We Look but Don't Always See
A man walks into a bar. The man's name is Burt Reynolds. Yes, that Burt Reynolds. Except this is early in his career, and nobody knows him yet--including a guy at the end of the bar with huge shoulders.
Reynolds sits down two stools away and begins sipping a beer and tomato juice. Out of nowhere, the guy starts harassing a man and a woman seated at a table nearby. Reynolds tells him to watch his language. That's when the guy with the huge shoulders turns on Reynolds. And rather than spoil what happens next, I'll let you hear it from Burt Reynolds himself, who recounted the story years ago in an interview with Playboy magazine:
I remember looking down and planting my right foot on this brass rail for leverage, and then I came around and caught him with a tremendous right to the side of the head. The punch made a ghastly sound and he just flew off the stool and landed on his back in the doorway, about 15 feet away. And it was while he was in mid-air that I saw . . . that he had no legs.
Only later, as Reynolds left the bar, did he notice the man's wheelchair, which had been folded up and tucked next to the doorway.
As mistakes go, punching out a guy with no legs is a lulu. But for our purposes the important part of the anatomy in this story is not the legs but the eyes. Even though Reynolds was looking right at the man he hit, he didn't see all that he needed to see. In the field of human error, this kind of mistake is so common that researchers have given it its own nickname: a "looked but didn't see" error. When we look at something--or at someone--we think we see all there is to see. But we don't. We often miss important details, like legs and wheelchairs, and sometimes much larger things, like doors and bridges.
We See a Fraction of What We Think We See
To understand why we do this, it helps to know something about the eye and how it works. The eye is not a camera. It does not take "pictures" of events. And it does not see everything at once. The part of the visual field that can be seen clearly at any given time is only a fraction of the total. At normal viewing distances, for instance, the area of clear vision is about the size of a quarter. The eye deals with this constraint by constantly darting about, moving and stopping roughly three times a second.
What is seen as the eyes move about depends, in part, on who is doing the seeing. Men, for instance, have been shown to notice different things from those that women do. When viewing a mock purse snatching by a male thief, for instance, women tended to notice the appearance and actions of the woman whose purse was being snatched; men, on the other hand, were more accurate regarding details about the thief. Right-handed people have also been shown to remember the orientation of certain objects they have seen more accurately than left-handers do. Years ago, after the Hale-Bopp comet made a spectacular appearance in the evening skies, investigators in England asked left- and right-handers if they could remember which way the comet had been facing when they saw it. Right-handers were significantly more likely than lefties to remember that the comet had been facing to the left. Handedness is also the best predictor of a person's directional preference. When people are forced to make a turn at an intersection, right-handers, at least in the United States, prefer turning right, and lefties prefer turning left. As a result, advised the authors of one study, "one should look to the left when searching for the shortest lines of people at stores, banks and the like."
The Expert's Quiet Eye
In fact, what we see is, in part, a function not only of who we are but of what we are. Researchers have demonstrated that different people can view the same scene in different ways. Say you're a golfer, for instance. Even better, say you're a great golfer with a low handicap. You're playing your buddy, who's not so great. You've teed off and played through the fairway, and now it comes time to putt. Do you and your buddy look at the ball in the same way?
Why? Because experts and novices tend to look at things in different ways. One of these differences involves something known as the "quiet-eye period." This is the amount of time needed to accurately program motor responses. It occurs between the last glimpse of our target and the first twitch of our nervous system. Researchers have documented expert-novice differences in quiet-eye periods in a number of sports, ranging from shooting free throws in basketball to shooting rifles in Olympic-style competition. The consistent finding is that experts maintain a longer quiet-eye period.
In the final few seconds of the putt, good golfers with low handicaps tend to gaze at the ball much longer and rarely shift their sight to the club or to any other location. Less-skilled golfers, on the other hand, don't stare at the ball very long and tend to look at their club quite often. Superior vision is so important in golf that many of the world's best players, including Tiger Woods and at least seven other PGA Tour winners, have had Lasik surgery to correct their vision, usually to twenty-fifteen or better. That means they can see clearly at twenty feet what people with twenty-twenty vision could see clearly only at fifteen feet. The sportswear giant Nike has even introduced a new putter, the IC, designed to reduce visual distractions. The shaft and the grip of the $140 putter are both green (to blend in with the color of the grass and reduce distraction), but the leading edge of the blade and the T-shaped alignment line are a blazing white, to help focus a golfer's eyes on the part of the club that contacts the ball.
We Notice on a Need-to-Know Basis
Regardless of whether we are experts or amateurs, even those of us with otherwise perfect vision are subject to fleeting but nonetheless startling kinds of blindness. One of the most fascinating forms is known as change blindness. It occurs when we fail to detect major changes to the scenes we are viewing during a brief visual disruption--even so brief as a blink.
The profound impact of change blindness was demonstrated a decade ago in an impish experiment by Daniel Simons and Daniel Levin, both of them at the time at Cornell University. The design of their experiment was simple: they had "strangers" on a college campus ask pedestrians for directions. As you might suspect, the experiment involved a twist. As the stranger and the pedestrian talk, the experimenters arranged for them to be rudely interrupted by two men who pass between them while carrying a door. The interruption is brief--lasting just one second. But during that one second, something important happens. One of the men carrying the door trades places with the "stranger." When the door is gone, the pedestrian is confronted with a different person, who continues the conversation as if nothing had happened. Would the pedestrians notice that they were talking to someone new?
In most cases, it turns out, the answer was no.
Only seven of the fifteen pedestrians reported noticing the change.
