do they differ from similar mechanisms in a
grown-up?
4. A single TMS stimulus applied over the primary motor cortex
of a person who generates a low level of hand force into
flexion produces a quick response in the wrist flexor (flexor
carpi radialis), followed by a silent period in the EMG. The
same person, after a fatiguing exercise, matches the
background hand force, and a TMS stimulus with the same
parameters is applied over the same spot. What changes
would you expect to see in characteristics of the muscle
response? Explain your answer.
5. A standing healthy elderly person is unexpectedly pushed
from behind by another person (not too strongly!). Describe all
the differences in posture-stabilizing responses from those
seen in a young person.
6. Older persons frequently show increased tremor across a
variety of actions. Describe all possible mechanisms that
could bring about the increased tremor, assuming no
diagnosed neural pathology exists.
Motor Disorders
Chapter 35
PROBLEM 35.1
Suggest a physiological mechanism for myotonia.
PROBLEM 35.2
How is it possible that monosynaptic reflexes are unchanged in
stiff-man syndrome if the problem is in reduced presynaptic
inhibition?
35.5 Mononeuropathies
Peripheral neuropathies are classified into mononeuropathies (when
a single nerve is affected), multiple mononeuropathies (when a
number of nerves are affected within a particular body area), and
polyneuropathies (these are systemic disorders affecting nerves
throughout the body). There are commonalities across these
disorders that include the slowing or complete interruption of
conduction along peripheral neural pathways as the main cause of
signs and symptoms.
Mononeuropathies typically are associated with slowed
conduction in a single nerve; the amplitude of motor or sensory
potentials is reduced. If a motor nerve is affected, signs of
denervation may be seen in the target muscle. Probably the best
known mononeuropathy is the carpal tunnel syndrome, which
represents entrapment of the median nerve in the carpal tunnel at
the wrist level. The carpal tunnel contains both the tendons of
extrinsic flexor muscles and the median nerve. There may be
different reasons causing increased pressure on the median nerve in
the carpal tunnel. These include thickening of the protective sheaths
that surround each of the tendons, resulting from repetitive overuse
of these muscles (for example, during work with power tools); the
tendon sheaths may also become swollen. There is also a genetic
predisposition for carpal tunnel syndrome that is related to the size of
the carpal tunnel.
Early symptoms of carpal tunnel syndrome commonly involve
painful tingling in one or both hands, particularly during the night.
These are felt in digits innervated by the median nerve up to the
medial half of the ring finger. As symptoms increase, tingling may be
felt during the day, commonly in the thumb, index, middle, and ring
fingers. There is also an impairment of the motor function leading to
poor ability to squeeze objects. In advanced cases, the thenar
muscle at the base of the thumb shows atrophy and strength loss.
Persons with this condition may start to appear clumsy and have
problems with simple tasks that involve finger coordination, such as
tying shoelaces or manipulating small objects. Figure 35.2 illustrates
schematically the delayed conduction of action potentials over the
carpal tunnel (cf. Nobuta et al. 2005); delays over 10 ms are
considered severe.
PROBLEM 35.4
Imagine that a young adult person with diabetes stands on a
platform, and the platform suddenly starts to move at a moderate
speed. Would you expect to see an ankle strategy or a hip
strategy in the early reaction to the perturbation?
35.7 Polyneuropathies
Polyneuropathies may be associated with a demyelinating process
or with toxic effects on peripheral axons. Demyelination may result
from a chronic inflammatory process or from an autoimmune
process, in which the body’s immune system attacks parts of the
peripheral nervous system. This latter condition is called Guillain-
Barré syndrome (Hughes and Cornblath 2005). The loss of myelin
may lead to blocked conduction along affected axons, which results
in a reduced ability to recruit motor units in muscles innervated by
the affected nerves. Monosynaptic reflexes such as knee jerks are
usually lost.
The Guillain-Barré syndrome has an annual incidence of 0.6 to
2.4 cases per 100,000. It is more frequent in males. About two-thirds
of patients have a history of gastrointestinal or respiratory infection 1
to 3 wk prior to the onset of the first clinical signs. The disease peaks
in the age ranges of 15 to 35 yrs and 50 to 75 yrs.
The first symptoms of this disorder include varying degrees of
weakness or tingling sensations in the legs. The onset of symptoms
is sudden and unexpected. In many instances, the weakness and
abnormal sensations spread to the arms and upper body. These
symptoms can increase in intensity until the patient is almost totally
paralyzed. Most people reach the stage of greatest weakness within
the first 2 wk after symptoms appear. A majority of patients, however,
recover from even the most severe cases of Guillain-Barré
syndrome, although some continue to have some degree of
weakness. The mortality rate ranges from 5% to 10%.
Guillain-Barré syndrome shares features of multiple sclerosis
(demyelination of neural pathways within the central nervous
system) and myasthenia gravis (an autoimmune disease). There are
several ways to treat the complications of the disease. Currently,
plasmapheresis and high-dose immunoglobulin therapy are used
most frequently.
