FIGURE 4-1
Asch experiment
The subject becomes intensely aware that the entire group is awaiting his answer. He thinks, I don’t see how they can answer so quickly. I thought it was Three, but I can see how One might be the same length. My eyes must be going bad. Feebly, he states, “One.” The fidgety guy lights up another cigarette and takes a long drag, getting immense enjoyment from the drama playing out in the seat next to him.
It does not get any better for the subject. Of the eighteen trials, the group is unanimously wrong on twelve. The unwitting subject, despite what his eyes see, goes along with the group 100 percent of the time.
After the experiment, everyone but the true subject is dismissed. Asch pulls him aside and asks, “How often did you answer as the others did, against your own first choice?” This was a standard question asked of all the subjects. Invariably, even the most conformist of subjects underestimated the number of times that he went along with the herd.
The subject replies, “Possibly as many as one-fourth or one-third.” The subject thinks about his estimate and continues, “Mostly I wasn’t sure, I was undecided. Plus, some of the choices were difficult.”
In fact, the choices were not difficult. It was the group pressure to give the wrong answer that made the choices seem difficult.
Although most subjects were not totally conformist like the one just described, the amount of conformity was, nevertheless, shocking. Without a group giving wrong answers, 95 percent of subjects performed without a single error. But with the group, only one-fourth of the subjects were able to maintain this perfect performance. Most subjects caved to group pressure about one-third of the time.
Asch’s debriefing procedure revealed that most subjects had some awareness of what they were doing, although as just illustrated, most subjects underestimated by a large degree the number of times they went along with the group. Some just accepted the group’s judgment as an indication that their own perceptions were wrong. One particularly conformist subject stated, “I just wondered what was wrong with me.”
Another said, “There were so many against me that I thought I must be wrong.”
Others gave no evidence that they were even aware of the fact that they were wrong.
See as I Say
Asch was an iconoclast himself, single-handedly creating the field of social psychology. As a Polish Jew conducting experiments in the years following WWII, Asch aimed to understand how millions of Germans could blithely follow the Nazi ideology of extermination. Asch’s approach, which stands as one of the most influential psychology experiments of the twentieth century and has been replicated hundreds of times, showed that even when you strip away all the ambiguity of what an individual sees, and there is no possibility of personal gain or reprisal, people will still go along with the group.
The social psychologists who followed Asch favored this explanation of conformity: we know what we see, and we know right from wrong, but with enough social pressure, we cave in to the fear of standing alone. Ironically, this explanation of conformity contains a certain heroic element that accounts for its persistence. If we grant that we are all a bit reticent at times to stand up for our personal opinions, this leaves the door open to act as individuals when we choose. It is a noble grasp for free will. But—and this is the kicker—we must be brave enough. This was Asch’s point. Even in a neutral laboratory setting, most people are not that brave.
Now you may think, Surely, I would be brave enough to stand my ground. True, not everyone in Asch’s experiment went along with the group. But even if some are not that brave, they might think that they could decide, for whatever reason, whether to go along or not. But what if that is wrong, and we do not have as much free will as we’d like to think? What if groups of people change how we see the world? Then we are dealing with a much more pernicious form of conformity: a form of conformity we might not even be aware of and one that dooms the would-be iconoclast before he even knows it.
The extent to which perception can be altered by fear is a question that has lingered over social psychology. It is a difficult question to answer, for if a person’s perception were truly altered, he might not even know it. Asch realized that at least two distinct mental processes go into making a perceptual judgment like his line task. The first is perception itself. As we saw in the first chapter, perception is shaped not only by what the eyes transmit but by an individual’s expectation of what they are seeing. The second process—judgment—is a type of decision making. In Asch’s experiment, the decision was simple: picking the two lines that were the same length. It is important to note that these are distinct cognitive processes and are potentially mediated by different circuits in the brain.
Ever since Asch, most social psychologists have assumed that conformity is exerted at the decision-making stage in a sort of spineless capitulation to the majority. Nevertheless, hints of perceptual shifts can be found throughout the literature on social conformity. Even Asch reported several subjects who marched along with the group and yet seemed blithely unaware that they were wrong. Their unawareness suggested, but by no means proved, that their perception had been altered. The traditional entrée into a person’s perception is to ask him or her—either directly, or indirectly through experiments. But no matter the method, perception has never been fully separated from the judgment process. At least until fMRI came along.
Although neuroscientists have been using fMRI to systematically map out the neurobiology of processes such as memory, emotion, attention, and perception, nobody had thought to use fMRI to answer the looming question of whether other people can change what you see. To answer this question, in 2005 my research group tackled the problem of how fear might change perception in a modern variation of Asch’s experiment.10
The logic of our experiment was simple. If other people change what you see (or what you think you see), then fMRI should detect changes in perceptual regions of the brain. If, on the other hand, conformity occurs at the decision-making level, we should see changes in decision-making regions. As we saw in the first chapter, visual information originating from the eyes first hits the brain in the back of the head in the occipital cortex. The information then flows forward, splitting into two paths—one through the high road in the parietal cortex, the other through the low road in the temporal lobe—and they meet again in the frontal lobes. Because perception occurs largely in the first stages of this process, in the back of the brain, we hoped that we could separate the perceptual side of conformity from the decision-making side.
