The Optimism Bias: A Tour of the Irrationally Positive Brain Page 2
Illusions of the Human Brain
January 3, 2004, Sharm el-Sheikh. One hundred and forty-eight passengers and crew board Flash Airlines Flight 604 bound for Paris via Cairo. The Boeing 737-300 takes off at exactly 4:44 a.m. Two minutes later, it disappears from the radar.
Sharm el-Sheikh is located on the southern tip of the Sinai Peninsula. It is a tourist destination because of its year-round warm weather, beautiful beaches, and marvelous snorkeling and diving. The majority of passengers on Flight 604 are French tourists escaping the European winter to spend their Christmas vacation near the Red Sea. Entire families are on board Flight 604, on their way back home.1
The crew is largely Egyptian. The pilot, Khadr Abdullah, is a decorated war hero, because of his performance flying the MiG-21 in the Egyptian air force during the Yom Kippur War. He has 7,444 flying hours under his belt, although only 474 of those are on the Boeing 737 he is piloting on this day.2
According to its designated route, the aircraft should have ascended for a short while after takeoff and then turned left, heading toward Cairo. Instead, less than a minute into the flight, the plane turns right and quickly assumes a dangerous angle. Flying completely on its side, the jet begins spiraling downward toward the Red Sea. Just before impact, the pilot appears to gain control over the now upside-down plane, but it is too late.3 Flight 604 crashes into the water moments after takeoff. There are no survivors.
At first, the authorities suspect a bomb had been planted on the plane by terrorists. This hypothesis arises because no distress signal was sent from the aircraft. However, when the sun comes up and pieces of the jet are discovered, it becomes apparent this theory is wrong. The pieces of the plane are detected close together, and there are not many of them.4 This suggests that when the plane hit the water, it was intact, rather than having exploded in midair, which would have resulted in many fragments scattering across the sea. What, then, caused Flight 604 to drop violently from the sky?
For the mystery to be solved, it is essential that the plane’s black box be found. The area of the sea where the plane crashed is one thousand meters deep, which makes it difficult to detect the signals emitted from the box. Furthermore, the black box’s battery will last for only thirty days; after that, the probability of finding it will be, realistically, nil. Egyptian, French, and U.S. search teams participate in the effort. Luckily, two weeks into the search, the black box is detected by a French ship.5
The information from both the data recorder and the voice cockpit log contain clues that guide the investigators in a number of different directions. No less than fifty different scenarios are identified, then ruled out one by one, on the basis of the available data. No evidence of any airplane-related malfunction or failure can be found.6 The investigators are left with a handful of scenarios, which they then try out in a plane simulator. After examining the remaining scenarios thoroughly, all but one are deemed inconsistent with the data at hand. The U.S. research team concludes that “the only scenario identified by the investigative team that explained the accident sequence of events, and was supported by the available evidence, was a scenario indicating that the captain experienced spatial disorientation.”7
During spatial disorientation, also known as vertigo, a pilot is unable to detect the position of the aircraft relative to the ground. This usually happens when no visual cues are available, such as when the plane is flying in a dense cloud or in pitch-darkness over the ocean. The pilot may be convinced that he is flying straight when, in fact, the plane is in a banked turn, or when coming out of a level turn, he may feel he is diving. Trying to correct the (false) position of the aircraft only makes matters worse. During a rapid deceleration, a pilot sometimes feels the plane is facing downward. To rectify this illusion, the pilot may then pull up the nose of the plane, which often leads the aircraft to fall into a catastrophic spin known, for obvious reasons, as the “graveyard spin.” The graveyard spin is what seems to have happened to the Piper plane piloted by John F. Kennedy, Jr. It crashed into the Atlantic Ocean on July 16, 1999, after Kennedy suffered spatial disorientation while flying at night in bad weather en route to Martha’s Vineyard.8