35.8 Radiculopathies
The term radiculopathy refers to a group of disorders that originate
from a mechanical or inflammatory damage to a spinal root or a
group of spinal roots that have progressed enough to cause the
development of neurological symptoms in the areas supplied by the
affected root. These disorders can lead to sensory (tingling and pain)
or motor problems (weakness), depending on whether the dorsal or
ventral roots are affected. The separation of sensory and motor
symptoms makes radiculopathies different from peripheral
neuropathies that typically affect both sensory and motor neural
fibers that run together in common nerves.
One of the most frequent causes for radiculopathies is spinal disk
herniation. In this condition, the outer layer (annulus) of the disk
cracks, and the gel-like center (nucleus) breaks through. This causes
the disk to protrude, putting pressure on the nerve that exits the
spinal column at that point. Spinal disks can also suffer from age-
related changes, leading to a degenerative disk disease. As humans
age, the water content in the disks diminishes, causing the disk to
shrink. Without sufficient cushioning, the vertebrae may begin to
press against each other, pinching the nerve, or they can form bony
spurs. Finally, radiculopathies may result from spinal stenosis, a
condition in which the space in the center of the vertebrae narrows
and squeezes the spinal column and nerve roots.
Most common radiculopathies occur at the cervical and
lumbosacral levels. For example, in the United States, lumbosacral
radiculopathies occur in about 2% of the population. Of these cases,
10% to 25% develop persisting symptoms.
CHAPTER 35 IN A NUTSHELL
Peripheral motor disorders may
originate from a dysfunction in the
muscles, neuromuscular synapses,
peripheral axons, and bodies of α-
motoneurons. Muscular dystrophies are
a group of genetic diseases
characterized by progressive weakness
and degeneration of the skeletal
muscles. Duchenne and Becker
dystrophies have been linked to
defects in the same dystrophin gene.
Myotonic dystrophy is characterized by
myotonia, a prolonged episode of
muscle activity after its voluntary
contraction. Myasthenia gravis is a
disorder of transmission at the
neuromuscular synapse. It is likely to
be a consequence of an autoimmune
process. Peripheral neuropathies lead
to both sensory and motor consequences
in areas innervated by the affected
nerves. Guillain-Barré syndrome leads
to demyelination of peripheral axons
that may result from an autoimmune
process. Immunosuppressive treatment
and plasmapheresis are common
therapeutic strategies for both
myasthenia gravis and Guillain-Barré
syndrome. Advanced diabetes leads to
peripheral neuropathies that may be
associated with impaired coordination
and balance.
Chapter 36
PROBLEM 36.1
Can you suggest why there is no spasticity in cases of lumbar
spinal cord injuries?
PROBLEM 36.2
Spinal cord injury at what level is likely to lead to the destruction of
the neuronal apparatus of locomotor central pattern generators?
PROBLEM 36.3
Suggest a physiological mechanism underlying increased muscle
tone. What about decreased muscle tone?
Positive signs
Flexor/extensor spasms
Clonus
Clasp-knife phenomenon
“Babinski reflex”
Exaggerated cutaneous reflexes
Autonomic hyperreflexia
Tone abnormalities
Negative signs
Weakness (paresis or plegia)
Loss of dexterity
Fatigability
Contractures
Although it has been commonly assumed that spasticity is
associated with a deficit in spinal inhibitory mechanisms, including
both postsynaptic and presynaptic inhibition, there is no consensus
about what causes these deficits in the first place. To say that they
are due to the disruption of the normal functioning of certain
descending systems does not help much since these systems are
not well defined, and their role in voluntary motor control is unclear.
The relations between spasticity and muscle reflexes are also not as
unambiguous as implied by the definitions presented. Let us
consider typical changes in muscle reflexes associated with
spasticity.
• Spasticity can be associated with exaggerated, unchanged, and
even absent monosynaptic reflexes, including the H-reflex, although
an increase in monosynaptic reflexes is more typical.
• A common correlate of spasticity is spasm-like bursts of activity
in leg muscles in response to tactile stimulation of the sole of the foot
(figure 36.1). This response is sometimes imprecisely called the
Babinski reflex or a defensive reaction. It is quite variable across
patients and may involve bursts of activity in all major flexor muscles,
a sustained contraction of the flexor muscles, either with or without a
comparable activation of extensor muscles. However, this reflex can
be absent in certain patients with spasticity.
• Another typical sign of spasticity is clonus (figure 36.2), which
represents a series of alternating bursts of activity in the flexor and
extensor muscles of a joint at a frequency of about 6 to 8 Hz in
response to a single quick movement of the joint performed by the
experimenter (passively) or by the patient (if there is enough
voluntary motor control left). Clonus may last for only about a
second, or it may continue for tens of seconds or even minutes until
it is stopped mechanically—for example, by clamping the joint and
preventing it from moving. Clonus is likely to represent an auto-
oscillation in the hyperexcitable stretch reflex loop (Iansek 1984;
Latash et al. 1989; Hidler and Rymer 1999): When a muscle is
stretched, a monosynaptic stretch reflex leads to its phasic
contraction, leading to a reversal of the joint’s movement direction.
