The Tell-Tale Brain: A Neuroscientist's Quest for What Makes Us Human

A common fallacy is that science begins with naïve unprejudiced observations about the world while in fact the opposite is true. When exploring new terrain, you always begin with a tacit hypothesis of what might be true—a preconceived notion or prejudice. As the British zoologist and philosopher of science Peter Medawar once said, we are not “cows grazing on the pasture of knowledge.” Every act of discovery involves two critical steps: first, unambiguously stating your conjecture of what might be true, and second, devising a crucial experiment to test your conjecture. Most theoretical approaches to aesthetics in the past have been concerned mainly with step 1 but not step 2. Indeed, the theories are usually not stated in a manner that permits either confirmation or refutation. (One notable exception is Brent Berlin’s pioneering work on the use of the galvanic skin response.)

Can we experimentally test our ideas about peak shift, supernormal stimuli, and other laws of aesthetics? There are at least three ways of doing so. The first one is based on the galvanic skin response (GSR); the second is based on recording nerve impulses from single nerve cells in the visual area in the brain; and the third is based on the idea that if there is anything to these laws, we should be able to use them to devise new pictures that are more attractive than what you might have predicted from common sense (what I refer to as the “grandmother test”: If an elaborate theory cannot predict what your grandmother knows using common sense, then it isn’t worth much).

You already know about GSR from previous chapters. This test provides an excellent, highly reliable index of your emotional arousal when you look at anything. If you look at something scary, violent, or sexy (or, as it turns out, a familiar face like your mother or Angelina Jolie), there is a big jolt in GSR, but nothing happens if you look at a shoe or furniture. This is a better test of someone’s raw, gut-level emotional reactions to the world than asking what she feels. A person’s verbal response is likely to be inauthentic. It may be contaminated by the “opinions” of other areas of the brain.

So GSR gives us a handy experimental probe for understanding art. If my conjectures about the appeal of Henry Moore sculptures are correct, then the Renaissance scholar who denies an interest in such abstract works (or, for that matter, the English art historian who feigns indifference to Chola bronzes) should nevertheless register a whopping GSR to the very images whose aesthetic appeal he denies. His skin can’t lie. Similarly, we know that you will show a higher GSR to a photo of your mother than to a photo of a stranger, and I predict that the difference will be even greater if you look at a caricature or evocative sketch of your mother rather than at a realistic photo. This would be interesting because it’s counterintuitive. As a control for comparison, you could use a countercaricature, by which I mean a sketch that deviates from the prototype toward the average face rather than away from it (or indeed, a face outline that deviates in a random direction). This would ensure that any enhanced GSR you observed with the caricature wasn’t simply because of the surprise caused by the distortion. It would be genuinely due to its appeal as a caricature.

But GSR can only take us so far; it is a relatively coarse measure because it pools several types of arousal and it can’t discriminate positive from negative responses. But even though it’s a crude measure, it’s not a bad place to start because it can tell the experimenter when you are indifferent to a work of art and when you are feigning indifference. The criticism that the test can’t discriminate negative arousal from positive arousal (at least not yet!) isn’t as damaging as it sounds because who is to say that negative arousal isn’t also part of art? Indeed, attention grabbing—whether initially positive or negative—is often a prelude to attraction. (After all, slaughtered cows pickled in formaldehyde were displayed in the venerable MOMA [Museum of Modern Art] in New York, sending shock waves throughout the art world). There are many layers of reaction to art, which contribute to its richness and appeal.

A second approach is to use eye movements, in particular, a technique pioneered by the Russian psychologist Alfred Yarbus. You can use an electronic optical device to see where a person is fixating and how she is moving her eyes from one region to another in a painting. The fixations tend to be clustered around eyes and lips. One could therefore show a normally proportioned cartoon of a person on one side of the image and a hyperbolic version on the other side. I would predict that even though the normal cartoon looks more natural, the eye fixations will cluster more around the caricature. (A randomly distorted cartoon could be included to control for novelty.) These findings could be used to complement the GSR results.