Module 6
Vision, Proprioception, and Movement
a. Light and the Visual Apparatus
Vision is an impressive capability. There are approximately 97 million light
receptors in the human eye, and a complex network of cells connecting them to each
other and to the optic nerve. The optic nerve boasts about 1.2 million axons,
compared with 30,000 in the auditory nerve. The high ratio of retinal receptors to
optic nerve axons tells us that a great deal of processing goes on in the retina itself.
The optic nerve transmits information to the brain at an estimated 100 million bits per
second, comparable to Ethernet data transmission rates (K. Koch et al., 2006). What
our brain does with the information it receives from the eye is equally remarkable.
The topics of vision and visual perception form an exciting story, one of high-tech
research and conflicting theories and dedicated scientists’ lifelong struggles to
understand our most amazing sense.
To understand vision, we need to start at the beginning by describing the
adequate stimulus, as we did with audition. To say that the stimulus for vision is light
seems obvious, but the point needs some elaboration. Visible light is a part of the
electromagnetic spectrum. The electromagnetic spectrum includes a variety of energy
forms, ranging from gamma rays at one extreme of frequency to the radiations of
alternating current circuits at the other.
The visible part of the electromagnetic spectrum accounts for only 1/70 of the
frequency range. Most of the frequencies are not useful for producing images; for
instance, AM, FM, and analog television waves pass right through objects. Some of
the other energy forms, such as X-rays and radar, can be used for producing images,
but they require powerful energy sources and special equipment for detecting the
images. Heat-producing objects give off infrared energy, which some nocturnal
animals (such as the sidewinder rattlesnake) use to detect their prey in darkness.
Humans can convert infrared images to visible ones with the aid of specialized
equipment, and this capability is very useful to the military and law enforcement for
detecting heatproducing individuals, heat-producing vehicles, and armament at night.
Sometimes this ability can be used to deceive heat-detecting equipment; during the
first Gulf War, the Iraqi army set up plywood silhouettes of tanks with heaters behind
them to distract Allied airplanes. But infrared is best for detecting nearby objects—
distant ones have blurred edges and fuzzy detail. The electromagnetic energy within
our detectable (visible) range produces well-defined images because it is reflected
from objects with minimal distortion. We are adapted to life in the daytime, and we
sacrifice the ability to see in darkness in exchange for crisp, colorful images of faces
and three-dimensional (3-D) objects in daylight. In other words, our sensory
equipment is adapted for detecting the energy that is most useful to us, just as the
night-hunting sidewinder rattlesnake is equipped to detect the infrared radiation
emitted by its prey and a bat’s ears are specialized for the high-frequency sound
waves it bounces off small insects.
Light is a form of oscillating energy and travels in waves just as sounds do.
We could specify visible light (and the rest of the electromagnetic spectrum) in terms
of frequency, just as we did with sound energy, but the numbers would be extremely
large. So we describe light in terms of its wavelength—the distance the oscillating
energy travels in one complete cycle. (We could do the same with sound, but those
numbers would be just as inconveniently small.) The unit of wavelength is the
nanometer (nm), which is a billionth of a meter; visible light ranges from about 380 to
800 nm. Notice that different wavelengths correspond to different colors of light; for
example, when light in the range of 500 to 570 nm strikes the receptors in our eye, we
normally report seeing green.
The eye is a spherically shaped structure filled with a clear, thick liquid. The
white outer covering, or sclera, is opaque except for the cornea, which is transparent.
Behind the cornea is the iris, which gives your eye its color. The iris is a circular
muscle that controls the amount of light entering your eye by contracting reflexively
in bright light and relaxing in dim light. You can observe this response in yourself by
watching in a mirror while you change the level of light in the room. The pupil is not
a structure but simply an opening surrounded by the iris muscle; it looks black
because light that enters your eye is absorbed by the retina and isn’t reflected out.
Behind the iris is the lens. Notice in the figure that the lens projects the image upside
down and backward onto the retina. You can demonstrate this by touching the side of
your eye through your closed lid—you’ll be manually stimulating the retina on one
side, but you’ll see a flashing spot on the other side of your eye—if you move your
finger down, the spot moves up. Because the normally flat lens is a remarkably
flexible tissue, the ciliary muscles attached to it can contract to make the lens rounder
to focus on a near object through a process called accommodation.
The retina, the light-sensitive tissue at the rear of the eye, is made up of two
main types of light-sensitive receptor cells, called rods and cones, and the neural cells
connected to them. The photoreceptors connect to bipolar cells, which in turn connect
to ganglion cells, whose axons form the optic nerve. Notice that light must pass
through the bipolar and ganglion cells to reach the photoreceptors; the neural cells are
very small and transparent, though, so they block very little light. The photoreceptors
are filled with light-sensitive chemicals called photopigments. The photopigment
absorbs energy from the light that causes some of the molecules to break down into
two components, and the ensuing chemical reaction ultimately results in a neural
response. The two components then recombine to maintain the supply of
photopigment.
The rod photopigment is called rhodopsin; the name refers to its color (from
the Latin rhodon, “rose”), not to its location in rods. Rhodopsin absorbs light much
faster than cone photopigment, so it is used exclusively for low-light situations. In
very bright light, the rhodopsin in your eyes remains broken down most of the time,
so the rods are not useful. The brief delay in adjusting to a darkened movie theater is
due to the time it takes the rhodopsin to resynthesize. Iodopsin, the cone
photopigment, requires a high light intensity level, so your cones function well in
daylight but are nonfunctional in dark situations. Three varieties of iodopsin, located
in different cone types, respond best to different wavelengths of light; this means that
cones detect only certain different wavelengths, whereas rods differentiate only
among different levels of light and dark (which is why you cannot recognize colors in
dim light).
Rods and cones also differ in their retinal distribution and in their amount of
neural interconnection. Cones are most concentrated in the fovea, a 1.5-millimeter-
diameter circle in the middle of the retina, and drop off rapidly with distance from that
point. Rods are most concentrated at 20 degrees from the fovea; from that point, they
decrease in number in all directions and fall to zero in the fovea. In the center of the
fovea, one cone synapses on a single bipolar cell, which synapses on a single ganglion
cell (Masland, 2017); the number of cones per ganglion cell increases with distance
from the center but remains small compared with rods. Because few cones share
ganglion cells, the fovea has higher visual acuity, or ability to distinguish details.
Acuity gets a further boost at the fovea because the connecting cells are pushed to the
side, eliminating the slight scattering of light that occurs in the rest of the retina. In
contrast to cones, many rods share a single ganglion cell; this reduces their resolution
but enhances their already greater sensitivity to dim light. The area of the retina from
which a ganglion cell (or any other cell in the visual system) receives its input is the
cell’s receptive field. So we can say that receptive fields are smaller in the fovea and
larger in the periphery.
