Phantoms in the Brain: Probing the Mysteries of the Human Mind (29 page)

From the experimenter's point of view, your sweaty palms are the most important aspect of your emotional response to the threatening face. The dampness of your hands is a sure giveaway of how you feel toward that person. Moreover, we can measure this reaction very easily by placing electrodes on your palm and recording changes in the electrical resistance of your skin. (Called the galvanic skin response or GSR, this simple little procedure forms the basis of the famous lie detector test. When you tell a fib, your palms sweat ever so slightly. Because damp skin has lower electrical resistance than dry skin, the electrodes respond and you are caught in the lie.) For our purposes, every time you look at your mother or father, believe it or not, your body begins to sweat imperceptibly and your galvanic skin response shoots up as expected.

So, what happens when Arthur looks at his mother or father? My hypothesis predicts that even though he sees them as resembling his parents (remember, the face recognition area of his brain is normal), he should
not
register a change in skin conductance. The disconnection in his brain will prevent his palms from sweating.

With the family's permission, we began testing Arthur on a rainy winter day in our basement laboratory on campus. Arthur sat in a comfortable chair, joking about the weather and how he expected his father's car to float away before we finished the morning's experiments. Sipping hot tea to take the chill from his bones, Arthur gazed at a video screen saver while we affixed two electrodes to his left index finger. Any tiny increase in sweat on his finger would change his skin resistance and show up as a blip on the screen.

Next I showed him a sequence of photos of his mother, father and grandfather interleaved with pictures of strangers, and I compared his galvanic skin responses to that of six college undergraduates who were shown an identical sequence of photos and who served as controls for comparison. Before the experiment, subjects were told that they would be shown pictures of faces, some of which would be familiar and some unfamiliar. After the electrodes were attached, they were shown each photograph for two seconds with a fifteen− to twenty−five−second delay between pictures so skin conductance could return to baseline.

In the undergraduates, I found that there was a big jolt in the GSR in response to photos of their parents—as expected—but not to photos of strangers. In Arthur, on the other hand, the skin response was uniformly low.

There was no increased response to his parents, or at times there would be a tiny blip on the screen after a 117

long delay, as if he were doing a double take. This result provided direct proof that our theory was correct.

Clearly Arthur was not responding emotionally to his parents, and this may be what led to the loss of his galvanic skin response.

But how could we be sure that Arthur was even seeing the faces? Maybe his head injury had damaged the cells in the temporal lobes that would help him distinguish between faces, resulting in a flat GSR whether he looks at his mother or at a stranger. This seemed unlikely, however, since he readily acknowledged that the people who took him to the hospital—his mother and father—looked like his parents. He also had no difficulty in recognizing the faces of famous people like Bill Clinton and Albert Einstein. Still, we needed to test his recognition abilities more direcdy.

To obtain direct proof, I did the obvious thing. I showed Arthur sixteen pairs of photographs of strangers, each pair consisting of either two slightly different pictures of the same person or snapshots of two different people.

We asked him, Do the photographs depict the same person or not? Putting his nose close to each photo and gazing hard at the details, Arthur got fourteen out of sixteen trials correct.

We were now sure that Arthur had no problem in recognizing faces and telling them apart. But could his failure to produce a strong galvanic skin response to his parents be part of a more global disturbance in his emotional abilities? How could we be certain that the head injury had not also damaged his limbic system?

Maybe he had no emotions, period.

This seemed improbable because throughout the months I spent with Arthur, he showed a full range of human emotions. He laughed at my jokes and offered his own funny stories in return. He expressed frustration, fear and anger, and on rare occasions I saw him cry. Whatever the

situation, his emotions were appropriate. Arthur's problem, then, was neither his ability to recognize faces nor his ability to experience emotions; what was lost was his ability to
link
the two.

So far so good, but why is the phenomenon specific to close relatives? Why not call the mailman an impostor, since his, too, is a familiar face?

