The Tell-Tale Brain

An analogy might be helpful. A rare blood disorder called sickle cell anemia is caused by a defective recessive gene that causes red blood cells to assume an abnormal “sickle” shape, making them unable to transport oxygen. This can be fatal. If you happen to inherit two copies of this gene (in the unlikely event that both your parents had either the trait or the disease itself), then you develop the full-blown disease. However, if you inherit just one copy of this gene, you do not come down with the disease, though you can still pass it on to your children. Now it turns out that, although sickle-cell anemia is extremely rare in most parts of the world, where natural selection has effectively weeded it out, its incidence is ten times higher in certain parts of Africa. Why should this be? The surprising answer is that the sickle-cell trait actually seems to protect the affected individual from malaria, a disease caused by a mosquito-borne parasite that infects and destroys blood cells. This protection conferred on the population as a whole from malaria outweighs the reproductive disadvantage caused by the occasional rare appearance of an individual with double copies of the sickle-cell gene. Thus the apparently maladaptive gene has actually been selected for by evolution, but only in geographic locations where malaria is endemic.

A similar argument has been proposed for the relatively high incidence of schizophrenia and bipolar disorder in humans. The reason these disorders have not been weeded out may be because having some of the genes that lead to the full-blown disorder are advantageous—perhaps boosting creativity, intelligence, or subtle social-emotional faculties. Thus humanity as a whole benefits from keeping these genes in its gene pool, but the unfortunate side effect is a sizable minority who get bad combinations of them.

Carrying this logic forward, the same could well be true for synesthesia. We have seen how, by dint of anatomy, genes that lead to enhanced cross-activation between brain areas could have been highly advantageous by making us creative as a species. Certain uncommon variants or combinations of these genes might have the benign side effect of producing synesthesia. I hasten to emphasize the part about benign: Synesthesia is not deleterious like sickle-cell disease and mental illness, and in fact most synesthetes seem to really enjoy their abilities and would not opt to have them “cured” even if they could. This is only to say that the general mechanism might be the same. This idea is important because it makes clear that synesthesia and metaphor are not synonymous, and yet they share a deep connection that might give us deep insights into our marvelous uniqueness.⁶

Thus synesthesia is best thought of as an example of subpathological cross-modal interactions that could be a signature or marker for creativity. (A modality is a sensory faculty, such as smell, touch, or hearing. “Cross-modal” refers to sharing information between senses, as when your vision and hearing together tell you that you’re watching a badly dubbed foreign film.) But as often happens in science, it got me thinking about the fact that even in those of us who are nonsynesthetes a great deal of what goes on in our mind depends on entirely normal cross- modal interactions that are not arbitrary. So there is a sense in which at some level we are all “synesthetes.” For example, look at the two shapes in Figure 3.7. The one on the left looks like a paint splat. The one on the right resembles a jagged piece of shattered glass. Now let me ask you, if you had to guess, which of these is a “bouba” and which is a “kiki”? There is no right answer, but odds are you picked the splat as “bouba” and the glass as “kiki.” I tried this in a large classroom recently, and 98 percent of the students made this choice. Now you might think this has something to do with the blob resembling the physical form of the letter B (for “bouba”) and the jagged thing resembling a K (as in “kiki”). But if you try the experiment on non-English-speaking people in India or China, where the writing systems are completely different, you find exactly the same thing.

Why does this happen? The reason is that the gentle curves and undulations of contour on the amoeba-like figure metaphorically (one might say) mimic the gentle undulations of the sound bouba, as represented in the hearing centers in the brain and in the smooth rounding and relaxing of the lips for producing the curved booo-baaa sound. On the other hand, the sharp wave forms of the sound kee-kee and the sharp inflection of the tongue on the palate mimic the sudden changes in the jagged visual shape. We will return to this demonstration in Chapter 6 and see how it might hold the key to understanding many of the most mysterious aspects of our minds, such as the evolution of metaphor, language, and abstract thought.⁷

FIGURE 3.7 Which of these shapes is “bouba” and which is “kiki”? Such stimuli were originally used by Heinz Werner to explore interactions between hearing and vision.

I HAVE ARGUED so far that synesthesia, and in particular the existence of “higher” forms of synesthesia (involving abstract concepts rather than concrete sensory qualities) can provide clues to understanding some of the high-level thought processes that humans alone are capable of.⁸ Can we apply these ideas to what is arguably the loftiest of our mental traits, mathematics? Mathematicians often speak of seeing numbers laid out in space, roaming this abstract realm to discover hidden relationships that others might have missed, such as Fermat’s Last Theorem or Goldbach’s conjecture. Numbers and space? Are they being metaphorical?

One day in 1997, after I had consumed a glass of sherry, I had a flash of insight—or at least thought I had. (Most of the “insights” I have when inebriated turn out to be false alarms.) In his original Nature paper, Galton described a second type of synesthesia that is even more intriguing than the number-color condition. He called it “number forms.” Other researchers use the phrase “number line.” If I asked you to visualize the numbers 1 to 10 in your mind’s eye, you will probably report a vague tendency to see them mapped in space sequentially, left to right, as you were taught in grade school. But number-line synesthetes are different. They able to visualize numbers vividly and do not see the numbers arranged sequentially from left to right, but on a snaking, twisting line that can even double back on itself, so that 36 might be closer to 23, say, than it is to 38 (Figure 3.8). One could think of this as “number-space” synesthesia, in which every number is always in a particular location in space. The number line for any individual remains constant even if tested on intervals separated by months.

FIGURE 3.8 Galton’s number line. Notice that 12 is a tiny bit closer to 1 than it is to 6.

As with all experiments in psychology, we needed a method to prove Galton’s observation experimentally. I called upon my students Ed Hubbard and Shai Azoulai to help set up the procedures. We first decided to look at the well-known “number distance” effect seen in normal people. (Cognitive psychologists have examined every conceivable variation of the effect on hapless student volunteers, but its relevance to number-space synesthesia was missed until we came along.) Ask anyone which of two numbers is larger, 5 or 7? 12 or 50? Anyone who has been through grade school will get it right every time. The interesting part comes when you clock how long it takes people to spit out each of their answers. This latency between showing them a number pair and their verbal response is their reaction time (RT). It turns out that the greater the distance between two numbers the shorter the RT, and contrariwise, the closer two numbers are, the longer it takes to form an answer. This suggests that your brain represents numbers in some sort of an actual mental number line which you consult “visually” to determine which is greater. Numbers that are far apart can be easily eyeballed, while numbers that are close together need closer inspection, which takes a few extra milliseconds.

We realized we could exploit this paradigm to see if the convoluted number-line phenomenon really existed or not. We could ask a number-space synesthete to compare number pairs and see if her RTs corresponded to the real