As a result, the antagonist muscle is stretched and demonstrates a
monosynaptic stretch reflex. And so on. Remember that joint
movements in healthy persons do not normally induce monosynaptic
reflexes. There is an alternative hypothesis claiming that clonus is a
result of the functioning of a central generator within the spinal cord
(Dimitrijevic et al. 1980; Beres-Jones et al. 2003).
Figure 36.1 An example of a spasm in leg muscles induced by tactile stimulation
of the sole of a foot in a spastic patient. This reaction is sometimes imprecisely
called the Babinski response. TA = tibialis anterior; SOL = soleus; QUAD =
quadriceps (rectus femoris); HAM = hamstrings.
Reprinted by permission from M.L. Latash, R.D. Penn, D.M. Corcos, and G.L. Gottlieb,
“Short-Term Effects of Intrathecal Baclofen in Spasticity,” Experimental Neurology, 103
(1989): 167. ©1989, with permission from Elsevier.
Figure 36.3 (a) Changes in an H-reflex in the soleus muscle induced by vibration
of the muscle tendon in a healthy person. Note the suppression of the H-reflex. (b)
Changes in an H-reflex in a person with spasticity; no suppression of the H-reflex.
PROBLEM 36.4
Based on these two conflicting hypotheses, predict changes in the
clonus frequency in conditions when the limb is loaded inertially
and when the strength of contractions is reduced by a drug.
Two clinical scales have been successfully used for quantitative
assessment of spasticity. The most frequently used is the Ashworth
scale, which reflects the degree of muscle resistance to passive limb
movements (table 36.2). The other is the spasm scale, which reflects
the frequency of spasms, their duration, and their general or local
character (table 36.3). Both scales are subjective and reflect a
physician’s general impression of the patient’s state, which is likely to
be important from the clinical view but is not very helpful for
understanding the mechanisms of the disorder.
Figure 36.4 (a) In a healthy person, the brain can shift the control variable λ
beyond the anatomical range of muscle length values, from λMIN to λMAX. These
shifts can lead to large muscle force or muscle relaxation at any muscle length
(LMIN to LMAX). (b) In a person with spasticity, the controller loses its ability to shift
λ over the whole range, only from λ– to λ+. As a result, the muscle may not be able
to relax at high length values (shown as “spasms”) and to produce voluntary
activation at low length values (shown as “paralysis”).
To account for spasticity, it has been assumed that the controller
loses the ability to shift λ over the whole range. Moreover, it has
been assumed that large λ shifts can be produced by sensory
feedback signals. Figure 36.4b illustrates that a contracted range of
voluntary λ shifts may be associated with unintentional muscle
activation when muscle length exceeds the range of λ shifts. The
same scheme suggests that no voluntary movement can be possible
when muscle length is shorter than the smallest value of λ within the
contracted range of its changes, leading to paralysis. A number of
recent studies linked the decreased range of λ changes in patients
with spasticity with their clinical state in support of the scheme in
figure 36.4 (Turpin et al. 2017; Subramaniam et al. 2018).
This hypothesis on the origins of spasticity leads to a few
predictions. Most importantly, it unites exaggerated muscle reflexes
and poor voluntary control into a single scheme. As such, it
eliminates the distinction between positive and negative signs of
spasticity and predicts that changes in one group of the signs may
be expected to be accompanied by changes in the other group. This
prediction received support in a number of studies that quantified the
effects of treatment on both muscle reflexes and voluntary
movements (Corcos et al. 1986; Latash et al. 1990; Latash and Penn
1996).
PROBLEM 36.5
What conclusion can be drawn regarding the site of baclofen
action based on the observations of similar suppression of
monosynaptic reflexes on both sides of the body of a patient with
a hemisyndrome, effective suppression of spastic signs, and no
apparent changes in control of the intact side of the body?
CHAPTER 36 IN A NUTSHELL
Spinal cord injury leads to sensory-
motor consequences reflecting the
disruption of transmission along
ascending and descending neural
pathways and the destruction of the
spinal neuronal apparatus. Spasticity
is a typical consequence of spinal
cord injury; it is characterized by a
partial or complete loss of voluntary
control over muscles, partial or
complete loss of sensation,
uncontrolled spasms and increased
reflexes, possibility of chronic pain,
and disruption of functions of
internal body organs below the level
of trauma. Treatment of spasticity
with intrathecal drug delivery, in
particular by intrathecal baclofen,
has been most successful. Suppression
of spasms and reflexes can be
accompanied by an unmasking of more
normal voluntary movements while
voluntary control of unaffected
muscles does not change. Adaptive
changes to the original trauma are
likely to play an important role in
the selective action of drugs.