No matter how it turned out, the answer would have wide-ranging implications for how we, as individuals, make decisions in groups. Take, for example, the democratic institutions that have been erected in the last two centuries. They all depend on self-determination. In the Bill of Rights, we protect the liberties of individuals. In return, there is an implicit contract that requires everyone to participate in the governance of society. This includes voting for representatives, adherence to the rule of law, trial by jury, and the conduct of business according to accepted practices. We allow for differences of opinions, but these differences are resolved through institutions such as voting. Beneath all these institutions lies the assumption that individuals “call it as they see it.”
Like Asch, we hired actors to play the roles of experimental subjects. When the real subject arrived for the experiment, he saw that four other people were participating. At least he thought they were participating like him. As a group, everyone was told they would be doing a task of visual perception on a computer. Everyone would have their own computer, and everyone would see the same thing. In addition, each participant would be able to see each of the others’ answers.
The task was actually quite simple. Each trial began with two abstract three-dimensional shapes appearing on the screen. The shapes were rotated with respect to each other. As in Asch’s experiment, all the subject had to do was decide whether the two shapes were the same or different
. The kicker was that, unknown to the real subject, the actors had been given instructions to answer incorrectly. Figure 4-2 shows what the subject saw.
The actors’ faces appeared on the right of the screen, along with the answers that each of them registered. In this example, the shapes are actually the same because they can be mentally rotated to match each other. Under these circumstances, many people think their eyes are playing tricks on them.
Of course, nothing was wrong with their eyes. The visual task, although a little harder than Asch’s, was not as difficult as most of the participants thought. When doing the mental rotation task by themselves, the participants arrived at the correct answer 86 percent of the time. But when the group gave the wrong judgment, the rate of correct answers dropped to 59 percent—statistically no better than if they had flipped a coin.
FIGURE 4-2
Shape rotation experiment
Source: Reprinted from Gregory S. Berns et al., “Neurobiological Correlates of Social Conformity and Independence During Mental Rotation,” Biological Psychiatry 58, no. 3 (2005): 247, with permission from Elsevier.
We debriefed our subjects, and, like Asch, found a wide range of insight into what had happened. Some of our subjects stuck to their own intuitions and were not swayed by the group. Others went along with them almost 100 percent of the time. Most lay somewhere between these two extremes. Nobody had strong recollections of why they did what they did and just vaguely remembered going with the group sometimes and not others.
The fMRI data told the story.
The mental rotation task by itself caused a network of specific brain regions to come online. Visual processing regions, located in the back of the head, were highly active while the person stared at the screen, examining the myriad facets of the shapes. The visual representations get reassembled in the parietal cortex and the temporal lobe. Within these two regions, the brain decides what it’s seeing (the low road) and where it’s located (the high road). Our task of mental rotation, because it required all these elements to work in coordination, resulted in strong activation in the visual, parietal, and temporal regions.
You would not normally think of conformity as a visual process, but that is precisely what we found. The group altered the patterns of activity in the visual and parietal regions of the subjects’ brains when the subjects went along with them incorrectly. When a subject capitulated to the group, and the group was wrong, we observed more activity in the parietal cortex, as if it were working harder. A plausible explanation is that the group’s wrong answers imposed a “virtual” image in the subject’s mind. In the case of conformity, this virtual image beat out the image originating from the subject’s own eyes, causing the subject to disregard her own perceptions and accept the group’s. This didn’t happen all the time, but when it did, the shift in patterns in brain activation was striking. Even more interesting was the fact that we didn’t find nearly as striking a change the frontal lobes. If anything, there was a slight decrease in activity when the subject conformed, suggesting that the group’s answers took some of the load off the decision-making process in the frontal lobe.
Even when the subjects stood their ground and gave the correct answer in the face of a unanimously wrong group we found changes in brain activity. Not in the perceptual regions, in this case. Instead, nonconformity went along with increased activity in an almond-sized region of the brain called the amygdala. Recall from the last chapter that the amygdala has direct connections with the arousal system of the brain, notably the hypothalamus. When the amygdala fires, a cascade of neural events is unleashed that prepares the body for immediate action. It is the first step in the “fight-or-flight” system, but the end result of amygdala activation is a rise in blood pressure and heart rate, more sweating, and rapid breathing. Lots of things trigger the amygdala, but fear is, by far, the most effective. Its activation during nonconformity underscored the unpleasant nature of standing alone—even when the individual had no recollection of it. In many people the brain would rather avoid activating the fear system and just change perception to conform with the social norm.