How can a pilot be convinced that he is flying up when he is actually heading down? Or that he is moving straight ahead when he is, in fact, in a dangerous bank? The human brain’s navigational system has evolved to detect our movement on earth, not in the sky. It calculates our position by comparing signals from the inner ear (which has tubes of liquid that shift when we move) to the fixed sensation of gravity that points down to the center of the earth.9 This system works extremely well when we are on the ground, as it was developed to function in this context (our ancestors did not spend much of their time airborne). However, in a speeding plane in midair, the system gets confused. Our brain interprets irregular signals, such as angular accelerations or centrifugal force, as the normal force of gravity. As a result, it miscalculates our position in relation to the earth. The liquid in the inner ear does not quite catch up with the fast rate of the plane’s directional change, causing false signals to be transmitted to the brain. When our eyes cannot confirm directional change, either, because visual cues are lacking, the change in position can go undetected. The result is that the plane can be flying on its side, while the pilot is utterly convinced it is parallel to the ground; he feels as if he were relaxing on his couch at home.10
Now, here is the problem: Throughout life, we have learned to rely on our brain’s navigational system to give us the correct position of our body relative to the ground. We seldom suspect it is giving us misinformation, and thus we do not normally second-guess our sense of position. At this very moment, while reading this book, you know for sure that the sky is above you and the ground is beneath. You are probably right. Even in the dead of night, with no visual cues, you can still tell with certainty which way is up.
So the first thing a pilot must learn is that although he may feel 100 percent certain that his plane is going in a specific direction, this may be an illusion. This is not an easy concept to grasp. An illusion is an illusion because we perceive it at face value—as reality. “The most difficult adjustment that you must make as you acquire flying skill is a willingness to believe that, under certain conditions, your senses can be wrong,” says one student pilot training guide.11
The good news is that there is a solution for a pilot’s vertigo; it is the plane’s navigational system. This is why, thankfully, most planes do not end up in the ocean, although almost every pilot has had a brush with vertigo at least once in his career. If a pilot is familiar with the plane’s navigational system and knows he must rely on it even when it communicates information that contradicts that conveyed by his brain, he will avoid tragedy. The problem in the case of John F. Kennedy, Jr., was that he was not certified in instrument flight rules (IFR), only in visual flight rules (VFR). He was not trained to fly in conditions that did not allow for the use of visual cues—conditions in which one must rely on instruments alone to navigate, such as that dark, stormy night his plane crashed.12
Khadr Abdullah, the experienced pilot on the Flash jet, was certified in both IFR and VFR. However, on that fatal day, his brain seemed to trick him into believing he was flying level as he guided the plane into a dangerous right overbank nose-down. How could this happen to an experienced pilot? The U.S. investigative team suggests the following scenario: Shortly after takeoff, the plane was over the Red Sea at night; thus, no visual cues (such as ground lights) were available to indicate ground or sea level. Second, the plane’s change in spatial position was so gradual that it could not be picked up accurately by the crew’s vestibular systems. In fact, once the angle had greatly increased, the pilot may have perceived that the plane was turning slightly left rather than dangerously right.13 This scenario is supported by the recordings from the cockpit voice tape. On the tape, the first officer can be heard informing the pilot that the plane is turning right. In a surprised tone, the pilot is then heard responding, “Right? How right?” indicating that he has detected a mismatch between the information provided by the first officer and his own perception.14
Because of the lack of visual cues and the gradual shift in position, the only way the pilot could have accurately perceived the relative location of the plane to the ground was by constantly monitoring the plane’s navigational system. There is evidence, however, that the flight instruments were not being monitored constantly. At the time the plane was entering a right bank, it was allowed to travel at thirty-five knots below the required airspeed and was climbing over the standard pitch. It appears the pilot did not detect these changes because his attention was focused on engaging and disengaging the autopilot.15 Without monitoring the plane’s navigational system, the pilot had only his brain’s navigational system to rely on, and that was receiving misinformation from his inner ear and no information from his eyes—resulting in disaster.