Spasticity has been interpreted as a
problem in changing the threshold of a
stretch reflex over its whole normal
range. This interpretation links
typical spastic signs to problems with
voluntary muscle activation. It also
suggests a definition for muscle tone.
Chapter 37
Figure 37.4 Four major signs of Parkinson’s disease include tremor, rigidity,
bradykinesia, and deficit in postural reactions. The latter deficit has two
components: a deficit in APAs and a poorly controlled increase in later,
preprogrammed reactions.
PROBLEM 37.1
There are obvious similarities between the symptoms of spasticity
and of Parkinson’s disease. How would you differentiate between
multiple sclerosis and Parkinson’s disease?
PROBLEM 37.2
Why are muscle forces and EMGs not adequate reflections of
central commands during voluntary movements by healthy
persons?
PROBLEM 37.3
Suggest an experiment that would test the hypothesis that tremor
in Parkinson’s disease is an oscillation in a long-latency reflex
loop.
Figure 37.6 The index of the multifinger synergy (ΔVZ) stabilizing total force
during the four-finger force production task by the right hand. The task involved
steady-state force production followed by a quick pulse into a target (starting at t0).
Note the higher steady-state value of the synergy index in the healthy control
subjects (dashed line, darker error shade) compared to patients with Parkinson’s
disease (PD, solid line, light error share). Note also the large ASA prior to the force
pulse initiation in the control group, but not in the PD patients.
Adapted by permission from J. Park, Y.-H. Wu, M.M. Lewis, X. Huang, and M.L. Latash,
“Changes in Multi-Finger Interaction and Coordination in Parkinson’s Disease,” Journal of
Neurophysiology 108 (2012): 915-924.
Similar changes in both stability and agility indices are seen also
in other tasks, including whole-body tasks such as standing. Both
indices are sensitive to dopamine replacement therapy (Park et al.
2014; Falaki et al. 2017). It is possible that the lack of ASAs
contributes to the impaired ability of patients with Parkinson’s
disease to initiate and modify actions, which can lead to episodes of
freezing, an inability to make a step in certain conditions. Freezing
can be a major contributor to falls and to the overall disability of
patients at later stages of the disease.
37.5 Treatment of Parkinson’s
Disease
Despite the recent progress in the understanding of the origins of
Parkinson’s disease and associated neurophysiological
mechanisms, treatment of this disorder is mostly symptomatic. The
most common strategy has been to take pills containing a precursor
of dopamine (L-DOPA). This drug does indeed induce a strong
therapeutic effect. However, it is also associated with a number of
side effects, including nausea, dyskinesias (to be discussed in
section 37.9), mood swings, hallucinations, delusions, and paranoid
psychosis. These problems are amplified by the fact that patients
develop tolerance to L-DOPA and require a continuous increase in
the dosage. In addition, high dosages of L-DOPA and large diurnal
variations in its concentration can accelerate the death of neurons in
the substantia nigra. Some of the aforementioned problems have
been partly avoided or minimized by using drugs with slow release of
L-DOPA. This prevents large daily variations in the drug’s
concentration.
Other pharmaceutical approaches to Parkinson’s disease include
using inhibitors of enzymes, COMT (catechol-0-methyltransferase)
and monoamineoxydase (MAO-B) involved in the L-DOPA
transformation in the brain. This helps keep the dopamine in the
brain for longer times. There is also a view that healthy brain
functioning depends on the balance between two types of neural
projections, those using dopamine and those using acetylcholine as
neuromediators. This view justifies using anticholinergic agents in
combination with dopamine replacement therapy.
Novel approaches to the treatment of Parkinson’s disease have
been developed lately. They involve, in particular, the implantation of
electrical brain stimulators into structures within the basal ganglia,
typically into the external part of the globus pallidus or into the
subthalamic nucleus, thus affecting the indirect loop within the basal
ganglia. Promising clinical results have also been reported in studies
with stimulation of thalamic structures (Benabid et al. 1998; Lozano
and Mahant 2004; Vaillancourt et al. 2004). This method, deep brain
stimulation (DBS), has so far led to variable results, from very strong
therapeutic effects to minor effects and even reports of an increase
in the incidence of falls in patients with DBS (Marconi et al. 2008;
Cossu and Pau 2017). Studies of the effects of DBS on motor
synergies have shown positive effects on indices of agility (longer
and larger ASAs) but no significant effects on indices of stability
(Falaki et al. 2018).
PROBLEM 37.4
Should all newborns be tested for the gene responsible for
Huntington’s disease? Give your reasons pro and con.
PROBLEM 37.5
Some of the gait abnormalities in Huntington’s disease are very
similar to those in Parkinson’s disease (slow, stiff, unsteady gait
with episodes of freezing). However, one of these diseases is
classified as a hypokinesia, while the other is a hyperkinesia.
What can you conclude from these observations?
