Connected: The Amazing Power of Social Networks and How They Shape Our Lives

The powerful effect of social networks on individual behaviors and outcomes suggests that people do not have complete control over their own choices. Interpersonal influence in social networks therefore raises moral questions. Our connections to others affect our capacity for free will. How much blame does Giacomo in Corsica deserve for his actions, and how much credit does Dan Lavis in Ontario deserve for his? If they acted merely as links in a chain, how can we understand their freedom to choose their actions at all?

Some scholars explain collective human behavior by studying the choices and actions of individuals. Others dispense with individuals and focus exclusively on groups formed by social class, race, or political party affiliation, each with collective identities that cause people in these groups to mysteriously and magically act in concert. The science of social networks provides a distinct way of seeing the world because it is about individuals and groups, and about how the former actually become the latter.

If we want to understand how society works, we need to fill in the missing links between individuals. We need to understand how interconnections and interactions between people give rise to wholly new aspects of human experience that are not present in the individuals themselves. If we do not understand social networks, we cannot hope to fully understand either ourselves or the world we inhabit.

CHAPTER 2 When You Smile, the World Smiles with You (#ua458f89f-1c8d-59a6-b438-14bafc5eada5)

A strange thing happened in Tanzania in 1962. At a mission boarding school for girls near Lake Victoria in the Bukoba District, there was an epidemic of laughter. And this was not just a few schoolgirls sharing a joke. An irresistible desire to laugh broke out and spread from person to person until more than one thousand people were affected.

The affliction had an abrupt onset, and the initial bout of laughter lasted between a few minutes and a few hours in those affected. This was followed by a period of normal behavior, then typically a few relapses over the course of up to sixteen days. In what was to be a clue about the real nature of this epidemic, the victims often described feeling restless and fearful, despite their laughter.

The physicians who first investigated and reported on the out-break—Dr. Rankin, a faculty member at Makerere University, and Dr. Philip, the medical officer of the Bukoba District—were extremely thorough.

(#litres_trial_promo) They found that each new patient had recent contact with another person suffering from the malady. They were able to observe that the incubation period between contact and onset of symptoms ranged from a few hours to a few days. Thankfully, as they intoned without irony, “no fatal cases have been reported.” Afflicted persons recovered fully.

The epidemic began on January 30, 1962, when three girls aged twelve to eighteen started laughing uncontrollably. It spread rapidly, and soon most people at the school had a serious case of the giggles. By March 18, ninety-five of the 159 pupils were affected, and the school was forced to close. The pupils went home to their villages and towns. Ten days later, the uncontrollable laughter broke out in the village of Nshamba, fifty-five miles away, where some of the students had gone. A total of 217 people were affected. Other girls returned to their village near the Ramanshenye Girls’ Middle School, and the epidemic spread to this school in mid-June. It too was forced to close when forty-eight of 154 students were stricken with uncontrollable laughter. Another outbreak occurred in the village of Kanyangereka on June 18, again when a girl went home. The outbreak started with her immediate family and spread to two nearby boys’ schools, and those schools were also forced to close. After a few months, the epidemic petered out.

Rankin and Philip looked hard for biological causes for the epidemic. They performed physical examinations and lab studies on the patients, did spinal taps, examined the food supply for toxins, and ascertained that there was no prior record of a similar epidemic in the region. The villagers themselves did not know what to make of it. In Bukoba, where the illness aroused great interest, there was the “belief that the atmosphere had been poisoned as a result of the atom bomb explosions.” Others described it as a kind of “spreading madness” or “endwara yokusheka,” which means simply, “the illness of laughing.”

As the villagers and the scientists investigating this outbreak realized, the epidemic was no laughing matter. It did not involve the spread of real happiness and joy—though this can happen too, albeit not in quite the same way. Rather, the outbreak was a case of epidemic hysteria, a condition that takes advantage of a deep-rooted tendency of human beings to exhibit emotional contagion. Emotions of all sorts, joyful or otherwise, can spread between pairs of people and among larger groups. Consequently, emotions have a collective and not just an individual origin. How you feel depends on how those to whom you are closely and distantly connected feel.

