It's a Jungle in There: How Competition and Cooperation in the Brain Shape the Mind (8 page)

Merzenich and his colleagues found that cells in the somatosensory cortex that had previously responded to touch on the middle finger but not to touch on the adjacent ring or index finger became responsive to touch on the still-present adjacent fingers after the middle finger was removed. In other words, the middle-finger region of the somatosensory cortex, which had been unresponsive to touch on the ring or index finger, became responsive to touch on the ring and index digits after the middle finger was removed.

How should you interpret this result? The best interpretation, I believe, is that it’s a jungle in there. A neuron engaged in transmitting signals from the
index
finger to the brain, or a neuron engaged in transmitting signals from the
ring
finger to the brain, can be said to have always sought to make contact with the middle-finger region. That region was next to the place where those
neurons mainly projected. But something kept those neurons from extending their tendrils to the middle-finger zone. Keeping them out were the neural tendrils from the middle finger. After surgical removal of the middle finger, afferents to the middle-finger region of the somatosensory cortex no longer delivered signals to that cortical zone, so index-finger inputs and ring-finger inputs had less competition for entry to the middle-finger area. Once they made that entry and could do so on repeated occasions, they could form reliable connections to it.

Do such changes occur only after unfortunate events like amputations? The answer, happily, is no. Enlargements of neural territories also accompany practice. Merzenich and his colleagues demonstrated practice effects in other, more benign experiments. In one, they gave monkeys practice on a task requiring fine tactile discrimination by the middle finger. After prolonged practice on this task, more brain tissue in the somatosensory cortex became responsive to touch on that finger. Other neighboring regions that were previously responsive to touch from the adjacent fingers
lost
neural space in the process as they became responsive to stimulation of the highly practiced middle finger. One zone’s gain was, therefore, another zone’s loss.
²⁰

Neural plasticity has also been found for modalities other than touch. People who go blind have been found to develop sound sensitivity in brain regions that would have otherwise been devoted to vision.
²¹
The neurons in the so-called visual cortex that have this crossover property lie next to the primary auditory cortex, consistent with the view that neural plasticity mainly reflects infiltration by neural neighbors.

Does neural plasticity have consequences for experience, or is it just something picked up by neurophysiologists in laboratories—a curious phenomenon, perhaps, but not of any practical value? Many findings indicate that the malleability of neural representations affects experience outside the lab. Some of these findings relate to the fact that when more brain area comes to be devoted to a function, better perception or performance results. For perception, this translates to better detection or discrimination. For performance, it translates to finer coordination or dexterity.

Consider the super-sized thumb of the sensory homunculus, shown in
Figure 2
. The thumb is a very sensitive area. When you try to determine whether someone is touching you with one toothpick or two, your discrimination threshold is small on your thumb. You’re very sensitive there and can
pick up a small difference. Only one other part of the body is more sensitive than the thumb when it’s subjected to two-point discrimination tests: the tongue.

Sensitivity to touch as measured through two-point discrimination is related to how much brain tissue is devoted to that body part. The tongue has lots of brain representation and so does the thumb. But the abdomen, whose two-point discrimination threshold is much larger, is much less “brainy,” with much less neural territory devoted to the abdomen on a per-square-centimeter basis than to the tongue or thumb. The thigh and calf are likewise scantily represented in the somatosensory cortex.
²²
What this shows is that there is a relation between the amount of neural territory devoted to or associated with touch and the fineness of tactile discrimination that the associated patch of skin can support.

This relation is mediated by experience in humans. This was shown in studies of violin players. People who spend long periods practicing the violin develop enlarged receptive fields for the somatosensory cortex on the side of the brain receiving inputs from the left hand (the hand used for fingering on the violin).
²³
Similarly, blind individuals who learn to read Braille develop greater tactile sensitivity as a result of this experience. This change is partly due to the growth of tactile sensitivity in
visual
areas. Visual areas—so named because they’re typically tuned to visual inputs but not to other kinds of inputs—become touch sensitive in Braille readers.
²⁴

Sometimes, when an area that was responsive to input of one kind is usurped by input of another kind, experience can change in strange ways. The researcher who has done the most to explain these strange effects is Vilayanur Ramachandran, a psychologist at the University of California, San Diego. He has shown that a couple of bizarre phenomena are attributable to neural plasticity.
²⁵

One strange phenomenon is that touch on one part of the body can elicit feelings in other, absent body parts. Ramachandran discovered this while interacting with a man whose arm had been amputated. Acting on reports from the man, Ramachandran asked him to indicate what he felt when Ramachandran touched him gently on his face. “As I’ve told you,” the man said, “when my face is touched, I also feel my hand being touched.” “Do you mean the hand that’s no longer there?” Ramachandran asked. “Yes,” the man replied.
²⁶

The explanation that Ramachandran gave for this strange effect was based on his knowledge that the face and hand regions of the somatosensory
cortex are adjacent. Relying on the idea that neurons try to form connections, Ramachandran suspected that sensory inputs from the young man’s face formed stronger-than-normal connections with the young man’s somatosensory cortex. Ramachandran reasoned that if the hand sector no longer received afferent inputs from the hand, but if touch to the face activated the hand region, then higher centers would treat the inputs as hand-based. Touching the amputee’s face would therefore cause the amputee to think his actually absent hand had been touched.

Ramachandran’s second observation was even droller. It concerned a topic that might better be talked about behind closed doors. If you’re prudish, skip the next three paragraphs.

Some people get their jollies by having their feet fondled. Fondling the soles of the feet or sucking and kissing the toes turn on these foot fetishists. Why such stimulation can lead to so much pleasure is a bit of a mystery, except if you follow the line of reasoning that Ramachandran pursued.