37.7 Hemiballismus
Another hyperkinetic disorder associated with an injury to a structure
within the basal ganglia is ballism. Ballism is due to a lesion of the
subthalamic nucleus. Such a lesion may result from a
cerebrovascular accident (stroke) or from a malformation. Since the
left and right subthalamic nuclei are relatively far from each other,
cases of bilateral lesions of both subthalamic nuclei are extremely
unlikely. Most commonly, this disorder is associated with a lesion of
one of the subthalamic nuclei leading to clinical signs limited to the
contralateral side of the body. Therefore, the disorder is frequently
termed hemiballismus.
Lesion of the subthalamic nucleus may be expected to lead to an
increase in the amount of excitation cortical structures get from the
thalamus (figure 37.8), leading to excessive movements on one side
of the body. Hemiballismus typically leads to uncontrollable rapid
movements of the contralateral limbs. These movements are
somewhat similar to those of chorea but typically faster and over
larger amplitudes. Treatment of hemiballismus is symptomatic,
similar to that of chorea.
37.8 Dystonia
Dystonia is defined as a syndrome of sustained muscle contractions
that produce twisting and repetitive movements and abnormal
postures. This rather vague definition unites cases of different
etiologies. However, a substantial number of patients with dystonia
show abnormalities in the basal ganglia, which justifies description of
this hyperkinetic disorder in this chapter.
Figure 37.8 Atrophy of the subthalamic nucleus leads to ballism (hemiballismus).
It results in changes in the indirect loop and higher excitatory input from the
thalamus to the cortex. Excitatory projections are shown with solid lines and open
circles; inhibitory projections are shown with dashed lines and filled circles.
Projections that are weaker than in a healthy person are shown with thin lines,
projections that are stronger with thick lines. Compare to figure 37.1.
PROBLEM 37.6
What conclusions could you draw about the origins of writer’s
cramp based on the fact that the same muscles and joint may
show no dystonic signs in other activities?
Figure 37.9 Typical patterns of a voluntary wrist flexion movement in a person
with dystonia. EMG patterns are characterized by multiple, irregular bursts. The
trajectory is “bumpy” and may have hesitations and reversals.
Reprinted by permission from M.L. Latash and S.R. Gutman, “Abnormal Motor Patterns in
the Framework of the Equilibrium-Point Hypothesis: A Cause for Dystonic Movements?”
Biological Cybernetics 71 (1995): 87-94. © 1995 Springer-Verlag.
CHAPTER 37 IN A NUTSHELL
Disorders of the basal ganglia may
lead to movement poverty (hypokinesia)
or excessive movements (hyperkinesia).
Parkinson’s disease is a consequence
of a loss of dopamine-producing
neurons in the substantia nigra. Its
symptoms include poverty of movements,
tremor at about 5 to 6 Hz, rigidity,
bradykinesia (slowness), and deficits
in postural control. A deficit in
anticipatory postural adjustment and a
poorly controlled increase in
corrective postural reactions have
been reported as contributing to
postural deficits. The most common
treatment for Parkinson’s disease is a
precursor of dopamine (L-DOPA).
Promising clinical results have been
obtained with deep brain stimulation
of nuclei in the thalamus and basal
ganglia. Huntington’s disease is a
genetic disorder associated with a
dysfunction of the caudate nucleus. It
leads to choreic movements and
dementia. Hemiballismus is a
consequence of a lesion of one of the
subthalamic nuclei; it leads to poorly
controlled, fast, large-amplitude
movements in the contralateral side of
the body. Dystonia is a disorder that
in some cases is linked to a
dysfunction of the basal ganglia.
Voluntary movements in dystonia are
characterized by twisted, sustained
postures of the limb segments, limbs,
neck, or trunk. Long-lasting
pharmacological treatment of chronic
neurological disorders can lead to
tardive dyskinesia. Its clinical signs
involve repetitive chewing movements
and tongue popping.
Chapter 38
Cerebellar Disorders
PROBLEM 38.1
What kinds of changes in descending effects on γ-motoneurons
may be expected to lead to decreased “muscle tone”?
PROBLEM 38.3
How would you distinguish the gait of a patient with a cerebellar
disorder from that of a person with Huntington’s chorea?
PROBLEM 38.4
What kinds of motor mistakes would you expect in a baseball
player with a mild cerebellar disorder who tries to hit or to catch a
baseball?
PROBLEM 38.5
Suggest a physiological mechanism for the check-and-rebound
phenomenon.
38.6 Ataxias
In humans, cerebellar disorders are accompanied by ataxia of
voluntary movements, a decomposition of movement into a number
of jerky segments (reviewed in Gilman 2004). Ataxia has been
associated with problems with movement initiation, termination, and
velocity control. A particular case of ataxia, when movement of
speech articulators is affected, is called ataxic dysarthria.