Our Ancestors Had Feelings

We all have emotions. And they consist of several elements. First, we usually have a conscious awareness of our emotions: when we are happy, we know it. Second, emotions typically affect our physical state: we show how we feel on our faces, in our voices, even in our posture; given the role emotions play in social networks, these physical manifestations are especially important. Third, emotions are associated with specific neurophysiological activity; if you are shown a scary picture, the flow of blood to structures deep in your brain instantly changes. Finally, emotions are associated with visible behaviors, like laughing, crying, or shrieking.

(#litres_trial_promo)

Experiments have demonstrated that people can “catch” emotional states they observe in others over time frames ranging from seconds to weeks.

(#litres_trial_promo) When college freshmen are randomly assigned to live with mildly depressed roommates, they become increasingly depressed over a three-month period.

(#litres_trial_promo) Emotional contagion can even take place between strangers, after just ephemeral contact. When waiters are trained to provide “service with a smile,” their customers report feeling more satisfied, and they leave better tips.

(#litres_trial_promo) People’s emotions and moods are affected by the emotional states of the people they interact with. Why and how does this happen?

We might consider another question first: Why aren’t emotions merely internal states? Why don’t we just have our own private feelings? Having feelings is surely evolutionarily advantageous to us. For example, the ability to feel startled is probably good for us in situations where we need to react quickly to survive. But we do not just feel startled, we show that we are startled. We jump or shriek or curse or clench, and these actions do not go unnoticed. They are copied by others.

Given the organization of early hominids into social groups, the spread of emotions served an evolutionarily adaptive purpose.

(#litres_trial_promo) Early humans had to rely on one another for survival. Their interactions with the physical environment (weather, landscape, predators) were modulated or affected by their interactions with their social environment. Humans bonded with others in order to face the world more effectively, and mechanisms evolved to support this bonding, most obviously verbal communication but also emotional mimicry. The development of emotions in humans, the display of emotions, and the ability to read the emotions of others helped coordinate group activity by three means: facilitating interpersonal bonds, synchronizing behavior, and communicating information.

Emotions and emotional contagion probably first arose to facilitate mother-infant pair bonding and then evolved to extend to kin members and ultimately to nonkin members. Emotional contagion fosters interaction synchrony. At the level of mother-child pairs, emotional contagion may have prompted mothers to be more attentive to and protective of their babies when their babies needed attention. Indeed, we are sadder when our family members are sad than when strangers are sad. There is an advantage in coordinating our moods with those to whom we are related.

Eventually this type of synchrony in mood or activity may have been beneficial for larger group activities, such as warding off enemies or hunting prey. If you are trying to coordinate a hunting party, it helps if members of the group are all upbeat and fired up. Conversely, if you are part of a group and someone in it appears afraid, perhaps that person has seen a predator that you have not seen. Quickly adopting his emotional state can enhance your prospects for survival. Indeed, it is thought that positive emotions may work especially well to increase group cohesiveness (“I’m happy; stay with me”) and that negative emotions may work well as communication devices (“I smell smoke; I’m scared”).

Emotions may be a quicker way to convey information about the environment and its relative safety or danger than other forms of communication, and it seems certain that emotions preceded language. What emotions lack in specificity compared to oral language, they may make up for in speed. You can tell whether your spouse is mad at you very quickly, but having her explain it to you may take a good deal more time (especially if she insists that you guess why she is mad before she tells you). You can walk through the door at home at the end of the day and immediately know whether the environment is safe or dangerous, and that is quite a trick our ancestors bequeathed us.

Of course, rapidly coordinated emotions are not always a good thing. If you come home and are in a bad mood, your partner will often detect it long before you resort to the more laborious process of explaining why you are in a bad mood. And before you have a chance to explain, she might already have caught your bad mood, which may lead to an argument and a downward spiral.

Emotional Contagion

Emotions spread from person to person because of two features of human interaction: we are biologically hardwired to mimic others outwardly, and in mimicking their outward displays, we come to adopt their inward states. If your friend feels happy, she smiles, you smile, and in the act of smiling you also come to feel happy. In bars and bedrooms, at work and on the street, everywhere people interact, we tend to synchronize our facial expressions, vocalizations, and postures unconsciously and rapidly, and as a result we also meld our emotional states.

