It's a Jungle in There: How Competition and Cooperation in the Brain Shape the Mind (7 page)

At this point, I could ask you to join me on a more detailed tour of the brain, considering what the many parts of the brain do—what functions, in other words, those brain regions seem to carry out with regard to cognition, perception, action, and emotion. We could move from place to place, noting the apparent specialization of each locale. The review would show that things change gradually as we move from one neural neighborhood to the next.

I will refrain from providing such a tour, however, because we could lose sight of the larger principle I want to emphasize. It suffices to say, in my opinion, that there are three aspects of brain specialization that bear on the jungle principle. These pertain to the methods used to infer the specialization, the claim that brain functions are localized, and the issue of whether the brain is hard- or soft-wired.
¹¹
I’ll take up each of these issues in this and the next two sections.

Regarding method, it pays to consider how the functional properties of different brain regions are discerned. One method—perhaps the most straightforward—is to ask what happens when a part of the brain is damaged. The logic is straightforward. If damage to some brain region disrupts some activity, then that region can be said to play some role in that activity.

Sometimes people go beyond this modest inference and say that if damage to a part of the brain impairs an activity, then that part of the brain
controls
that activity or is
necessary
for that activity. Such a conclusion is premature, however, and here’s an anecdote that shows why.

Suppose you’re a songwriter and, like many aspiring tune-and-lyric creators, you write songs at home. One day, something annoying happens. The sewer line running from your house backs up and your basement starts to fill with the most foul-smelling ooze. I think of this disgusting example because this happened at my house one day while I was writing this book. I couldn’t get the experience out of my mind and thought of it as I started to write this section.

If your sewer line backs up while you’re trying to compose music, your writing suffers. In fact, your composing comes to a screeching halt, not because your sewer line is strictly necessary for your music generation, but because the unexpected plumbing problem interferes with your ability to concentrate on your art. This homely example shows how careful one must be about drawing causal conclusions from neural damage. If neural damage disrupts an activity, it doesn’t follow that the activity depends in a direct way on the region that’s impaired.

Now consider another possible outcome—that damage to some area of the brain does
not
impair or affect a function. What can you conclude from that outcome? You have to be careful here as well. You’d be incorrect to say that the lack of an effect following damage to that brain area implies that the brain area plays no role in the function. The brain area
might
play a role, but there might also be a backup system that takes over if the area takes a hit.

These cautions aside, it’s unquestionable that a vast amount has been learned about the brain and nervous system by studying the effects of damage to its components. Such damage can result from accidents such as bullet wounds, interruptions to blood flow, or physiological mayhem wrought by cancer or infection. Interruptions of brain activity can also come about by deactivating parts of the brain through temporary (reversible) freezing, a technique used in many laboratories.
¹²

Other methods also exist for inferring the role of brain systems in psychological functions. These fall into two broad classes:
stimulation
and
recording
. Neurophysiologists stimulate the brain and other parts of the nervous system with electrodes and then observe the consequences of the stimulation. When stimulation of a brain site gives rise to some effect, it’s possible to infer that the brain region plays some role in the observed function. Saying that the region is
necessary
for the function is too large a leap, however, as already indicated.

It’s also possible to
record
from the brain. Doing so can be achieved with tiny electrodes that pick up the electrical activity of individual neurons or
small sets of neurons. Alternatively, it’s possible to record from the brain as a whole or section by section, picking up larger swaths of activity. Within this class of recording methods, there are technologically advanced techniques known by an alphabet-soup’s worth of acronyms: EEG, MEG, ERP, PET, CAT, MRI, fMRI, DOI. I won’t review the methods here; doing so would take us far afield.
¹³
Suffice it to say that the methods have supported the view that different parts of the brain serve different functions, a principle known as localization.

One of the most famous sources of evidence for localization of function in the brain is the discovery of feature detectors in the visual cortex. These neurons were found by David Hubel and Torsten Wiesel in the early 1960s.
¹⁴
While recording from cells in this region of the cat’s brain, Hubel and Wiesel noticed that when particular visual stimuli were shown, particular cells fired. For example, a given cell emitted a burst of action potentials when a dark bar was shown. The cell fired most when the bar was oriented at 90 degrees, but the more the bar’s orientation shifted from 90 (or 270) degrees, the less vigorously the cell fired. Another cell fired the most when the same dark bar was oriented at 45 (or 135) degrees. Again, the more the bar’s orientation departed from that angle, the less vigorously the cell fired. Similar distinctions applied to other cells. The stimuli that had these effects were of various kinds. They could be bars moving along paths with particular orientations, blobs in different parts of space, and so on.

Hubel and Wiesel referred to these neurons as
feature detectors
. The two scientists also distinguished between detectors that differed with respect to the complexity of the stimuli to which the detectors responded. “Simple” cells responded to elementary feature combinations. “Complex” cells responded to more complicated feature combinations. “Hyper-complex” cells responded to still more complicated feature combinations.

