Eye, Brain, and Vision

INTRODUCTION

Intuition tells us that the brain is complicated. We do complicated things, in
immense variety. We breathe, cough, sneeze, vomit, mate, swallow, and uri-
nate; we add and subtract, speak, and even argue, write, sing, and compose
quartets, poems, novels, and plays; we play baseball and musical instruments.
We perceive and think. How could the organ responsible for doing all that not
be complex?
We would expect an organ with such abilities to have a complex structure.
At the very least, we would expect it to be made up of a large number of
elements. That alone, however, is not enough to guarantee complexity. The
brain contains 10ˆ12 (one million million) cells, an astronomical number by any
standard. I do not know whether anyone has ever counted the cells in a human
liver, but I would be surprised if it had fewer cells than our brain. Yet no one
has ever argued that a liver is as complicated as a brain.
We can see better evidence for the brain's complexity in the interconnections
between its cells. A typical nerve cell in the brain receives information from
hundreds or thousands of other nerve cells and in turn transmits information
to hundreds or thousands of other cells. The total number of interconnections
in the brain should therefore be somewhere around 10ˆ14 to 10ˆ15, a larger num-
ber, to be sure, but still not a reliable index of complexity. Anatomical com-
plexity is a matter not just of numbers; more important is intricacy of organi-
zation, something that is hard to quantify. One can draw analogies between
the brain and a gigantic pipe organ, printing press, telephone exchange, or
large computer, but the usefulness of doing so is mainly in conveying the
image of a large number of small parts arranged in precise order, whose func-
tions, separately or together, the nonexpert does not grasp. In fact, such analo-
gies work best if we happen not to have any idea how printing presses and
telephone exchanges work. In the end, to get a feeling for what the brain is and
how it is organized and handles information, there is no substitute for examin-
ing it, or parts of it, in detail. My hope in this book is to convey some flavor of
the brain's structure and function by taking a close look at the part of it con-
cerned with vision.
The questions that I will be addressing can be simply stated. When we look
at the outside world, the primary event is that light is focused on an array of
125 million receptors in the retina of each eye. The receptors, called rods and
cones, are nerve cells specialized to emit electrical signals when light hits them.
The task of the rest of the retina and of the brain proper is to make sense of
these signals, to extract information that is biologically useful to us. The result
is the scene as we perceive it, with all its intricacy of form, depth, movement,
color, and texture. We want to know how the brain accomplishes this feat.
Before I get your hopes and expectations too high I should warn you that
we know only a small part of the answer. We do know a lot about the machin-
ery of the visual system, and we have a fair idea how the brain sets about the
task. What we know is enough to convince anyone that the brain, though
complicated, works in a way that will probably someday be understood—and
that the answers will not be so complicated that they can be understood only
by people with degrees in computer science or particle physics.
Today we have a fairly satisfactory understanding of most organs of our
body. We know reasonably well the functions of our bones, our digestive
tubes, our kidneys and liver. Not that everything is known about any of
these—but at least we have rough ideas: that digestive tubes deal with food,
the heart pumps blood, bones support us, and some bones make blood. (It
would be hard to imagine a time, even in the dark twelfth century, when it
was not appreciated that bones are what make our consistency different from
that of an earthworm, but we can easily forget that it took a genius like
William Harvey to discover what the heart does.) What something is for is a
question that applies only to biology, in the broad sense of the word "biol-
ogy." We can ask meaningfully what a rib is for: it supports the chest and
keeps it hollow. We can ask what a bridge is for: it lets humans cross a river—
and humans, which are part of biology, invented the bridge. Purpose has no
meaning outside of biology, so that I laugh when my son asks me, "Daddy,
what's snow for?" How purpose comes into biology has to do with evolution,
survival, sociobiology, selfish genes—any number of exalted topics that keep
many people busy full time. Most things in anatomy—to return to solid
ground—even such erstwhile mysterious structures as the thymus gland and
the spleen, can now have quite reasonable functions assigned to them. When I
was a medical student, the thymus and spleen were question marks.
The brain is different. Even today large parts of it are question marks, not
only in terms of how they work but also in terms of their biological purpose.
