Klin had Peter put on a hat with a very simple, but powerful, eye-tracking device composed of two tiny cameras. One camera recorded the movement of Peter’s fovea—the centerpiece of his eye. The other camera recorded whatever it was Peter was looking at, and then the two images were superimposed. This meant that on every frame of the movie, Klin could draw a line showing where Peter was looking at that moment. He then had people without autism watch the movie as well, and he compared Peter’s eye movements with theirs. In one scene, for example, Nick (George Segal) is making polite conversation, and he points to the wall of host George’s (Richard Burton’s) study and asks, “Who did the painting?” The way you and I would look at that scene is straightforward: our eyes would follow in the direction that Nick is pointing, alight on the painting, swivel back to George’s eyes to get his response, and then return to Nick’s face, to see how he reacts to the answer. All of that takes place in a fraction of a second, and on Klin’s visual-scanning pictures, the line representing the gaze of the normal viewer forms a clean, straight-edged triangle from Nick to the painting to George and back again to Nick. Peter’s pattern, though, is a little different. He starts somewhere around Nick’s neck. But he doesn’t follow the direction of Nick’s arm, because interpreting a pointing gesture requires, if you think about it, that you instantaneously inhabit the mind of the person doing the pointing. You need to read the mind of the pointer, and, of course, people with autism can’t read minds. “Children respond to pointing gestures by the time they are twelve months old,” Klin said. “This is a man who is forty-two years old and very bright, and he’s not doing that. Those are the kinds of cues that children are learning naturally—and he just doesn’t pick up on them.”
So what does Peter do? He hears the words “painting” and “wall,” so he looks for paintings on the wall. But there are three in the general vicinity. Which one is it? Klin’s visual-scanning pictures show Peter’s gaze moving frantically from one picture to the other. Meanwhile, the conversation has already moved on. The only way Peter could have made sense of that scene is if Nick had been perfectly, verbally explicit—if he had said, “Who did that painting to the left of the man and the dog?” In anything less than a perfectly literal environment, the autistic person is lost.
There’s another critical lesson in that scene. The normal viewers looked at the eyes of George and Nick when they were talking, and they did that because when people talk, we listen to their words and watch their eyes in order to pick up on all those expressive nuances that Ekman has so carefully catalogued. But Peter didn’t look at anyone’s eyes in that scene. At another critical moment in the movie, when, in fact, George and Martha (Elizabeth Taylor) are locked in a passionate embrace, Peter looked not at the eyes of the kissing couple—which is what you or I would do—but at the light switch on the wall behind them. That’s not because Peter objects to people or finds the notion of intimacy repulsive. It’s because if you cannot mind-read—if you can’t put yourself in the mind of someone else—then there’s nothing special to be gained by looking at eyes and faces.
One of Klin’s colleagues at Yale, Robert T. Schultz, once did an experiment with what is called an FMRI (functional magnetic resonance imagery), a highly sophisticated brain scanner that shows where the blood is flowing in the brain at any given time—and hence, which part of the brain is in use. Schultz put people in the FMRI machine and had them perform a very simple task in which they were given either pairs of faces or pairs of objects (such as chairs or hammers) and they had to press a button indicating whether the pairs were the same or different. Normal people, when they were looking at the faces, used a part of their brain called the fusiform gyrus, which is an incredibly sophisticated piece of brain software that allows us to distinguish among the literally thousands of faces that we know. (Picture in your mind the face of Marilyn Monroe. Ready? You just used your fusiform gyrus.) When the normal participants looked at the chair, however, they used a completely different and less powerful part of the brain—the inferior temporal gyrus—which is normally reserved for objects. (The difference in the sophistication of those two regions explains why you can recognize Sally from the eighth grade forty years later but have trouble picking out your bag on the airport luggage carousel.) When Schultz repeated the experiment with autistic people, however, he found that they used their object-recognition area for both the chairs and the faces. In other words, on the most basic neurological level, for someone with autism, a face is just another object. Here is one of the earliest descriptions of an autistic patient in the medical literature: “He never looked up at people’s faces. When he had any dealings with persons at all, he treated them, or rather parts of them, as if they were objects. He would use a hand to lead him. He would, in playing, butt his head against his mother as at other times he did against a pillow. He allowed his boarding mother’s hand to dress him, paying not the slightest attention to her.”
So, when Peter looked at the scene of Martha and George kissing, their two faces did not automatically command his attention. What he saw were three objects—a man, a woman, and a light switch. And what did he prefer? As it happens, the light switch. “I know for [Peter] that light switches have been important in his life,” says Klin. “He sees a light switch, and he gravitates toward it. It’s like if you were a Matisse connoisseur, and you look at a lot of pictures, and then you’d go, ahh, there is the Matisse. So he goes, there is the light switch. He’s seeking meaning, organization. He doesn’t like confusion. All of us gravitate toward things that mean something to us, and for most of us, that’s people. But if people don’t anchor meaning for you, then you seek something that does.”
Perhaps the most poignant scene Klin studied comes at a point in the movie when Martha is sitting next to Nick, flirting outrageously, even putting a hand on his thigh. In the background, his back slightly turned to them, lurks an increasingly angry and jealous George. As the scene unfolds, the normal viewer’s eyes move in an almost perfect triangle from Martha’s eyes to Nick’s eyes to George’s eyes and then back to Martha’s, monitoring the emotional states of all three as the temperature in the room rises. But Peter? He starts at Nick’s mouth, and then his eyes drop to the drink in Nick’s hand, and then his gaze wanders to a brooch on Martha’s sweater. He never looks at George at all, so the entire emotional meaning of the scene is lost on him.
“There’s a scene where George is about to lose his temper,” says Warren Jones, who worked with Klin on the experiment. “He goes to the closet and pulls a gun down from the shelf, and points it directly at Martha and pulls the trigger. And when he does, an umbrella pops out the front of the barrel. But we have no idea until it comes out that it’s a ruse—so there is this genuine moment of fear. And one of the most telltale things is that the classic autistic individual will laugh out loud and find it to be this moment of real physical comedy. They’ve missed the emotional basis for the act. They read only the superficial aspect that he pulls the trigger, an umbrella pops out, and they walk away thinking, those people were having a good time.”
Peter’s movie-watching experiment is a perfect example of what happens when mind reading fails. Peter is a highly intelligent man. He has graduate degrees from a prestigious university. His IQ is well above normal, and Klin speaks of him with genuine respect. But because he lacks one very basic ability—the ability to mind-read—he can be presented with that scene in Who’s Afraid of Virginia Woolf? and come to a conclusion that is socially completely and catastrophically wrong. Peter, understandably, makes this kind of mistake often: he has a condition that makes him permanently mind-blind. But I can’t help but wonder if, under certain circumstances, the rest of us could momentarily think like Peter as well. What if it were possible for autism—for mind-blindness—to be a temporary condition instead of a chronic one? Could that explain why sometimes otherwise normal people come to conclusions that are completely and catastrophically wrong?