The receptors’ response to light is different from what you might expect,
because they are most active when they are not being stimulated by light. In darkness,
the photoreceptor’s sodium and calcium channels are open, allowing these ions to
flow in freely. Thus, the membrane is partially depolarized; the receptor releases a
continuous flow of glutamate, and this inhibits activity in the bipolar cells. The
chemical response that occurs when light strikes the photopigments closes the sodium
and calcium channels, reducing the release of glutamate in proportion to the amount
of light. The bipolar cells release more neurotransmitters, which increases the firing
rate in the ganglion cells. (The photoreceptors and bipolar cells do not produce action
potentials.) In neural terms, ganglion cells undergo what is called “release from
inhibition.” In addition, amacrine cells connect across many ganglion cells. All this
interconnection might suggest to you that the retina does more than transmit
information about points of light to the brain. Along with the high ratio of receptors to
ganglion cells, this is a second indication that considerable processing occurs before
the signal leaves the retina. You will soon see evidence that this is true. With such
complexity, it’s no wonder most vision scientists consider the retina to be part of the
brain and refer to the optic nerve as a tract!
The axons of the ganglion cells join and pass out of each eye to form the two
optic nerves. Where the nerve exits the eye, there are no receptors, so it is referred to
as the blind spot. This point is located about 20 degrees toward the outside and down
slightly from the usual focal point of each retina. The blind spots of the two eyes fall
at different points in a visual scene, so you do not notice that any of your visual world
is missing; besides, your brain is good at “filling in” missing information, even when
a small part of the visual system is damaged. The two optic nerves run to a point just
in front of the pituitary gland, where they join for a short distance at the optic chiasm
before separating again and traveling to their first synapse in the lateral geniculate
nuclei of the thalamus. At the optic chiasm, axons from the nasal sides of the eyes
cross to the other side and go to the occipital lobe in the opposite hemisphere.
Neurons from the outside half of each eye (the temporal side) do not cross over but go
to the same side of the brain.
It seems like splitting the output of each eye between the two hemispheres
would cause a major distortion of the image can see that the arrangement keeps
related information together. Notice that the letter A, which appears in the person’s
left visual field (the part of the environment being registered on the retina), casts an
image on the right half of each retina. The information from the right half of each eye
will be transmitted to the right hemisphere. An image in the right visual field will
similarly be projected to the left hemisphere. This is how researchers who study
differences in the functions of the two cerebral hemispheres can project a visual
stimulus to one hemisphere. They present the stimulus slightly to the left or to the
right of the midline; the exposure time is too brief to allow the research participant to
shift the eyes toward the stimulus, and with brief exposures, the information does not
transfer to the other hemisphere.
b. Color Vision
After observing the effect of passing light through a prism, Sir Isaac Newton
proposed in 1672 that white light is composed of seven fundamental colors that
cannot themselves be resolved into other colors. If there are seven “pure” colors, this
would suggest that there must be seven receptors and brain pathways for
distinguishing color, just as there are five primary tastes (or six, if the fat receptor is
confirmed). In 1852, Hermann von Helmholtz revived an idea of Thomas Young
from a half century earlier. Because combining different amounts of just three colors
of light can produce any color, Young and Helmholtz recognized that this must be due
to the nature of the visual mechanism rather than the nature of light. They proposed a
trichromatic theory—now known as the Young-Helmholtz theory—that just three
color processes account for all the colors we are able to distinguish. They chose red,
green, and blue as the primary colors because observers cannot resolve these colors
into separate components as, for example, you can see red and blue in the color
purple. When you watch television or look at an electronic device screen, you see an
application of trichromatic color mixing: All the colors you see on the screen are
made up of different intensities of tiny red, green, and blue dots (pixels) of light.
The trichromatic theory accounted for some of the observations about color
perception very well, but it ran into trouble explaining why yellow also appears to
observers to be a pure color. Ewald Hering (1878) “solved” this problem by adding
yellow to the list of physiologically unique colors. But rather than assuming four color
receptors, he asserted that there are only two—one for red and green and one for blue
and yellow. Opponent process theory attempts to explain color vision in terms of
opposing neural processes. In Hering’s version, the photochemical in the red-green
receptor is broken down by red light and regenerates in the presence of green light.
The chemical in the second type of receptor is broken down in the presence of yellow
light and regenerates in the presence of blue light.
The trichromatic and opponent process theories appear to be contradictory.
Sometimes this means that one position is wrong and the other is right, but often it
means that each of the competing theories is partly correct but just too simple to
accommodate all the known facts. Hurvich and Jameson (1957) resolved the conflict
with a compromise: They proposed that three types of color receptors—red sensitive,
green sensitive, and blue sensitive—are interconnected in an opponent process
fashion at the ganglion cells.
Long-wavelength light excites “red” cones and the red-green ganglion cell to
give the perception of red. Medium-wavelength light excites the “green” cones and
inhibits the red-green ganglion cell, reducing its firing rate below its spontaneous
level and signaling green to the brain. Likewise, short- wavelength light excites
“blue” cones, which inhibits the yellow-blue ganglion cell, leading to a perception of
blue. Light midway between the sensitivities of the “red” and “green” cones would
stimulate both cone types. The firing rate in the red-green ganglion cell would not
change, because equal stimulation and excitation from the two cones would cancel
out; however, the cones’ connections to the yellow-blue ganglion cell are both
excitatory, so their combined excitation would excite the ganglion cell to produce a
perception of yellow. If all three cones are stimulated at equal intensities, both
ganglion cell types receive equal amounts of excitatory and inhibitory stimulation
and, therefore, produce a perception of white (or a shade of gray, depending on the
intensity of the light). According to this combined theory, there are three color
processes at the receptors and four beyond the ganglion cells.
Recently, researchers were able for the first time to study individual cones in
the eyes of human volunteers. Surprisingly, stimulation of two thirds of the cones
identified as red sensitive or green sensitive yielded a response of “white” much more
often than the expected “red” or “green”. This makes sense, because very few cones
are needed to determine an object’s color, but numerous cells are required for
discriminating fine details. This emphasis on light detection in the majority of cones
may be useful for some proposed treatments for retinal blindness (see A Further Look
on restoring vision), in that improving cone function will increase overall light
perception. You may wonder how the researchers were able to make repeated
measurements of photocells, given that the eye is constantly moving. Ingeniously,
they used the same technology that enables a telescope to track a star.