Nowhere do we show our emotions more than on our faces. It is not difficult to explain why our facial expressions change in response to environmental stimuli or how this may be evolutionarily adaptive. Recent research, for example, has provided insight into how two facial expressions, fear and disgust, moderate our reception of sensations coming from the outside world.

(#litres_trial_promo) When we are terrified, our eyes widen and our nostrils flare to help us see and smell more of our surroundings, just as the ears of a dog perk up when it hears something interesting. Similarly, when we are disgusted, such as by an offensive odor, our noses wrinkle and our eyes narrow to reduce the impact. Air intake increases when we are afraid and decreases when we are disgusted.

Yet, facial expressions appear to have evolved not just to modify our experience of the world as individuals but as a way to communicate with others. Over time, this aspect of facial expressions probably eclipsed their original role. Such changes happen often in evolution. Feathers may have arisen merely to insulate the bodies of prehistoric reptiles, but they wound up contributing to a different and more important advantage, the ability to fly.

We developed an ability to read the facial expressions of others. Hence, we benefit when our own faces are contorted in disgust and by being able to notice whether others’ faces are contorted in disgust. Humans have an extraordinary knack for detecting even small changes in facial expressions. This ability is localized in a particular area of the brain and can even be lost, a condition tongue-twistingly known as prosopagnosia. Reading the expressions of others was probably a key step on the way toward synchronizing feelings and developing the emotional empathy that underlies the process of emotional contagion.

Even as early as 1759, it was apparent to founding economist and philosopher Adam Smith that conscious thought was one way we could feel for others and hence feel like others: “Though our brother is upon the rack…by the imagination we place ourselves in his situation, we conceive ourselves enduring all the same torments, we enter as it were into his body, and become in some measure the same person with him, and thence form some idea of his sensations, and even feel something which, though weaker in degree, is not altogether unlike them.”

(#litres_trial_promo)

However, emotions spread in ways beyond simply reading faces and thinking about the experiences of others. There is actually a more primitive, less deliberative process of emotional contagion, a kind of instinctive empathy. People imitate the facial expressions of others, then, as a direct result, they come to feel as others do. This is called affective afference, or the facial-feedback theory, since the path of the signals is from the muscles (of the face) to the brain, rather than the more usual, efferent pathway from the brain to the muscles. The beneficial effects of facial expressions on a person’s mood are among the reasons, for example, that telephone operators are trained to smile when they work, even though the person at the other end of the line cannot see them. This theory also explains why it helps to smile when your heart is breaking.

One biological mechanism that makes emotions (and behaviors) contagious may be the so-called mirror neuron system in the human brain.

(#litres_trial_promo) Our brains practice doing actions we merely observe in others, as if we were doing them ourselves. If you’ve ever watched an intense fan at a game, you know what we are talking about—he twitches at every mistake, aching to give his own motor actions to the players on the field. When we see players run, jump, or kick, it is not only our visual cortex or even the part of our brain that thinks about what we are observing that is activated, but also the parts of our brain that would be activated if we ourselves were running, jumping, or kicking.

In one experiment related to emotional contagion, subjects listened to recordings of nonverbal vocal reactions communicating two positive emotions, such as amusement and triumph, and two negative emotions, such as fear and disgust. Investigators monitored the subjects’ brains for a response by placing them in a magnetic resonance imaging (MRI) machine.

(#litres_trial_promo) The subjects were told not to react to what they heard. While subjects did not visibly respond to the sounds, the MRI results showed that hearing the cues stimulated parts of the brain that command the corresponding facial expressions. It seems we are always poised to feel what others feel and to do what others do.

Emotional Stampedes

Everyone has experience with emotional contagion: we share a joke with a friend, we feel sad when a spouse cries, we rage against city hall with our neighbors, and we hug our kids tight when they’ve had a bad day. Yet one often overlooked aspect of all this sharing is that emotions spread not only to our friends but to our friends’ friends and beyond—even when we are not present. We are like a herd of buffalo quietly grazing on the plain until one of our neighbors starts to run. Then we start to run, and others start to run, and suddenly, mysteriously, the whole herd is barreling forward.