Imagine you’re a neuron—in this case, an afferent neuron projecting from sensory receptors in the foot to the foot region of the somatosensory cortex. Being such a neuron, each time your owner’s foot is touched, you send a signal to the foot zone of the somatosensory cortex. You have no idea, of course, being a neuron, that the signal you’re sending concerns foot touches
per se
. Nor do you know that the region to which you’re sending the signal is the region that gets more signals from the foot than from anywhere else. You’re also unaware that the area next to the foot region happens to be the region that gets signals from the genitals.

Being a typical neuron, however, you try to form connections with neighboring zones. Happily, from your point of view, some of your foot axons get hooked up with the nearby area, the genital area. Because the genital area sends signals to parts of the brain that elicit pleasure, the outcome is the one of interest: Pleasure comes from foot feels because inputs from the foot also feed the part of the brain that gets inputs from the genitals. If you’re someone in whom this cross-talk is especially strong, you may be especially fond of having your feet fondled.

A third strange phenomenon related to neural plasticity is not one that Ramachandran happened to have studied extensively. This one is
synesthesia
. People who experience synesthesia have vivid associations between sights and sounds (to name just two sensory modalities). When they
see
something, they also
hear
something, though what they see is objectively silent. What they hear is also consistently associated with what they see.
²⁷

Synesthesia is due in part to neural plasticity. Sensory inputs get crossed and go where they normally wouldn’t. A sight leads to an illusory sound because inputs that normally go to visual centers also penetrate auditory centers. Likewise for other cross-modal connections. Synesthesia arises, then, because the nervous system isn’t a perfectly regimented place. Rather, it’s a place where interactions can take on surprising twists and turns. This isn’t to say that what goes on in the nervous system is utterly wild and unpredictable. Neural connectivity has considerable orderliness, as indicated in the opening sections of this chapter, but it’s also subject to considerable variability, as indicated in the final sections, starting with the section on neural plasticity. These and the other functional properties of the brain reviewed here are consistent with the view that it’s a jungle in there. Even in a jungle—neural or otherwise—there’s considerable order, but there can also be staggering variety.

Earlier in this book you saw how important it is to resist the idea that your mind has a single head honcho, a top dog who decides things for you. The problem with postulating such a cognitive captain is that granting his or her existence begs the question of how that chief executive decides what to do. Who tells the CEO what commands to issue? You don’t want to end up with a Russian-doll problem, with beings inside beings endlessly passing the ruble.

If you deny that an executive decides things for you, you’re left with the question of how you make decisions. You can say your brain holds elections and, as it turns out, that account does a reasonably good job of explaining neural and behavioral data. But if you go with the election view, you’re still left with questions: How are elections set? What issues are worth voting on? How is the collective attention of the electorate directed to one thing or another? How, if at all, is attention managed? What is attention anyway?

To set sail for explaining attention in Darwinian terms, it helps to think about the problems that are solved by invoking attention. One is filtering. Imagine you’re on a train and you’re trying to read. People in nearby seats are jabbering away, oblivious to your quest for quiet. The topic they’re discussing doesn’t interest you in the least. Who cares if their Aunt Helen smoked more than she should have? Why should it matter if their Uncle Ben knew less about pickles than he claimed? But their chitchat cuts through anyway, leaving you unable to concentrate on the reading you want to do. Try as you might to filter out those other speakers, you can’t help overhearing them. Why can’t you tune them out? At some point you may succeed in doing so, ignoring their chatter to the point you’re no longer aware of doing so. But achieving this state is elusive, like falling asleep or falling in love. It comes unbidden.

The downside of filtering things out is missing what’s important. You may become so engrossed in your reading that you miss the conductor’s call for your station. Filtering needs to be done at the right level. Otherwise you could end up not where you want to go, but in
Nowheresville
.

Some of the earliest research on attention posited that the gate to the mind, if there is one, is on the sensory side. A psychologist at Oxford University, Donald Broadbent, was led to this view by focusing on a task similar to the one described above. In that task, people were asked to listen to two messages at once.
¹
One message was delivered to one ear. Another message was delivered to the other. When participants were instructed to focus on the message coming to one ear, they successfully recalled much of that message, but they recalled virtually nothing of the other. From that outcome, Broadbent drew a reasonable conclusion: People can attend to what reaches one ear but filter out what comes to the other.
²

This conclusion makes sense if you try to read a book on a train and people to your right prattle incessantly. Closing your right ear, as it were, keeps your left ear open. That way, if the conductor calls your station and his or her voice comes to your “open” left ear, you can hear the call; the prattle to your right won’t derail you. On the other hand, if the conductor’s call happens to come to your right ear when he or she calls your station, you won’t hear the call and you’ll miss your stop.

Of course, in a railroad car with ambient noise going to both ears at once, you’d be unlikely to miss a call if you happened to be paying more attention to one side than the other, so I mean this example figuratively, not literally. Still, people do miss their train stops on occasion, often as a result of missing calls for their stations. Perhaps they do so for the reason Broadbent suggested.

There’s a problem with Broadbent’s filter theory, though, that’s not about the prevalence of ambient noise in the environment. The problem is that the hypothesis is factually flawed, as shown by a couple of undergraduates who broadsided Broadbent’s hypothesis with a simple experiment.
³

The two students who disproved their teacher’s hypothesis carried out a variation of the simultaneous listening task. They noted that the version of the task that Broadbent used had subjects listening to two messages that were equally sensible if recalled by either ear. The student experimenters asked what would happen if people listened instead to two messages that made more sense if the listeners
switched
attention from ear to ear. The messages they presented were similar to those appearing below. Each row represents a moment in time. Each column represents an input coming to the left ear (left column) or right ear (right column).