The World Health Organization (2019) reports that at least 2.2 billion people
have a vision impairment or blindness; fortunately, researchers now have a variety of
promising strategies for restoring lost vision. When the photoreceptors have lost
function due to retinitis pigmentosa or macular degeneration, they can be bypassed by
electrodes inserted into the retina. The Argus II, for example, uses a video camera and
a processor to send a pattern of activation wirelessly to an array of 60 electrodes,
which stimulate the ganglion cells. The Alpha AMS replaces the camera with 1,600
light-sensing photodiodes inserted among the bipolar cells (Edwards et al., 2018). The
processor communicates between the photodiodes and the electrodes via a coil
mounted on the scalp behind the ear. Eliminating the external camera allows the user
to look around by moving the eyes rather than turning the head. The minimal
resolution of these two methods enables basic capabilities such as distinguishing
people from background and identifying food on a plate.
Of course, the ideal would be to make the damaged retina respond to light on
its own, and research teams are making progress toward accomplishing this.
Optogenetic methods have been used with some success in rodents to insert light-
sensitive channels or light-sensitive substances into bipolar cells and ganglion cells,
so those cells are able to respond to light (Baker & Flannery, 2018). Another
alternative is the use of stem cells. Inserting retinal pigment epithelium (a layer that
supports and nourishes the retina) grown from stem cells rescued degenerating
photoreceptors and improved vision in four of five patients; in another study, it
enabled reading at 60 to 80 words per minute in two individuals who had been unable
to read at all before (da Cruz et al., 2018; Kashani et al., 2018). Introducing the
RPE65 gene into the retinas of individuals deficient in an enzyme required for
receptor functioning and survival substantially improved vision in 65% of the
participants; the procedure won the first FDA approval for gene therapy for an
inherited retinal disease (Russell et al., 2017). But gene therapy requires finding the
right gene and discovering the best carrier for introducing the gene. Three attempts to
treat age-related macular degeneration have been unsuccessful, but additional trials
are in progress.
Additional confirmation of the trichromatic/opponent process theory came
when researchers found the predicted color-opponent cells in monkeys, both in the
retina and in the lateral geniculate nucleus of the thalamus of opponent cells, one that
is excited by red and inhibited by green (R+G−) and one that is excited by yellow and
inhibited by blue (Y+B−); Russell De Valois and his colleagues (1966) identified two
additional types that were the inverse of the previous two: green excitatory/red
inhibitory (G+R−) and blue excitatory/yellow inhibitory (B+Y−). A surprise was that
some of the color-opponent ganglion cells receive their input from cones that are
arranged in two concentric circles (Gouras, 1968; Wiesel & Hubel, 1966). The cones
in the center and those in the periphery have color-complementary sensitivities. Why
all this complexity? First, the opposition of cones at the ganglion cells provides
wavelength discrimination that individual cones are incapable of producing (E. B.
Goldstein, 1999)—an example of the “neural comparison” we referred to earlier. The
concentric-circle receptor fields also enhance information about color contrast
between adjacent areas. This mode of information sharpening will become clearer
when we look at how the retina distinguishes the edge of an object.
There are two major types of retinal color deficiency. A person who is red-
green colorblind sees these two colors but is unable to distinguish between them (this
is either protanopia with absence of red receptors, or deuteranopia with absence of
green receptors). We know something about what color-blind people experience by
noting which colors color-blind people confuse and from studying a few rare
individuals who are color-blind in only one eye. When a red-green color-blind
colleague of one of the authors was describing his experience of color, he explained
that green grass appeared to be the same color as peanut butter! We’re not sure what
peanut butter actually looked like to him, but he did say that he found grass and trees
“very beautiful.” People in the much rarer second color-blind group, tritanopia, have
difficulty distinguishing blue, so their world appears in variations of red and green.
People with red-green color blindness show a deficiency in either the red end
of the spectrum or in the green portion; this suggests that the person lacks either the
appropriate cone or the photochemical. Acuity is normal in both groups, so there
cannot be a lack of cones. Some individuals are unusually sensitive to green light, and
the rest are sensitive to red light; this suggests that in one case the normally
redsensitive cones are filled with green-sensitive photochemical and in the other the
normally green-sensitive cones are filled with red-sensitive chemical. The condition is
due to a mutation on the X chromosome, so it is considerably more prevalent in males
(5.0% to 8.0%, versus 0.5% to 1.0%). Tritanopia is usually due to an inherited lack of
blue-receptive cones, but it can be caused by blunt trauma to the eye. Unlike redgreen
color blindness, the deficiency is not X-linked, so it occurs with equal frequency in
males and females (about 1.0%).
c. Form Vision
Just as the auditory cortex is organized as a map of the cochlea, the visual
cortex contains a map of the retina. Russell De Valois and his colleagues
demonstrated this point when to monkeys that had been injected with radioactive 2-
deoxyglucose. The animals were sacrificed and their brains placed on photographic
film. Because the more active neurons absorbed more radioactive glucose, they
exposed the film more darkly this produced an image of the stimulus that appears to
be wrapped around the monkey’s occipital lobe.
This result tells us that just as there is a tonotopic map of the basilar
membrane in the auditory cortex, we have a retinotopic map in the visual cortex,
meaning that adjacent retinal receptors activate adjacent cells in the visual cortex.
However, this does not tell us how we see images; transmitting an object’s image to
the cortex like a television picture does not amount to perception of the object. Object
perception is a two-stage affair. In this section we will discuss form vision, the
detection of an object’s boundaries and features (such as texture). We will discuss the
second component, object recognition, a bit later. The story that unfolds here is about
more than perception; it provides a model for understanding how the brain processes
information in general. It is also a story that begins not in the cortex but in the retina
itself.
Detecting an object’s boundaries is the first step in form vision. The nervous
system often exaggerates especially important sensory information. In the case of
boundaries, the retina uses lateral inhibition, where each neuron’s activity inhibits the
activity of its neighbors and in turn they inhibit its activity, to enhance the contrast in
brightness that defines an object’s edge. The Hermann grid is the more dramatic of the
two illusions, but the simplicity of the Mach band graphic makes it easier to explain,
so we will focus on it. Each bar in the Mach band image is consistent in brightness
across its width, but it looks a bit darker on the left and a bit lighter on the right than it
does in the middle. (If you don’t see a difference at the edges, you may notice that the
bars seem slightly curved. This is because the illusion suggests subtle shadowing on
the left side of each bar.) An illusion is not simply an error of perception but an
exaggeration of a normal perceptual process, which makes illusions very useful in
studying perception.