There is a group of ataxic disorders that are inherited and
characterized by discoordination and problems with balance. These
are typically progressive neurodegenerative disorders that can lead
to significant impairment and death. Probably the most common of
inherited ataxias is Friedreich’s ataxia; it happens at a rate of 1
case per 30,000 to 50,000 among Caucasians. It is limited to certain
geographic areas such as Europe, Northern Africa, the Middle East,
and India. Clinical signs of Friedreich’s ataxia start before the age of
25; they include dysarthria, loss of reflexes, and axonal sensory
neuropathy, including loss of cutaneous and proprioceptive
sensation. The same genetic mutation has been shown to lead in
some people to a later-onset ataxia involving hyperreflexia and signs
of spasticity.
The second most common among inherited ataxias is ataxia
telangiectasia. This disorder starts early in childhood; the patients
are wheelchair bound by their second decade, and their life span
does not extend beyond early adulthood. Ataxia telangiectasia leads
to progressive ataxia, dysarthria, facial hypotonia, and oculomotor
abnormalities.
Other forms of inherited ataxias include spinocerebellar ataxias,
which are relatively rare and limited to particular subpopulations.
Ataxias rarely respond to pharmacological treatment. Supportive
care and rehabilitation remain the only options for helping these
patients and their families.
PROBLEM 38.6
Suggest an explanation for the suppressed tonic vibration reflex in
patients with cerebellar disorders.
CHAPTER 38 IN A NUTSHELL
In humans, cerebellar lesions are
associated with injuries, tumors, or
demyelinating processes affecting
neural tracts that carry information
to and from the cerebellum. Cerebellar
lesions lead to gross motor disorders
that can include rigidity, tremor,
hypotonia, dysmetria, ataxia, and
asynergia. Tremor in persons with
cerebellar disorders has a static,
postural component and a kinetic,
intentional component. Cerebellar
rigidity may be mediated by
abnormalities in inputs to α- or γ-
motoneurons. There are changes in
posture, which becomes unsteady, and
gait, which is characterized by
irregular steps and excessive lifting
of the swing foot. Voluntary movements
by persons with cerebellar disorders
are fragmented and show poor
interjoint coordination. They also
show major disorders of timing,
particularly in serial movements.
Motor learning and adaptation are
severely affected. Inherited cases of
ataxia, in particular Friedreich’s
ataxia, commonly represent progressive
neurological disorders leading to
serious impairment of motor function.
Cerebellar dysfunction leads to
impaired synergic control of movements
involving both reduced stability of
steady states and reduced agility when
variables have to be changed quickly.
Cerebellar cognitive affective
syndrome represents an impairment of
such general abilities as planning,
abstract reasoning, working memory,
and spatial cognition.
Chapter 39
Cortical Disorders
39.2 Stroke
A cerebrovascular accident (CVA), commonly known as a stroke,
involves an interruption of the normal blood supply to a brain area as
a result of a blood vessel rupture (hemorrhagic stroke) or blocked
blood flow (ischemic stroke). The clinical picture depends strongly on
the area affected by the stroke and the extent of the accident.
Strokes that affect the medulla are life-threatening. They can lead
to problems with breathing and cardiac function. Problems with
voluntary movements can also be seen, mostly caused by
interruption of transmission along descending pathways.
When a stroke affects the brainstem, there are consequences
related to dysfunction of the many nuclei that are located in the
brainstem. In particular, brainstem strokes commonly lead to
vestibular problems and associated problems with postural control.
Oculomotor problems may lead to nystagmus, while dysarthria
(disordered speech) may result from dysfunction of cranial nuclei
that innervate articulators. Discoordination of movements can be
seen, similar to that typical of cerebellar disorders produced by a
dysfunction of the climbing fiber input into the cerebellum. In cases
of an injury to the corticospinal pathways, patients can show signs of
spasticity typical of strokes that affect the large hemispheres.
PROBLEM 39.1
What parts of the body are likely to show spastic signs following a
unilateral stroke affecting corticospinal pathways?
PROBLEM 39.2
What other brain circuits, in addition to those involving the basal
ganglia, contribute significantly to motor synergies?
39.3 Myoclonus
Myoclonus represents a brief muscle jerk caused by a neuronal
discharge (reviewed in Toro and Hallett 2004). It may be
accompanied by a single muscle burst or repetitive muscle bursts of
activity. Other movement disorders may accompany myoclonus,
such as dystonia (chapter 37) and essential tremor (chapter 40).
Myoclonus may occur in healthy people in special conditions.
Nocturnal leg myoclonus is the most common example: Jerky leg
movements occur rather frequently while falling asleep or in the
middle of the night. When myoclonus is the only or most important
neurological problem, it is called essential myoclonus. Most cases of
essential myoclonus are familial. Myoclonus may be part of
idiopathic epilepsy, particularly in children. Secondary myoclonus
may be associated with various causes including trauma, intoxication
(bismuth), renal failure, Huntington’s disease, olivopontocerebellar
atrophy, and corticobasal degeneration.