Analogous cells for other sensory modalities were found as well. Other researchers found neurons that respond preferentially to sounds of different frequencies, to odors of different chemical compositions, to touches on different parts of the body, to an animal’s being in different spatial locations (so-called “place cells”), and so on.
¹⁵

Do these results imply that psychological functions are localized in the brain? The answer is no, and the reason is that detecting an aspect of a stimulus isn’t the same as experiencing it in all its aspects. In addition and no less
importantly, none of the neurons lives alone. Each does what it does by virtue of where it lives—that is, by virtue of the other neurons with which it communicates most directly. Finally, there are limits on how far one can go with the feature detector concept. If you have detectors for faces, for example, you may or may not have a detector for your grandmother’s face.

The search for “grandmother” cells has so far failed, presumably because grandmothers’ faces appear in infinitely many poses. Even if a grandmother cell were found, you’d be left wondering whether there would be a specialized cell for the sight of grandma sitting at the piano with a red tulip in the green vase atop the keyboard, a different specialized cell for the sight of grandma sitting at the piano with a
blue
tulip in the green vase atop the keyboard, and so on.

The problem with high-level detectors is that they can’t keep up with the endless combinations of experience. For any assortment of features, some other assortment can be imagined. It’s hard to believe that every possible scene can be recognized only through activation of some particular specialized cell whose
raison d’être
is recognition of just that environment. All possible stimuli can’t be anticipated, so all possible detectors can’t be pre-formed. By having relatively low-level features that can combine in novel ways, however, you have the basis for perceiving endlessly varied inputs. Low-level feature detectors have a good chance of surviving because they’re called on regularly. High-level features—or ensembles of features that correspond to more complex, varied arrangements—can survive as well but take longer to form and may have a more tenuous future.

Another aspect of neurophysiology that points to the plausibility of the inner jungle principle is
neural plasticity
. To understand the idea of neural plasticity, it’s useful to think of a convenient but now outdated fiction about the brain. The fiction is that every region of the brain serves a fixed, hard-wired function. According to this view, cells for feeling are faithful to touch all their lives, cells for seeing are loyal to looking from cradle to grave, and so on. At a more microscopic level, the particular sensory features that individual cells are tuned to—lines of some orientation for seeing, sounds of some frequency for hearing—remain the same forever. Like sailors whose girlfriends’ names remain tattooed to their arms forever, neurons stay faithful to their first functional flames. That, anyway, was the belief associated with the hard-wired view.

It turns out that the functional properties of different brain regions are not engraved in stone but instead are malleable, or
plastic
. The term “plastic” conjures up images of squeeze bottles. More generally, though, plastic means reshapable. The brain is plastic in the sense that its functions are not hard-wired, but instead can be recast through experience.
¹⁶

Brain plasticity is central to the theme of this book. In fact, it was when I was teaching about neural plasticity that I first uttered the phrase, “It’s a jungle in there.” Neural plasticity illustrates the jungle principle vividly—perhaps more vividly than any other phenomenon in neuroscience.

What exactly is neural plasticity and how was it discovered? I’ll begin with the second question.

In the 1980s, a young neuroscientist named Michael Merzenich was puzzled by observations he made in his neurophysiological recordings of squirrel monkey brains.
¹⁷
From his training, Merzenich had a pretty good idea of the functions served by the brain structures he was studying. He knew that vision is served by one region, that hearing is served by another, that touch is served by still another, and so on.
¹⁸
He also knew that adjacent regions within each of these brain regions tend to have similar functions. Within the somatosensory cortex, for example, some neurons fire in response to touch on the left index finger, adjacent neurons fire in response to touch on the left middle finger, and so on.
¹⁹

Another principle that Merzenich was aware of was that some neural enclaves are bigger than others. Neurons turned on by tongue touches are more plentiful than neurons turned on by foot feels, neurons activated by digit dabs are more numerous than neurons aroused by belly brushes, and so on. The thumb’s representation in this part of the brain is especially large (
Figure 2
). What Merzenich expected was that the sizes of the brain regions from which he recorded would remain more or less the same in the brains of all the monkeys he studied.

To his surprise, Merzenich and his coworkers found that brains of different monkeys varied considerably in how much of their brains were devoted to the same functions. Though the animals were roughly the same age and size, the sizes of their brain regions devoted to any given function varied considerably.

What was going on here, Merzenich and his coworkers wondered? One possibility was that different animals were genetically disposed to have differently sized brain regions. According to this view, one animal, through inheritance, had lots of brain space for touch on its
index
finger; another animal, owing to
its
genetic constitution, had more neural real estate devoted to touches on its
middle
finger, and so on. According to this nature-rather-than-nurture view,
what determined the amount of brain space for a function was genetic determination. According to another view, the nurture-as-well-as-nature view, the differences in the sizes of the brain zones reflected experience as well as genes.

FIGURE 2.
Cartoon of the amount of space (number of neurons) responsive to touch on different parts of the body.

To decide between these hypotheses, Merzenich and his colleagues performed experiments in which they altered the experiences of their experimental subjects. Their logic was straightforward: If the organization of the brain is fixed, experience shouldn’t change it.

In one study, Merzenich et al. amputated the middle finger of a monkey to see how that would affect the monkey’s somatosensory cortexes. This manipulation provided a particularly strong test of the experience hypothesis and so was deemed ethically acceptable.