Cells in the lateral geniculate nucleus (LGN) have circular receptive fields just
like the ganglion cells from which they receive their input. The receptive fields of
visual neurons in the cortex, however, turn out to be surprisingly different. David
Hubel and Torsten Wiesel (1959) were probing the visual cortex of anesthetized cats
as they projected visual stimuli on a screen in front of the cats. Their electrode was
connected to an auditory amplifier so that they could listen for indications of active
cells. They were manipulating a glass slide with a black dot on it in the projector—a
very effective stimulus with ganglion cells and LGN cells—and getting only vague
and inconsistent responses.
In other layers of the cortex, Hubel and Wiesel found complex cells. A
complex cell continues to respond when a line or an edge moves to a different
location, as long as it is not too far from the original site. They explained the complex
cell’s ability to continue responding in essentially the same way they explained the
sensitivity of simple cells. They assumed that complex cells receive input from
several simple cells that have the same orientation sensitivity but whose fields are
adjacent to each other on the retina. Notice that as the edge moves horizontally,
different simple cells will take over, but the same complex cell will continue
responding. However, if the edge rotates to a different orientation, this complex cell
will stop responding, and another complex cell specific for that orientation will take
over. Connecting several simple cells to a single complex cell enables the complex
cell not only to keep track of an edge as it moves but also to detect movement.
The feasibility of this kind of arrangement has received support from an
interesting source—artificial neural networks. Lau, Stanley, and Dan (2002) trained a
network so that its output “neurons” gave the same responses to bar-shaped stimuli as
those recorded from complex cells in cats. Then they examined the hidden processing
layer and found that those “neurons” had rearranged their connections to approximate
simple cells, complete with “on” and “off” regions in their receptive fields as well as
directional sensitivity. In an earlier study, a neural network was trained to recognize
curved visual objects (Lehky & Sejnowski, 1990). Its “neurons” spontaneously
developed sensitivity to bars or edges of light even though they had never been
exposed to such stimuli, suggesting that the Hubel-Wiesel model is a very versatile
one. But so far, Hubel and Wiesel had seen only the beginnings of the intricate neural
organization that makes visual perception possible. They lowered electrodes
perpendicularly through a monkey’s striate cortex; as the electrode passed through
simple and complex cells, the cells’ preferred width and length changed, but they had
the same orientation (Hubel & Wiesel, 1974). As the researchers moved the electrode
slightly to the side, the preferred orientation of those cells shifted slightly but
systematically in a clockwise or counterclockwise direction; over a distance of 0.5 to
1.0 mm, the orientation would progress through the complete circle. A complete 360-
degree “set” of preferred orientations tends to be organized around a common point,
forming what are called orientation pinwheels or whorls. This arrangement is repeated
in the adjacent cortex, with the input coming from an adjacent part of the retina. This
sort of organization is typical of the cortex’s efficiency in processing and transmitting
information. Connections mostly run up and down in columns, with much shorter
lateral connections. In addition, similar functions are clustered together, increasing
communication speed and reducing energy requirements.
Although some cortical cells respond best to edges, other cells apparently are
not so limited. Think of an edge as an abrupt or highfrequency change in brightness;
the more gradual changes in brightness across the surface of an object are low-
frequency changes. According to De Valois, some complex cells are “tuned” to
respond to the high frequencies found in an object’s border, while others are tuned to
low frequencies, for example, in the slow transition from light to shadow that gives
depth to the features of a face (De Valois et al., 1985). Some cells respond better to
“gratings” of alternating light and dark bars—which contain a particular combination
of spatial frequencies—than they do to lines and edges. According to spatial
frequency theory, visual cortical cells do a Fourier frequency analysis of the
luminosity variations in a scene. According to this view, different visual cortical cells
have a variety of different sensitivities, not just those required to detect edges.
A few photographs should help you understand what we mean by spatial
frequencies, as well as the importance of low frequencies. The result is a number of
high-frequency transitions, and the image is not very meaningful. It seems
paradoxical that blurring an image would make it more recognizable, but blurring
eliminates the sharp boundaries. Inspired by reading about the research in a Scientific
American article, the Spanish artist Salvador Dalí incorporated the illusion into one of
his more famous paintings.
d. The Perception of Objects, Color, and Movement
One of the more interesting characteristics of the visual system is how it
dissects an image into its various components and analyzes them in different parts of
the brain. The separation begins in the retina and increases as visual information flows
through all four lobes of the brain, with locations along the way carrying out analyses
of color, movement, and other features of the visual scene. Thus, we will see how
visual processing is, as mentioned earlier, both modular and hierarchical. Modular
processing refers to the segregation of the various components of processing into
separate locations. Hierarchical processing means that lower levels of the nervous
system analyze their information and pass the results on to the next higher level for
further analysis. Some neuroscientists reject the modular notion, arguing that any
visual function is instead distributed, meaning that it occurs across a relatively wide
area of the brain. One study found evidence that sensitivity to faces, for example, is
scattered over a large area in the temporal lobe (Haxby et al., 2001). Different studies
have supported both of these views (J. D. Cohen & Tong, 2001), and it can be argued
that vision involves a mix of modular and distributed functioning, rather like the
arrangement we saw for language. In the following discussion, you will see that basic
visual analysis occurs in isolated modules, and more comprehensive analysis is
distributed across the width and breadth of the brain.
Most visual information follows two routes from the retina through the brain,
which make up the parvocellular system and the magnocellular system (M.
Livingstone & Hubel, 1988; P. H. Schiller & Logothetis, 1990). Parvocellular
ganglion cells are smaller than magnocellular cells, account for the large majority of
ganglion cells, and are most numerous in the fovea. They have circular receptive
fields that are small and color opponent, which suits them for the specialties of the
parvocellular system, the discrimination of fine detail and color. Magnocellular
ganglion cells have large circular receptive fields that are brightness opponent and
respond rapidly but only briefly to stimulation. As a result, the magnocellular system
is specialized for brightness contrast and for movement.
plest example is that at dusk our sensitivity to light increases, but we lose our
ability to see color and detail. You cannot read a newspaper under such conditions or
color coordinate tomorrow’s outfit, because the high-resolution, color-sensitive
parvocellular system is nearly nonfunctional. The magnocellular system’s sensitivity
to movement is most obvious in your peripheral vision. Hold your arms outstretched
to the side while you look straight ahead, and move your hands slowly forward while
wiggling your fingers. When you just notice your fingers moving, stop. Notice that
you can barely see your fingers but you are very sensitive to their movement. Notice
that you see considerable depth in (a); this is because the bicycle differs from the
background in brightness, so the image stimulates primarily the magnocellular
system. The bicycle in (b) looks “flat”; the image has color contrast but little
brightness contrast, so it stimulates the magnocellular system minimally.