Clinically, myoclonus is classified into spontaneous, reflex, and
action (reviewed in Obeso and Zamarbide 2004). Spontaneous
myoclonus may affect a single muscle or a small group of muscles
(focal), several muscle groups (multifocal), or nearly the whole body
(generalized). Spontaneous myoclonus may be associated with a
seizure, known as myoclonic epilepsy. Spontaneous myoclonic
bursts of muscle activity may be singular or rhythmic; in the latter
case, the frequency of the muscle EMG bursts is relatively low,
between 1 and 4 Hz. Myoclonic activity may even persist in sleep.
Reflex myoclonus is triggered by sensory stimuli that can be of
different modalities: visual, auditory, or somatosensory. The term
action myoclonus refers to myoclonic discharges associated with
attempts to perform a voluntary action. It can affect both postural
muscles and focal prime movers. Figure 39.4 illustrates
schematically action myoclonus that affects several limb muscles. In
particular, when a standing person performs an arm movement,
myoclonic muscle activity can be seen in both arm muscles and in
leg and trunk postural muscles. Action myoclonus may be negative;
this means that it leads not to a burst of muscle activity but rather to
a silence period in an ongoing steady-state muscle activation. An
example of negative myoclonus is shown by the arrow in the right
part of Figure 39.4. Note that rapid EMG bursts seen in a distal
muscle may be accompanied by transient gaps in steady-state
activation—negative myoclonus—of a proximal muscle (Obeso et al.
1983, 1985). Negative myoclonus in postural muscles may
compromise postural stability.
Figure 39.4 A schematic illustration of myoclonus in a distal limb muscle (lower
trace) and negative myoclonus (transient gap in steady-state muscle activation) in
a proximal muscle of the same limb.
PROBLEM 39.3
How can a myoclonic discharge be distinguished from a spastic
spasm?
39.4 Tics
A tic is defined as a brief, repetitive, and seemingly purposeless,
stereotyped action that may involve one muscle or muscle groups.
Typically, tics are seen in childhood, but they may persist for a
lifetime. In children, tics are commonly associated with Tourette
syndrome (next section).
Tics are classified as being either primary or secondary. Primary
tics may be transient (lasting from 1 mo to 1 yr) or chronic (lasting
over 1 yr). Secondary tics may follow an infection (encephalitis) or a
basal ganglia disorder, or they may be drug-induced. The range of
drugs reported to have induced tics is very broad and includes
stimulants, anticonvulsants, L-DOPA, antidepressants, and even
birth control pills.
Tics may be fast (clonic) or sustained (dystonic). Patients report
that motor tics and vocal tics are under voluntary control—they follow
an urge to produce an action or a sound. However, researchers and
physicians view tics as involuntary; unlike voluntary movements, tics
are not preceded by a readiness potential (chapter 8). Sometimes,
brief sensory experiences precede tics (these are called sensory
tics). Table 39.1 presents a summary of common motor, vocal, and
sensory tics, both simple and complex.
The neurophysiological origin of tics is unknown, They are
believed to result from abnormalities in corticostriatothalamic
pathways—consequences of a disorder involving signal transmission
through the basal ganglia. Therapies for tics reflect the potential
involvement of the basal ganglia in these hyperkinetic disorders:
dopamine blockers and acetylcholine-like drugs have been used to
treat tics.
PROBLEM 39.4
How would you distinguish a tic from a myoclonic discharge?
Simple
Frequent blinking Sniffing Burning
Blepharospasm Grunting Tightness
Grimacing Throat clearing Muscle tension
Pouting Barking Tingling
Head jerking Coughing Itching
Jaw opening Growling Impulsion
Shoulder shrugging Moaning
Fist clenching Humming
Motor Vocal Sensory
Complex
Head twisting Panting Inner tension
Spitting Belching Pain syndrome
Hitting (self or others) Stuttering Phantom tics
Jumping Echolalia
Squatting Coprolalia
Pelvic/abdominal thrusting
PROBLEM 39.5
Another rare disorder is associated with a basically intact cortex
and severely abnormal cerebellum. What striking differences
would you expect to see in persons with that disorder and in
children with Williams disease?
CHAPTER 39 IN A NUTSHELL
Cortical lesions lead to a variety of
motor, sensory, and cognitive
consequences, depending on their site
and extent. Strokes result from an
interruption of normal blood supply to
an area of the brain. They lead to a
mixture of positive and negative
signs, primarily in the contralateral
half of the body, including weakness,
discoordination, and spasticity.
Synergy indices show only minor
changes after stroke, but anticipatory
synergy adjustments are reduced in
patients, reflecting difficulties with
movement initiation. Exercise of the
affected body parts seems to be most
effective for recovery after stroke.
Myoclonus represents brief muscle
jerks that may be triggered by a
sensory stimulus (reflex myoclonus) or
by a voluntary action (action
myoclonus), or they may occur
spontaneously. Tics are brief,
repetitive, and seemingly purposeless
stereotyped actions. They can be
vocal, motor, or sensory. Tics are
sometimes associated with Tourette
syndrome, a condition in young
children that is typically transient.