At this point, you may find it tempting to think, "I would have noticed a change like that." And maybe you would have. But consider this: you've probably seen countless similar changes and never noticed them. Where? In the movies. Movie scenes, as many people know, are not filmed sequentially; instead, they are shot in a different order from how they appear in the film, usually months or even years apart. This process often results in embarrassing mistakes known in the trade as continuity errors.
Continuity errors have long bedeviled the motion picture industry. The Hollywood epic Ben-Hur is a good example. The 1959 movie, which starred the late Charlton Heston as Ben-Hur, won eleven Academy Awards--more than any other movie up to that point in history, including one for Best Picture. But it still has its share of errors, especially in the famous chariot scene, which lasts for eleven minutes but took three months to film. During the chariot race, Messala damages Ben-Hur's chariot with his saw-toothed wheel hubs. But at the end of the race, if you'll look closely, you'll see that Ben-Hur's chariot appears--undamaged! There's also a mix-up in the number of chariots. The race begins with nine chariots. During the race, six crash. That should leave three chariots at the end of the race. Instead, there are four.
Hollywood employs experts who are supposed to catch these things. Officially, they are known as continuity editors or script supervisors, though they are more commonly referred to as script girls because the role, traditionally, has been filled by women. But even they can't catch all the mistakes.
"It's not humanly possible," says Claire Hewitt, who has supervised scripts on a variety of movies, from documentaries and short films to full-length features and even kung-fu action flicks. The best you can do in any given scene, she says, is to try to spot the most important things. But even that is easier said than done.
One of Hewitt's more memorable lapses occurred in her second film as a script supervisor, a short film about a man and a woman who live next door to each other in an apartment building. Instead of filming the actors in separate rooms, though, the filmmakers cheated: they used the same room to film both actors. This required redecorating the room to make it appear in the various scenes to belong to either the man or the woman, but it saved on location costs.
The error occurs in a key scene of the movie, when the woman finally meets the man. "You see her leaning against the door, listening to whether he's out in the hall, and she comes out," says Hewitt. "But the door opens the wrong way!"
Hewitt never noticed the error on her own; it was instead brought to her attention by her mother's boyfriend. "People love doing that--catching you out," says Hewitt. Indeed, entire Web sites are devoted to pointing out continuity errors. (One of the more popular ones is the British Web site moviemistakes.com, run by Jon Sandys, who has been cataloging movie flubs since he was seventeen.) But Hewitt's experience with her mother's boyfriend carries an important lesson: errors that are obvious to others can be invisible to us, no matter how hard we try to spot them.
Okay, you might say, it's easy enough to miss changes to minor details like which way a door opens. Who cares? But what about changes to bigger, more important things?
That's what Levin and Simons wanted to find out. So they shot their own movie. This time, they didn't just change the scenery; they changed the actors. During each film, one actor was replaced by another. For example, in one film an actor walked through an empty classroom and began to sit in a chair. The camera then changed, or cut to a closer view, and a different actor completed the action. The films were shown to forty students. Only a third of them noticed the change.
We See What We Are
When we look at something, we intuitively feel that we can see everything in it in great detail and are quite confident that we would notice any changes. That, said Simons, is what makes change blindness such an interesting problem. "People consistently believe that if something unexpected changes, it will automatically grab their attention and they will notice it." As part of their "door" experiment, for instance, the two Dans polled a group of fifty students. They read them a description of the experiment, then asked them to raise their hands if they believed they would be able to detect the changes. All fifty raised their hands.
The eye, says Simons, has high resolution only at an angle of two degrees. That's not much. If you hold your fist out at arm's length and stick out your thumb, the width of the thumb is roughly two degrees. Superimpose that thumb on a movie theater screen and you get an idea of how little you see clearly. Beyond that, things get progressively blurry. True, you do see some things through your peripheral vision, which is why movies like March of the Penguins are popular on wide-screen formats like IMAX theaters. But what you gather through this peripheral vision, says Simons, is broad, blurry information. "You're not going to see the details of the penguins."
The details we do notice depend, to a degree, on how we define ourselves. In the door experiment, for instance, Simons and Levin found that the seven pedestrians who did notice the change had something in common: they were all students of roughly the same age as the "stranger" they encountered. In one sense, this finding wasn't surprising. Social psychologists have shown that we often treat members of our own social group differently from how we treat members of other groups. Black people encountering white people (or vice versa) may behave differently than when they encounter someone of their own group; ditto for rich people encountering poor people, old versus young, and men versus women. Nonetheless, wondered Simons and Levin, would those differences in the way we behave toward others extend to the way we see others?
To answer that question, they repeated the door experiment, using the same "strangers" they had initially used. Only this time the strangers weren't dressed casually, as students would be; they were dressed as construction workers, complete with hard hats. And this time, they approached only people of their own age. In all, the "construction workers" encountered twelve pedestrians. Of those twelve, only four reported noticing the switch when the door came through. Putting the experimenters in construction clothes, it seemed, had been enough to change the way they were seen by students. Rather than being seen as individuals--as they had been when they were dressed as students--the experimenters were now seen as members of another group.
One of the pedestrians who had failed to detect the change when the door was brought through said as much when she was told of the experiment and interviewed afterward. She said she had seen only a "construction worker" and had not really noticed the individual; that is, she had quickly categorized him as a construction worker and hadn't noted those details--like his hair or his eyes or his smile--that would allow her to see him as an individual. Instead, she had formed a representation of the category--a stereotype. In the process, she traded the visual details of the scene for a more abstract understanding of its meaning; she had skimmed.
From the Hardcover edition.