Magnocellular neurons arrive in V1 in areas that are responsive to orientation,
movement, and retinal disparity. The dorsal stream then proceeds through V2 to V5,
also known as MT because it is on the middle temporal gyrus in the monkey; neurons
there have strong directional sensitivity, which contributes to the perception of
movement. The dorsal stream travels then to the posterior parietal cortex, the area just
behind the somatosensory cortex; its role is primarily to locate objects in space, but
the behavioral implications of its functions are far more important than that simple
statement suggests. Movement perception is a good example of how modular and
distributed processing work together. V5/MT and a nearby area that receives input
from MT, known as MST (for medial superior temporal area), appear to be the most
important areas for perceiving movement. They receive most of their input from the
magnocellular pathway, including complex cells that are sensitive to movement; they
also respond when the motion is only implied in a photograph of an athlete in action
or a picture of a cup falling off a table. At the same time, there are many other areas
that are specialized for particular kinds of movement. Viewing movement of the
human body or its parts activates dorsal stream areas adjacent to V5/MT and MST, in
the parietal and frontal lobes and in the ventral stream in the temporal lobes. Images
move across your retinas every time you move your eyes, but you don’t see the world
moving around you. (Imagine trying to read otherwise.)
This is because the activity of movement-sensitive cells in MT and MST is
suppressed during eye movements. These cells are sensitive to movement in a
particular direction, and some of them reverse their preferred direction of movement
as the head moves, which allows them to continue responding to real movement of
objects. The brain’s visual movement areas are close to an area that processes input
from the vestibular organs, which monitor body motion you are already indirectly
familiar with this fact if you get motion sickness in a moving car when you read or
when you watch roadside objects too closely. The functions of the ventral “what” and
dorsal “where” streams are best illustrated by a comparison of patients with damage
in the two areas. People with damage in the temporal cortex (ventral stream) have
trouble visually identifying objects, but they can walk toward or around the objects
and reach for them accurately. People with damage to the dorsal stream have the
opposite problem. They can identify objects, but they have trouble orienting their gaze
to objects, reaching accurately, and shaping their hands to grasp an object using visual
cues. So the dorsal “where” stream also functions as a “how” area that is important for
action.
While some processing of face information occurs in the inferior temporal
cortex, recognizing individual faces requires additional structures. In humans, a part
of the fusiform gyrus located in the inferior temporal cortex is so important to face
recognition that it is referred to as the fusiform face area (FFA). This area begins
assisting us in recognizing caregiver faces at about 2 years of age (Bushnell, 2001)
but can be developmentally delayed in individuals with autism. Damage that results in
prosopagnosia is usually in the right hemisphere, but face processing is a cooperative
effort involving both sides of the brain. Face-like images produce activity in the left
fusiform gyrus that is proportionate to the image’s resemblance to a face; the right
fusiform gyrus becomes active 2 two seconds later, only when the image is of a
human face (M. Meng, Cherian, Singal, & Sinha, 2012). Another area that has been
implicated in face recognition is the medial temporal lobe, which is important in
memory and seems to be involved in decision making between familiar and
unfamiliar faces.
Until fairly recently, researchers thought the only way prosopagnosia occurred
was through brain damage. Then medical student Martina Grueter began to recognize
the symptoms in her husband’s behavior and made congenital prosopagnosia the
subject of her MD thesis (Grueter, 2007). An estimated 2.5% of the population has
symptoms of the disorder without any history of brain damage (Kennerknecht et al.,
2006); thus, they make errors in recognizing familiar faces, and they learn new faces
slowly. Afflicted individuals include noted primatologist Jane Goodall, actor Brad
Pitt, and Oliver Sacks, the neurologist who studied Dr. P. Face recognition ability has
a heritability of about 39% (Zhu et al., 2010), so its deficiency in the absence of brain
damage likely has a genetic origin. The defect, though, does not appear to be in the
fusiform face area itself; the FFA responds normally, but connections to more anterior
temporal and frontal cortex areas are diminished (Avidan & Behrmann, 2009). This
suggests that face recognition is a distributed function, despite modularity of its
components. These capabilities might be “hardwired” at birth to some extent, but they
also are amenable to learning. When researchers showed monkeys pictures of the
faces of lab workers, neurons in the inferior temporal cortex increased their firing
rates according to the monkeys’ familiarity with the workers (M. P. Young & Yamane,
1992). Isabel Gauthier and her colleagues (1999) trained humans to identify faces,
using pictures of fictitious creatures they called “greebles” to ensure initial
unfamiliarity. The fMRI that pictures of human faces activated the FFA, but greebles
did so only after the person had learned to recognize individual creatures.
People with prosopagnosia do respond emotionally to photographs of familiar
faces they do not recognize, as indicated by EEG-evoked potentials and skin
conductance response. This “hidden perception” is not without precedent. Patients
blinded by damage to V1 show a surprising ability to track the movement of objects
and discriminate colors, all the while claiming to be guessing (Zeki, 1992). Cortically
blind individuals also can identify emotions expressed in faces they do not otherwise
see (Tamietto et al., 2009), and they can avoid obstacles while walking. This ability to
respond to visual stimuli that are not consciously seen is called blindsight. Imaging
studies have found that blindsight depends on pathways passing through the superior
colliculus directly to extrastriate areas, bypassing V1.
e. The Body Senses
We get information about our body from the somatosensory system and from
the vestibular system. The somatosenses include proprioception; the skin senses,
which tell us about conditions at the surface of our body; and the interoceptive
system, concerned with sensations in our internal organs. The vestibular system
informs the brain about body position and movement. The interoceptive system
operates mostly in the background and participates less directly in behavior, so we
will limit our attention to the other systems.
Proprioception (from the Latin proprius, “belonging to one’s self”) is the sense
that informs us about the position and movement of our limbs, body, and head. Its
sensors report tension and length in muscles and the angle of the limbs at the joints. It
also keeps track of our head as we move about the world. Proprioception is not as
glamorous a sense as vision or audition, or even touch. However, without it, we would
have a great deal of difficulty, like Michael J. Fox, in maintaining posture, moving our
limbs, and grasping objects. This sense is what allows you to pick up your cell phone
from where you put it down, without looking. Oliver Sacks (1990) tells the story of
Christina, who lost all sense of proprioception following a bout with neuritis, an
inflammation of the nerves that is often caused by a viral infection. Deprived of
proprioceptive feedback, for the first month, she was floppy as a rag doll, unable even
to sit upright. After a year of rehabilitation, she was able to resume a reasonably
normal life, relying solely on vision to sit and stand erect and to walk. Ian Waterman,
who is similarly afflicted, crumples helplessly onto the floor if someone turns the
lights out (J. Cole, 1995). In other words, proprioception does more than provide
information; it is critically important in the control of movement.