Visual Illusions
Most of us have never flown a plane, so we are unfamiliar with the experience of vertigo that can result. Unknowingly, however, we are constant victims of the illusions created by our brain. Take a look at Figure 1, which portrays two squares—A and B. Which one is lighter? You probably see the same as I do: B is lighter. Right?
Figure 1. Checker Shadow Illusion
Edward H. Adelson, 1995.
Wrong. The squares are exactly the same color; I assure you that they are identical. So why do we perceive them as different shades of gray? It is a visual illusion created by our brain. Our visual system believes square B is in shadow, while square A is in light. They are not. The image was created using Photoshop. The squares convey the same amount of light, but our brain corrects for what it assumes to be the position of the squares (in shadow or in light) and concludes that square B must be lighter.16 The result? Square A looks darker than square B. Our subjective perception of reality differs from objective reality.
Although in this instance our brain has given us faulty information (and in a very convincing manner, too), it has done so for good reason. Our visual system was not built to interpret a cleverly constructed Photoshop image that does not follow physical rules. Like our navigational system, our visual system was developed to interpret the world it would encounter most frequently. To do so, it developed some shortcuts, some assumptions about the world, which it uses to function. These allow our brain to work efficiently in almost all situations. However, it does leave room for errors when those assumptions are not met.
Let’s explore another example. Look at Figure 2.
Figure 2. Smiling Girl
Adapted from P. Rotshtein, R. Malach, U. Hadar, M. Graif, and T. Hendler, “Feeling or Features: Different Sensitivity to Emotion in Higher-Order Visual Cortex and Amygdala,” Neuron 32 (2001): 747–57.
What do you see? An upside-down photo of a girl smiling. Okay, now rotate the book 180 degrees so you can see the photo the right way up. What do you see now? Suddenly, she is not that sweet-looking, is she? The illusion is called the Thatcher illusion, as it was first demonstrated in 1980 on a photo of former British prime minister Margaret Thatcher,17 who, to say the least, is not known for her cheerful expressions.
The illusion is created by inverting a face without inverting the mouth and eyes. Upside down, the face looks relatively normal and the expression perceived is the same as that conveyed by the original photo before it was “Thatcherized” (this is the term for inverting the face without rotating the mouth and eyes). So if the girl was originally smiling, she will be perceived as smiling after being Thatcherized. However, the Thatcherized face looks bizarre when upright, even grotesque. The mismatch between the orientation of the mouth and eyes relative to the rest of the face is easily detected.
This illusion, like many others, gives us clues as to how the brain functions, and the evolutionary constraints that guided its development. We walk around all day encountering upright faces. They are everywhere—on the street, next to us on the bus, or at the office. It is important that we accurately and efficiently recognize that a face is a face rather than, say, a football or a watermelon, because faces really should not be kicked around or split in two. It is also important that you easily distinguish between the face of your significant other and that of your boss or neighbor, as things could get quite awkward if you don’t. In fact, just being able to recognize the faces of your partner, boss, and neighbor is not enough. To get along in this world, we need to remember and distinguish thousands of faces. Luckily, most of us do so with ease, thanks to the part of the brain known as the fusiform face area (FFA), which is located in a region of the brain called the fusiform gyrus.18 The FFA is the part of our visual system that allows us to recognize that a face is a face, and to distinguish between the many faces we encounter on a daily basis. Without a functioning FFA, we may all become Prosopagnosic, which means we will be face-blind. People who suffer from lesions to their fusiform gyrus have difficulty identifying faces and may even be unable to recognize their own face. (Oliver Sacks famously wrote of such a case in his book The Man Who Mistook His Wife for a Hat.)19
Imagine living your life without knowing who’s who. True, our face recognition is not perfect. We are often approached by people who claim they have met us before but whom we are unable to recall. However, when you fetch your child from school, you usually pick out the right kid, even if he is wearing a new outfit or has just had a haircut. In fact, you do better than that. Not only are you able to detect your child in the mass of faces; you are also able to sense whether your kid had a good or bad day simply by glimpsing the expression on his face.