Epidemics of emotional states have been reported for centuries. They just have not involved laughter like the Bukoba outbreak. When emotions spread from person to person and affect large numbers of people, it is now called mass psychogenic illness (MPI) rather than the old-fashioned and more poetic epidemic hysteria. MPI is a specifically social phenomenon involving otherwise healthy people in a psychological cascade. Like a single startled buffalo within a herd, a single emotional reaction in one person can sometimes cause many others to feel the same thing, creating an emotional stampede.

There are two main types of MPI. In the pure-anxiety type, people may feel a variety of physical symptoms, including abdominal pain, headache, fainting, shortness of breath, nausea, dizziness, and so on. In the motor type, people may engage in hysterical dancing, pseudo-seizures, and—yes—laughing, though the actual feelings underlying these behaviors are fear or anxiety. Both types of MPI thus involve the same basic psychological processes.

Historical records of such phenomena date back to at least 1374, when, in close succession to the Black Death in Europe, “dancing manias” broke out. The first such manias occurred in what is now Aachen, Germany. As described by the German medical historian J. F. C. Hecker in his 1844 book The Epidemics of the Middle Ages, these consisted of people who “united by one common delusion, exhibited to the public both in the streets and in the churches the following strange spectacle. They formed circles hand in hand, and appeared to have lost all control over their senses, continued dancing, regardless of the bystanders, for hours together, in wild delirium, until at length they fell to the ground in a state of exhaustion. They then complained of extreme oppression, and groaned as if in the agonies of death.”

(#litres_trial_promo) These people were obviously no happier to be dancing than the African schoolgirls were to be laughing.

In a bygone era, demons and witchcraft were often seen as causes of these symptoms, but today toxic chemicals and environmental contamination are the triggers subjects typically identify. Yet, while toxins do cause some outbreaks of physical illness, they do not cause outbreaks of MPI. The source of the problem, as well as the mechanism of transmission, is psychological. Individuals afflicted in these outbreaks, and many observers, are often reluctant to ascribe the symptoms to a psychological source, however.

A relatively recent example of MPI occurred at the Warren County High School in McMinnville, Tennessee. At the time, the school had 1,825 students and 140 staff members. On November 12, 1998, a teacher believed she smelled gasoline, which caused her to complain of headache, shortness of breath, dizziness, and nausea. Seeing her response, some of her students soon developed similar symptoms. As the classroom was being emptied, other students, observing what was happening, began to report feeling unwell too. A schoolwide fire alarm was activated, and the school was evacuated. The teacher and several students were transported by ambulance to a nearby hospital, in full view of other students and teachers who were outside because of the alarm. Large numbers of police, firefighters, and emergency medical personnel from three counties responded. A total of one hundred people went to the hospital that day, and thirty-eight were admitted. Classes were canceled.

The school was closed for four days. It was inspected by the fire department, the gas company, and state officials from the Occupational Safety and Health Administration (OSHA), but no problems were identified. After the school had been deemed perfectly safe, the students and the teachers were allowed to return. Unfortunately many still smelled odors, and on November 17, seventy-one people were stricken. Ambulances were again called, and the school was evacuated and then closed.

The school’s principal was fed up. In a “no more Mr. Nice Guy” move, he decided to call several government agencies, including the famed Epidemic Intelligence Service of the Centers for Disease Control (CDC). Also involved were the federal Environmental Protection Agency, the Agency for Toxic Substances and Disease Registry, the National Institute for Occupational Safety and Health, OSHA, the Tennessee Department of Health, the Tennessee Department of Agriculture, and numerous other local emergency organizations and personnel. The investigation was extremely thorough. Aerial surveillance identified potential environmental sources of contamination; personnel explored caves in the vicinity of the school; the school’s air-handling, plumbing, and structural systems were thoroughly checked; core samples were drilled from the grounds around the school; and air samples (including from the days of the outbreak) and water and waste samples were tested. The air was evaluated with an astonishing array of technology, including colorimetric tubes, flameionization detectors, photoionization detectors, radiation meters, and combustible-gas indicators.

Two years later, a New England Journal of Medicine article described the extensive examination of possible environmental causes for the illness and reported the results of the investigation by the CDC. In the end, like Rankin and Philip studying the African laughter epidemic, the investigators concluded that psychogenic factors were to blame. They found that the illness was associated with directly observing another ill person during the outbreak and with being female.

(#litres_trial_promo) The diagnosis was epidemic hysteria.