It may be that when any normal person (including Arthur, prior to his accident) encounters someone who is emotionally very close to him— a parent, spouse or sibling—he expects an emotional "glow," a warm fuzzy feeling, to arise even though it may sometimes be experienced only very dimly. The absence of this glow is therefore surprising and Arthur's only recourse then is to generate an absurd delusion—to rationalize it or to explain it away. On the other hand, when one sees the mailman, one doesn't expect a warm glow and consequently there is no incentive for Arthur to generate a delusion to explain his lack of "warm fuzzy"

response. A mailman is simply a mailman (unless the relationship has taken an amorous turn).

Although the most common delusion among Capgras' patients is the assertion that a parent is an impostor, even more bizarre examples can be found in the older medical literature. Indeed, in a case on record the patient was convinced that his stepfather was a robot, proceeded to decapitate him and opened his skull to look for microchips. Perhaps in this patient, the dissociation from emotions was so extreme that he was forced into an even more absurd delusion than Arthur's: that his stepfather was not even a human being, but was a mindless android!4

About a year ago, when I gave a lecture on Arthur at the Veterans Administration Hospital in La Jolla, a neurology resident raised an astute objection to my theory. What about people who are born with a disease in which their amygdalas (the gateway to the limbic system) calcify and atrophy or those who lose their amygdalas (we each have two of them) completely in surgery or through an accident? Such people do exist, but they do not develop Capgras' syndrome, even though their GSRs are flat to all emotionally evocative 118

stimuli. Likewise, patients with damage to their frontal lobes (which receive and process information from the limbic system for making elaborate future plans) also often lack a GSR. Yet they, too, do not display Capgras'

syndrome.

Why not? The answer may be that these patients experience a general blunting of all their emotional responses and therefore do not have a baseline for comparison. Like a purebred Vulcan or Data on
Star Trek,
one could legitimately argue, they don't even know what an emotion is,

whereas Capgras' patients like Arthur enjoy a normal emotional life in all other respects.

This idea teaches us an important principle about brain function, namely, that all our perceptions—indeed, maybe all aspects of our minds—are governed by comparisons and not by absolute values. This appears to be true whether you are talking about something as obvious as judging the brightness of print in a newspaper or something as subtle as detecting a blip in your internal emotional landscape. This is a far−reaching conclusion, and it also helps illustrate the power of our approach—indeed of the whole discipline that now goes by the name cognitive neuroscience. You can discover important general principles about how the brain works and begin to address deep philosophical questions by doing relatively simple experiments on the right patients. We started with a bizarre condition, proposed an outlandish theory, tested it in the lab and—in meeting objections to it—learned more about how the healthy brain actually works.

Taking these speculations even further, consider the extraordinary disorder called Cotard's syndrome, in which a patient will assert that he is dead, claiming to smell rotten flesh or worms crawling all over his skin.

Again, most people, even neurologists, would jump to the conclusion that the patient was insane. But that wouldn't explain why the delusion takes this highly specific form. I would argue instead that Cotard's is simply an exaggerated form of Capgras' syndrome and probably has a similar origin. In Capgras', the face recognition area alone is disconnected from the amygdala, whereas in Cotard's perhaps all the sensory areas are disconnected from the limbic system, leading to a complete lack of emotional contact with the world. Here is another instance in which an outlandish brain disorder that most people regard as a psychiatric problem can be explained in terms of known brain circuitry. And once again, these ideas can be tested in the laboratory. I would predict that Cotard's syndrome patients will have a complete loss of GSR for all external stimuli—not just faces—and this leaves them stranded on an island of emotional desolation, as close as anyone can come to experiencing death.

Arthur seemed to enjoy his visits to our laboratory. His parents were pleased that there was a logical explanation for his predicament, that he wasn't just "crazy." I never revealed the details to Arthur because I wasn't sure how he'd react.

Arthur's father was an intelligent man, and at one point, when Arthur wasn't around, he asked me, "If your theory is correct, doctor—if the information doesn't get to his amygdala—then how do you explain how he has no problems recognizing us over the phone? Does that make sense to you?"