A Lesson in Conquering Fear
Martin Luther King Jr., perhaps the greatest iconoclast of the civil rights movement, knew firsthand the damaging effects of fear on perception. By championing the rights of blacks, he immediately incurred the wrath of the Southern whites who surrounded him in Atlanta. King and others were pummeled with intimidation, tactics designed to instill fear in blacks. Consider James Meredith, who in 1962 became the first black to enroll in the University of Mississippi, sparking riots across the state. After graduating, Meredith organized the March Against Fear from Memphis, Tennessee, to Jackson, Mississippi, only to be shot by a sniper along the way. But in 1963, in the March on Washington, King really laid out his philosophy in his “I have a dream” speech.
Taking cues from Gandhi, King adopted a philosophy of nonviolent civil disobedience. At its heart, the tenet of nonviolence was aimed directly at conquering the damaging effects of fear. It targeted blacks’ fear of white retaliation by showing how peaceful protest, in large numbers, provided a relatively safe haven for effecting social change. Rather than standing alone, the strategy of nonviolence played into the safety in numbers that is so deeply wired into the human brain. More important, however, was King’s deep-seated conviction that nonviolence was the only way to garner public support from whites. Not all the black leaders agreed with this approach. Notably, Malcolm X advocated for more direct tactics of confrontation.
King’s strategy of nonviolence came to a head on March 7, 1965, when five hundred supporters started marching out of Selma, Alabama, to protest the intimidation tactics used by whites to prevent blacks from registering to vote. The governor of Alabama, George Wallace, declared the march a threat to public safety and ordered the police to take action against the marchers. With television cameras rolling, the nation witnessed the police attack with billy clubs and tear gas on the peaceful demonstrators. Several of the marchers were hospitalized, leading the marchers to call it “Bloody Sunday.” Two days later, King organized a second march but turned back before crossing the bridge into Montgomery, avoiding another violent confrontation. Eventually, his strategy paid off. A week later, a federal judge ruled that the state of Alabama did not have a right to block the peaceful demonstration. The marchers finally reached the capital on March 24. Several months later, Lyndon Johnson signed into law the Voting Rights Act of 1965, which outlawed any qualification tests to vote.
King well understood how fear was damaging the perception of blacks, both of themselves and by whites. As he said in his Nobel Peace Prize acceptance speech: “Nonviolence has also meant that my people in the agonizing struggles of recent years have taken suffering upon themselves instead of inflicting it on others. It has meant, as I said, that we are no longer afraid and cowed. But in some substantial degree it has meant that we do not want to instill fear in others or into the society of which we are a part.”11
King saw nonviolence as the only way to eliminate the damaging effects of fear: “It is the method which seeks to implement the just law by appealing to the conscience of the great decent majority who through blindness, fear, pride, and irrationality have allowed their consciences to sleep.”12
We begin to see clues about how the individual must invoke conscious, rational thought processes to control fear. It is a theme to which I will return at the end of the chapter. But before we get to these strategies, it helps to understand where the power of intimidation comes from.
The Law of Large Numbers
It may seem counterintuitive that the human brain is so susceptible to the opinions of others that it is willing to disregard its own visual inputs, but viewed from a statistical perspective, this biological capitulation makes perfect sense. As we saw in the first chapter, the information that the eyes transmit to the brain does not uniquely determine what the individual perceives. The brain must make educated guesses to construct a visual perce
pt based on the context and the individual’s past experience. The evolutionary theory of perception, coupled with the efficiency of the brain, means that perception is a statistical process. For any given visual image transmitted by the eyes, the brain must choose one of several possible interpretations. Efficiency dictates that the brain will pick the most likely interpretation for what it is seeing.
The way in which an individual categorizes objects markedly influences his perception. We have considered this process to be a product of experience, but there is another, even more potent source of categorization that affects perception: other people.
In both the Asch experiment and our subsequent version with fMRI, the subjects made perceptual judgments that were seemingly straightforward. The fact that individuals were swayed by group opinion raises the ominous conclusion that they would be even more susceptible to group influence with judgments that were more subjective. What if we are to consider questions in which the answers are unknown, such as who will win the World Series this year, or at what level will the Dow Jones Industrial Average close next month? While everyone may have an opinion about these questions, their confidence in their answer may vary widely from one person to another. In such situations, it makes sense to look toward other individuals to see what they think. It is a fact that other people are more likely be correct than any given individual. The reason comes from the statistics of aggregating information.
The problem has much to do with the Asch effect and is not too different from guessing how many jelly beans are in a jar. In the classic bean jar game, you buy a guess for $1, and the person guessing the closest to the actual amount wins the pot of money and the jar of beans. If you gather up enough people and record their guesses, the distribution of guesses will be similar to a bell-shaped curve.13 Even more important, if the participants don’t share information or disclose their guesses, then the average of the guesses will be very close to the true number of beans in the jar. Typically, the average of the guesses will be better than 95 percent of the participants’. In other words, the average of a group of independent observers is better than any individual, and frequently the average is as good as, or better than, even the best individual in the group. The more diverse the group of participants, the better the group’s average. The only thing that matters is that the participants act independently of one another.14
Iconoclast: A Neuroscientist Reveals How to Think Differently Page 10