The commonly accepted skin senses are touch, warmth, cold, texture, and
pain. However, three studies involving mice have found evidence that itch and tickle
have their own receptors, neural pathway (spinothalamic tract), and receptor proteins;
itch (which also goes by the interesting name of pruriception) will most likely be
added to the list of skin senses. And there is evidence that tickle is processed by itch
receptors as well. Whatever the actual number of skin senses, the important point here
is that each is distinct from the others, with its own receptors and separate “labeled
line” pathway to the brain. To demonstrate this for yourself, move the point of a lead
pencil slowly across your face. You will feel the touch of the pencil continuously, but
the lead will feel cold only occasionally—because touch and cold are monitored by
different receptors. Although their range is limited to the surface of our body, changes
there are often due to external stimulation, so the skin senses inform us about both our
body and the world. (We experience these sensations deeper in the body as well, but
less often and with less sensitivity.)
There are two general types of receptors. Free nerve endings are simply
processes at the ends of neuronal dendrites; they detect warmth, cold, and pain. All
the other receptors are encapsulated receptors, which are more complex structures
enclosed in a membrane; their role is to detect touch. Why are there so many receptors
just for touch? Because touch is a complex sense that conveys several types of
information. In the superficial layers of the skin, Meissner’s corpuscles respond with a
brief burst of impulses, while Merkel’s disks give a more sustained response. Located
near the surface of the skin as they are, they detect the texture and fine detail of
objects. They also detect movement and come into play when you explore an object
with gentle strokes of your hand or when a blind person reads Braille. Pacinian
corpuscles and Ruffini endings are in the deeper layers, where they detect stretching
of the skin and contribute to our perception of the shape of grasped objects.
The other three skin senses are detected by free nerve endings, but this
statement is a bit misleading; those nerve endings have distinctly different receptors
that make the neurons stimulus specific (Basbaum, Bautista, Scherrer, & Julius,
2009). We can respond to a wide range of temperatures; at least two different
receptors detect different levels of warmth, and another receptor responds to cooling
of the skin. These receptors are all members of the transient receptor potential (TRP)
family of protein ion channels. Detection of pain also requires several receptors,
mostly because of the variety of pain sources; these sources are categorized as
thermal, chemical, and mechanical. Two TRP receptors respond to painful heat; a
receptor for painful cold has not been conclusively identified, but the coolness
receptor does not account for this sensation. Chemical receptors react to a wide range
of chemical irritants. Best known is the TRPV1 heat pain receptor, which also
responds to capsaicin, the ingredient in chili peppers that makes spicy foods painfully
hot. Ointments containing capsaicin alleviate the joint pain of arthritis, apparently
because continued stimulation of the receptors depletes the neuropeptide that the
neurons use to signal pain. This receptor also responds to the pain-inducing acid
released in bone cancer, and a TRPV1 antagonist relieves this pain (Ghilardi et al.,
2005). The TRPM8 coolness receptor produces the cool sensation of mint in
toothpaste and candies, as well as the cooling effect of menthol and camphor on the
skin; menthol and camphor creams are useful for treating muscle pain, skin irritations,
and canker sores in the mouth. TRPA1 receptors are responsible for the painful
irritation caused by vehicle exhaust, tobacco smoke, hydrogen peroxide, and tear gas,
and they account for the pungency of mustard, garlic, and wasabi, as well as the tingle
you get from a carbonated drink. As we will see shortly, the body produces its own
irritants when tissues are damaged; these continue to produce pain well after the
stimulus is past. The receptors for mechanical pain have not been determined, which
is unfortunate because persistent hypersensitivity to touch is a major problem
following tissue or nerve injury.
The semicircular canals and the utricle and saccule also send projections to the
cerebellum and the brain stem, and there is also a pathway to the cortex, specifically
to an area called the parieto-insular-vestibular (PIV) cortex. This is the likely location
where excessive eye movements, from reading in a moving car, for example, cause
dizziness and nausea. The PIV receives information from smell and taste receptors as
well, which is why nausea and disgust are linked. The same thing happens with
excessive body motion, for example, during a rough boat ride or from spinning
around quickly. Doing the latter eventually causes the vestibular fluid to move with
the head (like stirring a glass of water with a straw; at first the water resists
movement, but eventually the water moves with the straw). When that happens and
you stop, the fluid keeps moving, which makes the cupula continue to bend and
makes you feel like you’re spinning in the opposite direction. Figure skaters and
gymnasts train themselves to focus on a single point when spinning and then turn their
heads quickly to maintain that focus, which keeps the fluid from spinning. If you
easily get motion or seasickness, chances are that you have a more sensitive PIV
cortex. Taking an antihistamine drug (such as Dramamine) can decrease PIV
activation and therefore reduce nausea.
Pain begins when certain free nerve endings are stimulated by intense pressure
or temperature, by tissue damage, or by various chemicals. Tissue injury also causes
cells to release a wide array of signaling molecules, referred to as the inflammatory
soup; these include histamine, proteins (bradykinin and substance P), lipids
(prostaglandins), neurotransmitters (serotonin), and cytokines (Julius & Basbaum,
2001; Kidd & Urban, 2001). Some of these stimulate pain receptors, but they also
produce the familiar swelling and redness of inflammation, and they enhance
excitability of the pain neurons so much that the neurons respond even to light touch.
This effect is adaptive because it encourages guarding of the injured area, but the
resulting pain can be more troublesome than the original injury. From the injured area,
pain information travels to the spinal cord over large, myelinated A-delta fibers and
small, lightly myelinated or unmyelinated C fibers. Because A-delta fibers transmit
more rapidly than C fibers, you notice a sharp, stinging pain almost immediately
when you are injured, followed by a longer-lasting dull, aching pain (Basbaum &
Jessell, 2013). Sharp pain receptors tend to be superficially located and are much
more densely packed, which makes localizing the source of the sharp pain easier. Dull
pain receptors are deeper and respond to a wider array of painful stimuli, and the
sensations are harder to localize (Raja, Meyer, Ringkamp, & Campbell, 1999). Sharp
pain makes a good danger signal and motivates you to take quick action, while dull
pain hangs around for a longer time to remind you that you have been injured.