Williams syndrome is a rare disorder
characterized by reduced cerebral
volume and severely depressed
nonverbal IQ. In contrast, children
with Williams syndrome show high
verbal and grammatical fluency.
Chapter 40
Systemic Disorders
PROBLEM 40.1
Within a neural tract, thicker axons are more likely to stop
conducting action potentials in multiple sclerosis. Suggest an
explanation.
PROBLEM 40.2
Patients with multiple sclerosis frequently feel better in a cold
room and worse in a hot room. Suggest an explanation.
PROBLEM 40.3
Suggest an experiment that would be able to test a hypothesis
that essential tremor represents an auto-oscillation in a loop
involving sensory receptors.
Unless the tremor is disabling, it is not treated. In a large
percentage of patients, alcohol produces a dramatic temporary
suppression of the tremor. In cases where essential tremor
significantly interferes with daily activities, treatment strategies range
from beta-adrenergic blockers (e.g., propranolol), benzodiazepines
and other sedative drugs, peripheral agents inducing muscle paresis
(such as botulinum toxin), to anticonvulsant drugs, psychotherapy,
and hypnosis. In severe cases, thalamotomy and thalamic deep-
brain stimulation have been used.
PROBLEM 40.4
Children with cerebral palsy can sometimes show walking on tip-
toes. How would you distinguish this walk from Huntington’s
chorea?
PROBLEM 40.5
What effects of cholinergic drugs would you expect in Wilson’s
disease?
Figure 40.1 Effects of intrathecal baclofen on ankle clonus produced by a quick
passive ankle dorsiflexion in a person with cerebral palsy (solid lines = before;
dashed lines = after). Note the smaller bursts of muscle activity on the drug. GM =
gastrocnemius medialis; GL = gastrocnemius lateralis; SOL = soleus; TA = tibialis
anterior.
Reprinted by permission from M.L. Latash and R.D. Penn, “Changes in Voluntary Motor
Control Induced by Intrathecal Baciofen,” Physiotherapy Research International 1 (1996):
229-246. ©1996 Whurr Publishers Ltd.
Figure 40.2 Effects of intrathecal baclofen on movement kinematics and EMG
patterns during voluntary ankle plantarflexion by a person with cerebral palsy (solid
lines = before; dashed lines = after). Note the smoother trajectory and higher
velocity on the drug; the EMG levels are generally reduced. GM = gastrocnemius
medialis; GL = gastrocnemius lateralis; SOL = soleus; TA = tibialis anterior.
Reprinted by permission from M.L. Latash and R.D. Penn, “Changes in Voluntary Motor
Control Induced by Intrathecal Baciofen,” Physiotherapy Research International 1 (1996):
229-246. ©1996 Whurr Publishers Ltd.
CHAPTER 40 IN A NUTSHELL
Systemic disorders affect multiple
anatomical structures and pathways
within the central nervous system.
Amyotrophic lateral sclerosis leads to
progressive death of α-motoneurons,
resulting in muscle denervation and
loss of voluntary muscle control. It
is associated with neuronal loss in
cortical areas involved in the motor,
sensory, and cognitive functions.
Multiple sclerosis is a disorder
characterized by a loss of the myelin
sheath by the axons of neural tracts
within the central nervous system,
leading to various clinical
consequences depending on which tracts
are affected. Its cause is an
autoimmune process that can be
triggered by a viral infection.
Multisystem atrophy leads to a
combination of clinical signs and
symptoms that are seen in patients
with basal ganglia and cerebellar
dysfunctions. Its initial signs
commonly resemble those seen in
Parkinson’s disease or in patients
with cerebellar ataxia. Essential
tremor is a common disorder involving
various brain structures, in
particular the cortex, the cerebellum,
and the inferior olives. Cerebral
palsy is an inborn nonprogressive
disorder caused by problems during
gestation and delivery. Disorders of
the motor function in cerebral palsy
include discoordination, spasticity,
dystonia, and dysarthria. Wilson’s
disease is caused by copper deposits
in the brain (cortex and basal
ganglia). It leads to a combination of
signs resembling those in Parkinson’s
disease and unusual low-frequency
tremor characterized by wing-beating,
crescendo, and spreading.
Chapter 41
Motor Rehabilitation
PROBLEM 41.1
Minimization of energy spending has been viewed as a general
principle underlying motor coordination. Present examples of
behaviors when this is likely to be true and when it is likely to be
false.
PROBLEM 41.2
Plastic changes following training commonly lead to higher
excitability of involved cortical areas. Suggest a task when
practice can be expected to lead to decreased excitability of
involved cortical areas.
PROBLEM 41.3
Suggest a method of helping patients with Parkinson’s disease to
overcome episodes of freezing during gait based on the “lines
drawn on the floor” idea but useful for everyday walking tasks.
PROBLEM 41.4
Suggest a method of communication with a “locked-in” person
who cannot perform any voluntary movements.