Humans are very good at perceiving the emotional state of others. We do so unconsciously all the time, using all sorts of clues, such as tone of voice and gait. Mostly, however, we identify the emotional states of others by perceiving their facial expressions. We know a happy expression when we see it on someone’s face; we know when someone is sad, afraid, or angry by the exact way his mouth curls and his eyes open wide or become narrow. The clues may be subtle, but we are quite good at detecting someone else’s emotional state because we have become experts at identifying facial expressions. We can do so for familiar faces, faces we have not previously encountered, faces from our own culture or a foreign one, because emotional expressions are universal.20
The capability to convey and detect emotion is critical to our existence. Take, for example, our ability to differentiate between a fearful face and an angry one. An angry face signals that the person in front of us is upset, possibly at us, and may be a threat to our survival. A fearful face signals that there is a threat somewhere in the environment; however, the person in front of us is not the source of this threat. In this case, we should quickly scan our surroundings to try to detect where the danger is coming from, so it can be avoided.
Accurate recognition of both emotional expressions and identity is vital for social communication. Most of us can recognize thousands of faces; we can easily distinguish Margaret Thatcher from Boy George (apparently, they resemble each other),21 and a frown from a grin. However, turn faces upside down and we become almost as helpless as a pilot flying in pitch-darkness without navigational instruments.
The brain is used to detect upright faces and expressions. It processes the parts of the face (eyes, nose, and mouth) in unison, as this is the most efficient way to do so. In other words, rather than identifying each part separately, the brain processes the face and its expression as a whole.22 Now, because the brain does not encounter upside-down faces very often, it has not learned to process them as effectively as upright faces. When presented with a rotated face, we seem to process its features separately, rather than in a configural manner.23
Let us turn back to the rotated face of the girl in Figure 2. Although her face was rotated, her mouth and eyes were left upright. On their own, the mouth and eyes express emotion in a normal manner. Our brain processes them separately from the rest of the face and identifies the emotional clues conveyed. We thus conclude that the person is smiling. Rotate the Thatcherized face, however, and what is perceived are eyes and mouth in a shape never seen before. The look is deformed, and our emotional reaction to the distortion is disgust and fear.
It’s not only humans who are tricked by a Thatcherized face. Monkeys are fooled, too.24 A group of researchers at Emory University Thatcherized the face of a monkey using the same technique utilized in Figure 2. They then showed a group of monkeys four photos: a photo of a standard monkey face, an inverted photo of a standard monkey face, an inverted Thatcherized monkey face (as in Figure 2), and an upright Thatcherized monkey face (the one humans find bizarre). The monkeys were not very interested in the photos of the standard monkey face—whether the image was inverted or upright, they glanced briefly at the normal face and moved on. What about the Thatcherized face? When the image was inverted (as in Figure 2), the monkeys were no more interested in the Thatcherized face than in the normal face. However, when the image was presented upright, the monkeys spent much longer looking at the Thatcherized face than at any of the other faces. The monkeys’ response indicates that they found the upright Thatcherized face as odd as we do, but, like us, they, too, were tricked into perceiving the rotated Thatcherized face as normal. If monkeys are sensitive to the Thatcher illusion, this means that the processes underlying the illusion are evolutionarily old. The brain seems to have developed a specific bias for processing upright faces long ago.
As in most illusions, learning of the illusion and its roots does not erase the illusion. Although we now know that the squares in Figure 1 are the same, we still perceive B as being lighter than A. Our knowledge does not change our perception; the illusion is still there. Similarly, a pilot may acknowledge he is in a state of vertigo, in which the information provided by the instruments does not align with his perception, and still feel that he is climbing up while heading down. The illusion, which feels very real, is dissociated from the knowledge (when available) that the perception is false.