The most frequently used pain drugs are aspirin, ibuprofen, naproxen, and
acetaminophen. The first three (also called nonsteroidal anti-inflammatory drugs, or
NSAIDs) block inflammatory enzymes required for producing prostaglandins, so they
primarily reduce swelling and peripheral pain. Acetaminophen weakly blocks the
same enzymes, so it has little anti-inflammatory benefit; its major effect is in the
central nervous system. Ameliorating intense pain often requires more powerful
substances called opiates (morphine extracted from the opium poppy is the
acknowledged gold standard), but their addictiveness and patients’ rapidly developing
tolerance have spurred the development of numerous alternatives. The new MDAN
series of drugs, for example, targets the mu opioid receptor while blocking the delta
opioid receptor; these drugs are reportedly 50 times more potent than morphine,
without producing either tolerance or addiction (Dietis et al., 2009). While promising,
no mu opioid receptor drug has yet to make it to market; the promising drug
oxycodegol (NKTR-0181) program was cancelled after a unanimous negative vote by
the U.S. Food and Drug Administration (FDA), even after being fast tracked for
testing and completing a successful Phase 3 clinical trial (Anson, 2020).
Efforts are underway on a variety of other fronts as well. Tanezumab, an
antibody for nerve growth factor, has shown safety and effectiveness in clinical trials
with chronic back pain and the inflammatory pain of joint arthritis. An experimental
drug, NEO6860, that blocks the TRPV1 pain receptor produces modest reduction of
knee pain in human participants (Arsenault et al., 2018). Some antidepressants
(tricyclics and serotonin-norepinephrine reuptake inhibitors) also appear to be
successful in alleviating chronic pain and pain that has an emotional component (D.
M. Marks et al., 2009). In addition, cannabinoids appear to interact and suppress pain
messages throughout the central nervous system (Bettinger & Chu, 2020). There has
also been some preliminary success with gene therapy; patients whose pain from
cancer was not relieved by 200 mg per day of morphine experienced an 80%
reduction following treatment with a gene that increases endorphin production (D. J.
Fink et al., 2011). The fact that pain ultimately occurs in the brain has inspired a novel
approach that is also showing promise; chronic pain patients given continuous
functional magnetic resonance imaging (fMRI) feedback of activity in their cingulate
gyrus learned to reduce pain-related brain activity and their experience of pain
(deCharms et al., 2005).
During a spring break in New Orleans, Bob was touring one of the Civil War–
era plantation houses for which the area is noted. In a glass case was an assortment of
artifacts that had been found on the plantation grounds. An odd part of the collection
was a few lead rifle slugs with what were obviously deep tooth marks on them. When
makeshift surgery had to be performed with only a large dose of whiskey for
anesthesia, the unfortunate patient would often be given something to bite down on,
such as a piece of leather harness or a relatively soft lead bullet. (You can probably
guess the common expression that is associated with this practice.) As a toddler, when
you scraped your knee, you clenched your teeth and rubbed the area around the
wound. And—tribute to your childhood wisdom—it really did help and got you
through the pain without the benefit of either a lead bullet or whiskey. You might
think that tooth clenching and rubbing simply take attention from the pain. Ronald
Melzack and Patrick Wall (1965) had another idea. In their gate control theory, they
hypothesized that pressure signals arriving in the brain trigger an inhibitory message
that travels back down the spinal cord, where it closes a neural “gate” in the pain
pathway.
Research has confirmed the general idea of their theory, and we now know
that a descending pathway in the spinal cord is one of the ways the brain uses
endorphins to control pain. Pain causes the release of endorphins in the periaqueductal
gray (PAG), a brain stem structure surrounding the cerebral ventricles with a large
number of endorphin synapses. Activation of the endorphin circuit apparently has
multiple neural origins, including the cingulate cortex during placebo analgesia and
the amygdala in the case of fear-induced analgesia (Petrovic, 2005). Brain imaging
shows that placebo reduces pain through this circuit, and so does simple distraction.
Women, unfortunately, have fewer mu opioid receptors in the PAG than men, so they
receive less pain-relieving benefit from opiate drugs (Loyd, Wang, & Murphy, 2008).
The PAG also contains cannabinoid receptors, which respond to endogenous
cannabinoids and the active ingredient in marijuana.
f. Movement
A popular view of the brain is that it is mostly preoccupied with higher
cognitive processes, such as thinking, learning, and language. However, a surprising
proportion of the brain is devoted to planning and executing movements. We are
talking about more than simply moving the body from one place to another; consider
the surgeon’s coordinated hand movements during a delicate operation, the control of
mouth and throat muscles and diaphragm required to sing an aria, or a baseball
outfielder’s ability to track a fast-moving fly ball and arrive in time to catch it. Studies
of the control of movement provided one of the earliest windows into the brain’s
organization and functioning, and it is that facet of the research that will interest us
most. Before we launch into that topic, we need some understanding of the equipment
the brain has to work with.
The ones you are most familiar with are the skeletal muscles, which move the
body and limbs but can fatigue if overused; they are also called striated muscles
because of their striped cellular appearance. Smooth muscles produce rhythmic
contractions in the internal organs; for example, they move food through the digestive
system, constrict blood vessels, and void the bladder. Cardiac muscles are the non-
fatiguing muscles that make up the heart. Because our focus is on movement, we will
concentrate on the skeletal muscles. But despite differences in appearance, the
muscles function similarly. Like other tissues of the body, a muscle is made up of
many individual cells, or muscle fibers. The muscle cells are controlled by motor
neurons that synapse with a muscle cell at the neuromuscular junction using the
neurotransmitter acetylcholine. The number of cells served by a single axon
determines the precision of movement possible. The biceps muscles have about 100
muscle fibers per axon, but the ratio is around 3 to 1 in the eye muscles, which must
make very precise movements in tracking objects (Evarts, 1979).
Skeletal muscles are anchored to bones across a joint by tendons, which are
tough fibrous bands of connective tissue. Muscles can only contract; a limb is both
flexed and extended by pulling on the joint from one side or the other. You can see
that limbs have antagonistic muscles that produce opposite movements at a joint. In
this case, the biceps muscle decreases the arm angle to flex the arm, and the triceps
increases the angle, extending the arm. Rather than one muscle relaxing while the
other does all the work, movement involves opposing contraction from both muscles.
The simultaneous contraction of antagonistic muscles creates a smoother movement,
allows precise stopping, and maintains the limb angle with minimal tremor. Standing
requires complex coordination of antagonistic muscles in the legs as well as muscles
in the torso. The strength and duration of muscle contraction varies from moment to
moment, so the balance between antagonistic muscles is constantly shifting,
constantly correcting. If maintaining the balance between opposed pairs of muscles
required conscious, voluntary activity, we would find it very difficult to hold a camera
still enough to get a sharp picture or even to stand or sit erect. Adjustments this fast
must be controlled by reflexes at the level of the spinal cord.