"Well," I replied, "there is a separate pathway from the auditory cortex, the hearing area of the temporal lobes, to the amygdala. One possibility is that this hearing route has not been affected by the accident— only the visual centers have been disconnected from Arthur's amygdala."

This conversation got me wondering about the ot her well−known functions of the amygdala and the visual centers that project to it. In particular, scientists recording cell responses in the amygdala found that, in addition to responding to facial expression and emotions, the cells also respond to the direction of eye gaze.

For instance, one cell might fire if another person is looking directly at you, whereas a neighboring cell will fire only if that person's gaze is averted by a fraction of an inch. Still other cells fire when the gaze is way off 119

to the left or the right.

This phenomenon is not surprising, given the important role that gaze direction5 plays in primate social communications—the averted gaze of guilt, shame or embarrassment; the intense, direct gaze of a lover or the threatening stare of an enemy. We tend to forget that emotions, even though they are privately experienced, often involve interactions with other people and that one way we interact is through eye contact. Given the links among gaze direction, familiarity and emotions, I wondered whether Arthur's ability to judge the direction of gaze, say, by looking at photographs of faces, would be impaired.

To find out, I prepared a series of images, each showing the same model looking either directly at the camera lens or at a point an inch or two to the right or left of the lens. Arthur's task was simply to let us know whether the model was looking straight at him or not. Whereas you or I can detect tiny shifts in gaze with uncanny accuracy, Arthur was hopeless at the task. Only when the model's eyes were looking way off to one side was he able to discern correctly that she wasn't looking at him.

This finding in itself is interesting but not altogether unexpected, given the known role of amygdala and temporal lobes in detecting gaze direction. But on the eighth trial of looking at these photos, Arthur did something completely unexpected. In his soft, almost apologetic voice, he exclaimed that the model's identity had changed. He was now looking at a new person!

This meant that a mere change in direction of gaze had been sufficient to provoke Capgras' delusion. For Arthur, the "second" model was apparently a new person who merely resembled the "first."

"This one is older," Arthur said firmly. He stared hard at both images. "This is a lady; the other one is a girl."

Later in the sequence, Arthur made another duplication—one model was old, one young and a third even younger. At the end of the test session he continued to insist that he had seen three different people. Two weeks later he did it again on a retest using images of a completely new face.

How could Arthur look at the face of what was obviously one person and claim that she was actually three different people? Why did simply changing the direction of gaze lead to this profound inability to link successive images?

Answers lie in the mechanics of how we form memories, in particular our ability to create enduring representations of faces. For example, suppose you go to the grocery store one day and a friend introduces you to a new person—Joe. You form a memory of that episode and tuck it away in your brain. Two weeks go by and you run into Joe in the library. He tells you a story about your mutual friend, you share a laugh and your brain files a memory about this second episode. Another few weeks pass and you meet Joe again in his office—he's a medical researcher and he's wearing a white lab coat—but you recognize him instantly from earlier encounters. More memories of Joe are created during this time so that you now have in your mind a

"category" called Joe. This mental picture becomes progressively refined and enriched each time you meet Joe, aided by an increasing sense of familiarity that creates an incentive to link the images and the episodes.

Eventually you develop a robust concept of Joe—he tells great stories, works in a lab, makes you laugh, knows a lot about gardening, and so forth.

Now consider what happens to someone with a rare and specific form of amnesia, caused by damage to the hippocampus (another important brain structure in the temporal lobes). These patients have a complete inability to form new memories, even though they have perfect recollection of all events in their lives that took place before the hippocampus was injured. The logical conclusion to be drawn from the syndrome is not that memories are actually stored in the hippocampus (hence the preservation of old memories), but that the hippocampus is vital for the acquisition of new memory traces in the brain. When such a patient meets a new person (Joe) on three consecutive occasions—in the supermarket, the library and the office—he will not 120