The Tell-Tale Brain

isn’t identical to the first. What seems to matter for recognition is not the exact pattern of nerve impulses but which neurons fire and how much they fire—a principle known as Muller’s law of specific nerve energies. Proposed in 1826, the law states that the different perceptual qualities evoked in the brain by sound, light, and pinprick—namely, hearing, seeing, and pain—are not caused by differences in patterns of activation but by different locations of nervous structures excited by those stimuli.

That’s the standard story, but an astonishing new discovery by two neuroscientists, Wolf Singer of the Max Planck Institute for Brain Research in Frankfurt, Germany, and Charles Gray from Montana State University, adds a novel twist to it. They found that if a monkey looks at a big object of which only fragments are visible, then many cells fire in parallel to signal the different fragments. That’s what you would expect. But surprisingly, as soon as the features are grouped into a whole object (in this case, a lion), all the spike trains become perfectly synchronized. And so the exact spike trains do matter. We don’t yet know how this occurs, but Singer and Gray suggest that this synchrony tells higher brain centers that the fragments belong to a single object. I would take this argument a step further and suggest that this synchrony allows the spike trains to be encoded in such a way that a coherent output emerges which is relayed to the emotional core of the brain, creating an “Aha! Look here, it’s an object!” jolt in you. This jolt arouses you and makes you swivel your eyeballs and head toward the object, so you can pay attention to it, identify it, and take action. It’s this “Aha!” signal that the artist or designer exploits when she uses grouping. This isn’t as far-fetched as it sounds; there are known back projections from the amygdala and other limbic structures (such as the nucleus accumbens) to almost every visual area in the hierarchy of visual processing discussed in Chapter 2. Surely these projections play a role in mediating the visual “Aha!”

The remaining universal laws of aesthetics are less well understood, but that hasn’t stopped me from speculating on their evolution. (This isn’t easy; some laws may not themselves have a function but may be byproducts of other laws that do.) In fact, some of the laws actually seem to contradict each other, which may actually turn out to be a blessing. Science often progresses by resolving apparent contradictions.

The Law of Peak Shift

My second universal law, the peak-shift effect, relates to how your brain responds to exaggerated stimuli. (I should point out that the phrase “peak shift” has a purportedly precise meaning in the animal learning literature, whereas I am using it more loosely.) It explains why caricatures are so appealing. And as I mentioned earlier, ancient Sanskrit manuals on aesthetics often use the word rasa, which translates roughly to “capturing the very essence of something.” But how exactly does the artist extract the very essence of something and portray it in a painting or a sculpture? And how does your brain respond to rasa?

A clue, oddly enough, comes from studies in animal behavior, especially the behavior of rats and pigeons that are taught to respond to certain visual images. Imagine a hypothetical experiment in which a rat is being taught to discriminate a rectangle from a square (Figure 7.6). Every time the animal approaches the rectangle, you give it a piece of cheese, but if it goes to the square you don’t. After a few dozen trials, the rat learns that “rectangle = food,” it begins to ignore the square and go toward the rectangle alone. In other words, it now likes the rectangle. But amazingly, if you now show the rat a longer and skinnier rectangle than the one you showed it originally, it actually prefers that rectangle to the original! You may be tempted to say, “Well, that’s a bit silly. Why would the rat actually choose the new rectangle rather than the one you trained it with?” The answer is the rat isn’t being silly at all. It has learned a rule—“rectangularity”—rather than a particular prototype rectangle, so from its point of view, the more rectangular, the better. (By that, one means “the higher the ratio of a longer side to a shorter side, the better.”) The more you emphasize the contrast between the rectangle and the square, the more attractive it is, so when shown the long skinny one the rat thinks, “Wow! What a rectangle.”

This effect is called peak shift because ordinarily when you teach an animal something, its peak response is to the stimulus you trained it with. But if you train the animal to discriminate something (in this case, a rectangle) from something else (the square), the peak response is to a totally new rectangle that is shifted away even further from the square in its rectangularity.

What has peak shift got to do with art? Think of caricatures. As I mentioned in Chapter 2, if you want to draw a caricature of Nixon’s face, you take all those features of Nixon that make his face special and different from the average face, such as his big nose and shaggy eyebrows, and you amplify them. Or to put it differently, you take the mathematical average of all male faces and subtract this average from Nixon’s face, and then amplify the difference. By doing this you have created a picture that’s even more Nixon-like than the original Nixon! In short, you have captured the very essence—the rasa—of Nixon. If you overdo it, you get a humorous effect—a caricature—because it doesn’t look even human; but if you do it right, you get great portraiture.

FIGURE 7.6 Demonstration of the peak shift principle: The rat is taught to prefer the rectangle (2) over the square (1) but then spontaneously prefers the longer, skinnier rectangle (3).

Caricatures and portraits aside, how does this principle apply to other art forms? Take a second look at the goddess Parvati (Figure 7.2a), which conveys the essence of feminine sensuality, poise, charm, and dignity. How does the artist achieve this? A first-pass answer is that he has subtracted the average male form from the average female form and amplified the difference. The net result is a woman with exaggerated breasts and hips and an attenuated hourglass waist: slender yet voluptuous. The fact that she doesn’t look like your average real woman is irrelevant; you like the sculpture just as the rat liked the skinnier rectangle more than the original prototype, saying, in effect, “Wow! What a woman!” But there’s surely more to it than that, otherwise any Playboy pinup would be a work of art (although, to be sure, I’ve never seen a pinup whose waist is as narrow as the goddess’s).

Parvati is not merely a sexy babe; she is the very embodiment of feminine perfection—of grace and poise. How does the artist achieve this? He does so by accentuating not merely her breasts and hips but also her feminine posture (formally known as tribhanga, or “triple flexion,” in Sanskrit). There are certain postures that a woman can adopt effortlessly but are impossible (or highly improbable) in a man because of anatomical differences such as the width of the pelvis, the angle between the neck and shaft of the femur, the curvature of the lumbar spine. Instead of subtracting male form from female form, the artist goes into a more abstract posture space, subtracting the average male posture from the average female posture, and then amplifies the difference. The result is an exquisitely feminine posture, conveying poise and grace.

Now take a look at the dancing nymph in Figure 7.7 whose twisting torso is almost anatomically absurd but who nevertheless conveys an incredibly beautiful sense of movement and dance. This is probably achieved, once again, by the deliberate exaggeration of posture that may activate—indeed hyperactivate—mirror neurons in the superior temporal sulcus. These cells respond powerfully when a person is viewing changing postures and movements of the body as well as changing facial expressions. (Remember pathway 3, the “so what” stream in vision processing discussed in Chapter 2?) Perhaps sculptures such as the dancing nymph are producing an especially powerful stimulation of certain classes of mirror neurons, resulting in a correspondingly heightened reading of the body language of dynamic postures. It’s hardly surprising, then, that even most types of dance— Indian or Western—involve clever ritualized exaggerations of movements and postures that convey specific emotions. (Remember Michael Jackson?)