More complex patterns of motor behavior are also controlled in the spinal
cord. It has been known for at least a century that cats with severed spinal cords,
eliminating control from the brain, can make rhythmic stepping movements and walk
when they are suspended with their feet on a treadmill (J. G. Jones, Tansey, & Stuart,
2011). This behavior depends on central pattern generators (CPGs), neuronal
networks that produce a rhythmic pattern of motor activity, such as those involved in
walking, swimming, flying, and breathing. CPGs are in the spinal cord and in the
brain and have been found in all vertebrate and in many invertebrate animals. In
humans, they are most obvious in infants below the age of 1 year, who can make
stepping movements when held with their feet on a treadmill (Lamb & Yang, 2000).
In adults, CPGs provide an important bit of automaticity to routine movements. They
can be elicited in individuals with spinal cord injury to produce rhythmic stepping
movements. Spinal reflexes produce quick, reliable responses, and CPGs provide
basic routines the brain can call up when needed, freeing the brain for more important
matters. But reflexes and CPGs cannot provide all our movement capabilities, so we
will turn our attention to the contributions the brain makes to movement.
You already know two functions of the prefrontal cortex that suit it for its role
in movement control: First, it plans actions with regard to their consequences; second,
it receives information from the ventral visual stream about object identity, which is
useful in identifying targets of motor activity. As an initial step in motor planning, the
prefrontal cortex integrates sensory information about the world with information
about the body (from the posterior parietal cortex); it then holds the information in
memory while selecting the appropriate movement and its target. Considering its
activities, it really makes more sense to say that the role of the prefrontal cortex is not
so much in planning movements as in planning for movements. These functions are
typically investigated in monkeys while they perform some variation of a delayed
match-to-sample task. The monkey is presented with a visual stimulus; then, after a
delay of a few seconds in which the stimulus is absent, the monkey is presented two
or more stimuli and required to select the original stimulus (by reaching for it) to
obtain a reward, such as a sip of juice. Some cells in the prefrontal cortex start firing
when the first stimulus is presented and continue to fire throughout the delay,
suggesting that they are “remembering” the stimulus. At response time, another group
of prefrontal cells starts firing before activity starts in the premotor areas; this
indicates that the prefrontal cortex selects the target of behavior and the appropriate
motor responses.
The premotor cortex begins programming a movement by combining
information from the prefrontal cortex and the posterior parietal cortex (Kalaska &
Rizzolatti, 2013). A good example comes from a study in which monkeys were cued
to reach for one of two targets, A or B, in different locations and to use the left arm on
some trials and the right arm on others. Some premotor neurons increased their firing
rate only if target A was cued, and other neurons were selective for target B. Other
cells fired selectively depending on which arm was to be used. Still other cells
combined the information of the first two kinds of cells; they increased their firing
only when a target was cued and a particular arm was to be used (Hoshi & Tanji,
2000). Two other cell types combine visual and somatosensory information to provide
visual guidance of reaching and object manipulation.
Output from the prefrontal cortex flows to the supplementary motor area,
which assembles sequences of movements, such as those involved in eating or in
playing the piano. In monkeys trained to produce several different sets of movement
sequences, different neurons increase their firing during a delay period depending on
which sequence has been cued for performance (Shima & Tanji, 2000; Tanji & Shima,
1994). An important form of movement sequencing is the coordination of movements
between the two sides of the body. For example, when a monkey’s supplementary
motor cortex is damaged in one hemisphere, its hands tend to duplicate each other’s
actions instead of sharing the task. Coordinating the actions of the limbs is one of the
biggest problems to be overcome on the way to developing mechanized prostheses, as
A Further Look explains.
The basal ganglia and cerebellum produce no motor acts themselves. Rather,
they modulate the activity of cortical and brain stem motor systems; in that role, they
are necessary for posture and smooth movement (Wolpert, Pearson, & Ghez, 2013).
The basal ganglia—the caudate nucleus, putamen, globus pallidus, and the substantia
nigra—use information from the primary and secondary motor areas and the
somatosensory cortex to integrate and smooth movements. The basal ganglia send
output directly to the primary motor cortex and supplementary motor area and to the
premotor cortex via the thalamus. The basal ganglia also are especially active during
complex sequences of movements (Boecker et al., 1998). It appears that they are
involved in learning movement sequences so that the movements can be performed as
a unit (Graybiel, 1998). In fact, one of the symptoms of Parkinson’s disease, which is
caused by degeneration in the basal ganglia, is impaired learning, whether motor
behavior is involved or not (Knowlton, Mangels, & Squire, 1996). Malfunction in the
basal ganglia results in postural abnormalities and involuntary movements in
Parkinson’s disease and Huntington’s disease, which can be treated using deep brain
stimulation.
When the cerebellum receives information from the motor cortex about an
intended movement, it determines the order of muscular contractions and their precise
timing. It also uses information from the vestibular system to maintain posture and
balance, refine movements, and control eye movements that compensate for head
movements (Lisberger & Thach, 2013). Once an intended movement has been
modified, the cerebellum sends the information back to the primary motor cortex. We
can see the contribution of the cerebellum in the deficits that occur when it is
damaged. For example, we begin to shape our hand for grasping while the arm is
moving toward the target, but a person with cerebellar damage reaches, pauses, and
then shapes the hand. A normal individual touches the nose in what appears to be a
single, smooth movement; cerebellar damage results in exaggerated, wavering
corrections. The effects of cerebellar damage on coordination and accuracy in limb
movements is like the effect of alcohol; the drunk driver who is pulled over by the
police has trouble walking a straight line, standing on one foot with the eyes closed,
or touching the nose with the tip of the finger. People with damage to the cerebellum
are often mistakenly believed to be drunk. The cerebellum lives up to the meaning of
its name, “little brain, ” by applying its expertise to a variety of tasks. It is necessary
for learning motor skills (D. A. McCormick & Thompson, 1984), but it also
participates in nonmotor learning (Canavan, Sprengelmeyer, Diener, & Hömberg,
1994) and in making time and speed judgments about auditory and visual stimuli
(Keele & Ivry, 1990). Also, patients with cerebellar damage have difficulty shifting
visual attention to another location in space (whether this involves eye movements or
not), taking 0.8 to 1.2 seconds, compared with 0.1 second for normal individuals
(Townsend et al., 1999). We should think of the cerebellum in terms of its general
functions, rather than strictly as a motor organ.