A huge, rich subject, neuroanatomy consists largely of a sort of geography of
structures, whose functions are still a partial or complete mystery. Our igno-
rance of these regions is of course graded. For example, we know a fair
amount about the region of brain called the motor cortex and have a rough
idea of its function: it subserves voluntary movement; destroy it on one side
and the hand and face and leg on the opposite side become clumsy and weak.
Our knowledge of the motor cortex lies midway along a continuum of rela-
tive knowledge that ranges all the way from utter ignorance of the functions of
some brain structures to incisive understanding of a few—like the understand-
ing we have of the functions of a computer, printing press, internal combus-
tion engine, or anything else we invented ourselves.
The visual pathway, in particular the primary visual cortex, or striate cortex,
lies near the bone or heart end of this continuum. The visual cortex is perhaps
the best-understood part of the brain today and is certainly the best-known
part of the cerebral cortex. We know reasonably well what it is "for", which is
to say that we know what its nerve cells are doing most of the time in a
person's everyday life and roughly what it contributes to the analysis of the
visual information. This state of knowledge is quite recent, and I can well
remember, in the 1950s, looking at a microscopic slide of visual cortex, show-
ing the millions of cells packed like eggs in a crate, and wondering what they
all could conceivably be doing, and whether one would ever be able to find
out.
How should we set about finding out? Our first thought might be that a
detailed understanding of the connections, from the eye to the brain and
within the brain, should be enough to allow us to deduce how it works.
Unfortunately, that is only true to a limited extent. The regions of cortex at
the back of the human brain were long known to be important for vision
partly because around the turn of the century the eyes were discovered to make
connections, through an intermediate way station, to this part of the brain.
But to deduce from the structure alone what the cells in the visual cortex are
doing when an animal or person looks at the sky or a tree would require a
knowledge of anatomy far exceeding what we possess even now. And we
would have trouble even if we did have a complete circuit diagram, just as we
would if we tried to understand a computer or radar set from their circuit
diagrams alone—especially if we did not know what the computer or radar set
was for.
Our increasing knowledge of the working of the visual cortex has come
from a combination of strategies. Even in the late 1950s, the physiological
method of recording from single cells was starting to tell us roughly what the
cells were doing in the daily life of an animal, at a time when little progress was
being made in the detailed wiring diagram. In the past few decades both fields,
physiology and anatomy, have gone ahead in parallel, each borrowing tech-
niques and using new information from the other.
1 have sometimes heard it said that the nervous system consists of huge
numbers of random connections. Although its orderliness is indeed not always
obvious, I nevertheless suspect that those who speak of random networks in
the nervous system are not constrained by any previous exposure to neuroa-
natomy. Even a glance at a book such as Cajal's Histologie du Systeme Nerveux
should be enough to convince anyone that the enormous complexity of the
nervous system is almost always accompanied by a compelling degree of or-
derliness. When we look at the orderly arrays of cells in the brain, the impres-
sion is the same as when we look at a telephone exchange, a printing press, or
the inside of a TV set—that the orderliness surely serves some purpose. When
confronted with a human invention, we have little doubt that the whole ma-
chine and its separate parts have understandable functions. To understand
them we need only read a set of instructions. In biology we develop a similar
faith in the functional validity and even ultimately in the understandability of
structures that were not invented, but were perfected through millions of years
of evolution. The problem of the neurobiologist (to be sure, not the only
problem) is to learn how the order and complexity relate to the function.
To begin, I want to give you a simplified view of what the nervous system is
like—how it is built up, the way it works, and how we go about studying it.
I will describe typical nerve cells and the structures that are built from them.
The main building blocks of the brain are the nerve cells. They are not the
only cells in the nervous system: a list of all the elements that make up the
brain would also include glial cells, which hold it together and probably also
help nourish it and remove waste products; blood vessels and the cells that
they are made of; various membranes that cover the brain; and I suppose even
the skull, which houses and protects it. Here I will discuss only the nerve cells.
Many people think of nerves as analogous to thin, threadlike wires along
which electrical signals run. But the nerve fiber is only one of many parts ot
the nerve cell, or neuron. The cell body has the usual globular shape we associate
with most cells (see diagram on this page) and contains a nucleus, mitochon-
dria, and the other organelles that take care of the many housekeeping func-
tions that cell biologists love to talk about. From the cell body comes the main
cylinder-shaped, signal-transmitting nerve fiber, called the axon. Besides the
axon, a number of other branching and tapering fibers come off the cell body: