Can we influence each other by our behavior?

2024 Author: Seth Attwood | [email protected]. Last modified: 2023-12-16 15:55

Folk wisdom "Tell me who your friend is, and I will tell you who you are" can hide more in itself than we used to think. Not only our closest friends, but also friends of friends have an influence on who we are: they help us quit smoking or contribute to the fact that we get fat, they also make us happy or lonely. True, in fairness, we ourselves also influence people whom we may not even know directly. Prepared an abridged translation of an article by journalist Clive Thompson for The New York Times, devoted to research and criticism of the theory of social connections and contagious behavior.

Eileen Belloli, 74, tries to maintain her friendships. She was born in the town of Framingham, Massachusetts, and there she met her future husband, 76-year-old Joseph. They both never left Framingham, as did many of Eileen's elementary school friends, so even 60 years later, they still get together every six weeks.

Last month, I visited the Belloli family and asked Eileen about her friends: she immediately pulled out a folder containing all the photos from her school days and class meetings. Eileen told me that every five years she helps organize a meeting and each time they manage to get a group of about 30 people together. As I leafed through the photos, I could see that Belloli and their friends had kept their health at a high level over the years. As they age, they have largely remained slender, even though many other residents of Framingham have died of obesity.

Eileen is especially proud to remain active. Perhaps her only vice was smoking: usually right after the end of the school day (Eileen worked as a biology teacher), she went to the nearest cafe, where she drank two cups of coffee and smoked two cigarettes. At the time, her addiction to cigarettes did not seem to be a problem: most of her friends also smoked. But in the late 1980s, some of them began to give up this bad habit, and pretty soon Eileen became uncomfortable holding a cigarette in her hands. She also quit smoking, and after a few years there were no people left in her circle who would continue to do this.

Photos from school meetings showed only one person whose health deteriorated markedly over the years. When he was younger, this man looked as healthy as everyone else, but every year he got bigger. He did not remain friends with his classmates, his only point of contact with them was these meetings, which he continued to attend until last year. Later it turned out that he had died.

I found this man's story especially relevant because Eileen and Joseph are involved in scientific research that may help explain his fate. The Framingham Heart Study is the world's most ambitious national heart disease project, dating back to 1948 and spanning three generations of town families.

Every four years, doctors examine every aspect of the subjects' health and assess their heart rate, weight, blood cholesterol, and more. For decades, Framingham's research has been a goldmine of information on heart disease risk factors …

… but two years ago, a couple of sociologists, Nicholas Christakis and James Fowler, used the information gathered over the years about Joseph, Eileen and several thousand of their neighbors to make a discovery of a completely different order

By analyzing Framingham's data, Christakis and Fowler said for the first time they found a solid foundation for a potentially powerful theory of social epidemiology: good behavior - such as quitting smoking, being positive, or staying lean - is transmitted from friend to friend in much the same way as speech was about infectious viruses. According to the available data, the participants in the Framingham study influenced each other's health through casual communication.

But the same was true for bad behavior: groups of friends seemed to "infect" each other with obesity, unhappiness, and smoking. It seems that good health is not only a matter of your genes and diet, but in part a result of your close proximity to other healthy people.

For decades, sociologists and philosophers have suspected that behavior could be "contagious." Back in the 1930s, the Austrian sociologist Jacob Moreno began drawing sociograms, small maps of who knows whom, and found that the form of social connections varied widely from person to person. Some were sociometric "stars" whom many chose as friends, while others were "isolated", virtually devoid of friends. In the 1940s and 1950s, some sociologists began to analyze how the shape of the social network can influence the behavior of people; others have explored how information, gossip and opinion spread within the network.

One of the pioneers of the trend was Paul Lazarsfeld, a sociologist at Columbia University, who analyzed how a commercial product became popular. Lazarsfeld argued that the rise in popularity of a product is a two-step process in which highly connected people first absorb the advertising of the product in the media and then share the product with their many friends.

Nowadays, it is customary to talk about social changes as epidemics (for example, the "obesity epidemic") and "superconnections", which interact so closely that they have a huge impact in society, almost single-handedly contributing to the emergence of certain trends.

However, in none of these case studies did scientists observe the "contagion" process in action. They, of course, reconstructed it after the fact: sociologists or marketers conducted interviews to try to reconstruct who told whom and what. But this, of course, implies a perception error: people may not remember how they were influenced or who they influenced, or they may not remember quite correctly.

Plus, studies like this have focused on small groups of people (a few hundred maximum), which means they don't necessarily reflect how contagious behavior is spreading - if it does at all - among the general public. Are "superconnectors" really important, people with the maximum number of connections? How many times does someone need to encounter a trend or behavior before "picking up" it? Of course, scientists already knew that a person can influence his closest colleague, but can this influence spread further? Despite the belief in the existence of social contamination, no one really knew how it worked.

Nicholas Christakis redefined the issue in 2000 after visiting terminally ill patients in working-class neighborhoods in Chicago. Christakis, a physician and sociologist at Harvard University, was sent to the University of Chicago and made a name for himself by studying the “widowhood effect,” the well-known propensity of spouses to die soon after the death of their partners. One of his patients was a terminally ill elderly woman with dementia who lived with her daughter, the latter acting as a nurse.

The daughter was tired of caring for her mother, and the daughter's husband fell ill due to the great stress of his wife. And then one day a friend of her husband called to Christakis's office, asking for help and explaining that he, too, felt depressed because of this situation. One woman's illness spread outward "through three degrees of separation": to the daughter, to the husband, to the friend of this man. After this incident, Christakis wondered how this phenomenon could be studied further.

In 2002, a mutual friend introduced him to James Fowler, then a graduate student in the Harvard School of Political Science. Fowler investigated the question of whether the decision to vote in an election for a particular candidate could be virally transmitted from one person to another. Christakis and Fowler agreed that social contagion was an important area of research and decided that the only way to answer the many unanswered questions was to find or collect a huge pool of data that would represent thousands of people.

At first, they thought they would do their own research, but later went on the hunt for an already existing dataset. They weren't optimistic: although there are several large surveys about adult health, medical researchers don't have the habit of thinking about social media, so they rarely ask who they know who their patients are.

And yet the Framingham study looked promising: it took more than 50 years to store data from more than 15,000 people over three generations. At least in theory, it could provide the right picture, but how to track social connections? Christakis is lucky.

During his visit to Framingham, he asked one of the study coordinators how she and her colleagues had managed to keep in touch with so many people for so long. The woman reached under the table and pulled out a green leaf - this was the form the staff used to collect information from each participant every time they came for an examination.

Everyone asked: who is your spouse, your children, parents, brothers and sisters, where they live, who is your doctor, where you work, live and who is your close friend. Christakis and Fowler could use these thousands of green shapes to manually reconnect Framingham's social connections decades ago.

Over the next several years, scientists led a team that carefully reviewed the records. When the work was completed, they received a map of how 5124 subjects were connected: it was a network of 53,228 connections between friends, family and colleagues.

They then analyzed the data, starting by tracking patterns of how and when the residents of Framingham got fat, and created an animated diagram of the entire social network, where each resident was depicted as a point that grew more or less as the person gained or lost weight. over the past 32 years. The animation made it possible to see that obesity was spreading in groups. People got fat for a reason.

The social effect was very powerful. When one resident of Framingham became obese, his friends' propensity for obesity rose to 57%. Even more surprising for Christakis and Fowler, the effect did not stop there: a resident of Framingham was about 20% more likely to be obese if a friend of his friend had a similar problem, and the close friend himself remained at the same weight.

“You may not know him personally, but a co-worker of your friend’s husband can make you fat. And your sister's friend's boyfriend can make you skinny,”Christakis and Fowler will write in their upcoming book, Webbed.

Obesity was just the beginning. Over the next year, the sociologist and political scientist continued to analyze Framingham's data, finding more and more examples of contagious behavior. In exactly the same way, drunkenness spread in society, as well as happiness and even loneliness. And in each case, the individual influence extended three degrees before disappearing altogether. Scientists have called this the rule of "three degrees of influence": we are connected not only with those around us, but with all other people in this web, which stretches much further than we think.

But how exactly could obesity or happiness spread along so many links? Some contagious behaviors, such as smoking, seem understandable. If a lot of people smoke around you, you will be subject to peer pressure, and if no one smokes, you are more likely to quit. But the simple peer pressure explanation doesn't work with happiness or obesity: We don't often urge people around us to eat more or be happier.

To explain the phenomenon, Christakis and Fowler hypothesized that this behavior is propagated in part through subconscious social signals that we receive from others, which serve as a kind of clues to what is now considered normal behavior in society. Experiments have shown that if a person sits next to someone who eats more, they will also eat more, unwittingly adjusting their perception of what is normal food.

Christakis and Fowler suspect that as friends around us get heavier, we gradually change the way we think about what "obesity" looks like, and silently allow ourselves to gain weight. In the case of happiness, these two argue that the infection can be even more deeply subconscious: according to them, the spread of good or bad feelings can be partially caused by "mirror neurons" in our brain, which automatically mimic what we see on the faces of people around US.

The subconscious nature of emotional reflection may explain one of the most curious findings of the study: if you want to be happy, the most important thing is to have many friends. Historically, we have tended to think that having a small group of close, longtime friends is critical to happiness. But Christakis and Fowler found that the happiest people in Framingham were the ones with the most connections, even if the relationship wasn't deep.

The reason these people were the happiest is probably because happiness doesn't just come from deep, heart-to-heart conversations. It is also shaped by the fact that you are faced with many small moments of contagious happiness in other people every day.

Of course, the danger of being in close contact with a lot of people is that you run the risk of meeting a large number of people in their bad mood. However, playing to increase sociability always pays off for one surprising reason: happiness is more infectious than unhappiness. According to the statistical analysis of scientists, each additional happy friend increases your mood by 9%, while each additional unhappy friend pulls you down by only 7%.

Findings from the Framingham study also suggest that different contagious behaviors are spread in different ways. For example, colleagues, unlike close friends, do not convey happiness to each other, but they convey an attitude towards smoking.

Obesity had its own peculiarity: spouses do not influence each other as much as friends. If a male subject from Framingham had a male friend who got fat, the risk doubled, but if the subject's wife got fat, the risk increased by only 37%. This is probably due to the fact that when it comes to body image, we compare ourselves primarily with people of the same gender (and in the Framingham study, all spouses were of the opposite sex). In the same way, heterosexual friends did not transmit obesity to each other at all: if a man became fat, his girlfriends did not suffer from it at all, and vice versa. Likewise, relatives of the same sex (two brothers or two sisters) influence each other's weight more than relatives of the opposite sex (brother and sister).

When it came to drinking, Christakis and Fowler found a gender effect of a different kind: Framingham women were significantly more powerful than men. A woman who started drinking increased her risk of alcohol consumption by those around her, while men who drank had less of an impact on others. Fowler believes that women have more influence precisely because they usually drink less. Therefore, when a woman begins to abuse alcohol, this is a strong signal for others.

The researchers' work has sparked a number of reactions from other scientists. Many health experts were delighted. After years of observing patients, they certainly suspected that the pattern of behavior was spreading in society, but now they have data to support this.

But many of those who study the networks have been more cautious in their reactions. Unlike medical experts, these scientists specialize in studying the networks themselves - from grid-connected areas to teenage Facebook friends - and they are familiar with the difficulty of establishing cause and effect in such complex structures. As they note, the Framingham study found intriguing correlations in human behavior, but this does not prove that social contamination is causing a phenomenon to spread.

There are at least two other possible explanations. One of them is "hetero / homophilia", a kind of tendency of people to gravitate towards their own kind. People who are gaining weight may well prefer to spend time with other people who are gaining weight, just as happy people may seek out others who are happy.

A second possible explanation is that a shared environment - not a social infection - may cause Framingham residents to share behavior within groups. If a McDonald’s opens in one of the Framingham neighborhoods, it could cause a group of people living nearby to gain weight or become a little happier (or sadder, depending on how they think about McDonald’s).

One of the most prominent critics of Christakis and Fowler is Jason Fletcher, assistant professor of public health at Yale University: he and economist Ethan Cohen-Cole even published two articles in which it was argued that Christakis and Fowler did not exclude all kinds of hetero- and homophilic effects from their calculations. … Initially, Fletcher wanted to replicate the analysis of the data by Christakis and Fowler, but he did not have access to the source.

Faced with this obstacle, Fletcher and a colleague decided instead to test Christakis and Fowler's mathematical methods on another dataset - the Add Health study, a federal government project that tracked the health of 90,118 students in 144 high schools between 1994 and 2002. …

Among the questionnaires circulated by the researchers was one in which students were asked to list up to 10 of their friends - this allowed Fletcher to build maps of how friends were connected in each school, and get a set of small social networks on which to check the math of Christakis and Fowler.

When Fletcher analyzed the forms using statistical tools he said, similar to those used by Christakis and Fowler, he found that social contagion did exist, however, behaviors and conditions that were contagious turned out to be completely implausible: they included acne, growth and headache. How can you get taller by associating with taller people?

This, Fletcher concluded, has questioned whether Christakis and Fowler's statistical methods actually eliminate hetero / homophilia or environmental influences and, he says, means that the Framingham study is equally dubious.

Fletcher said he believes the social contagion effect is real, but the evidence from Christakis and Fowler is simply not impressive

Other scholars have pointed out another important limitation in Christakis and Fowler's work, which is that their map showing the connections between the people of Framingham is necessarily incomplete. When the participants in the Framingham study were checked every four years, they were asked to list all members of their family, but to name only one person whom they considered a close friend. Perhaps this could mean that the named three-stage influence effects could be an illusion.

When I voiced my concerns to Christakis and Fowler, they agreed that their friendship map was imperfect, but said they believed there were far fewer holes in their map of connections in Framingham than critics claim. When Christakis and Fowler took stock of the green sheets, they were often able to establish a relationship between two people who did not identify each other as acquaintances, which reduced the number of false three-level links.

They also admitted that it is impossible to completely eliminate the problems of hetero / homophilia and environmental exposure, but this does not mean that they agree with Fletcher.

Both Christakis and Fowler point to two other findings to support their position in favor of social contagion rather than environmental impact. First, in the Framingham study, obesity could spread from person to person, even over long distances. When people moved to another state, their weight gain still affected friends in Massachusetts. In such cases, according to Christakis and Fowler, the local environment could not force both to gain weight.

Their other finding, more intriguing and perhaps more significant, was that they found that behavior seemed to spread differently depending on the type of friendship that existed between the two people. In Framingham's study, people were asked to name a close friend, but friendships were not always symmetrical.

Although Stephen might call Peter his friend, Peter might not think the same about Stephen. Christakis and Fowler found that this "focus" is important: according to them, if Stephen gets fat, it will not affect Peter in any way, because he does not consider Stephen his close friend.

On the other hand, if Peter gains weight, Steven's risk of obesity rises by almost 100%. And if two men consider each other mutual friends, the effect will be huge: one of them will gain weight, which will almost triple the risk of the other. At Framingham, Christakis and Fowler found this directional effect even in people who lived and worked very close to each other. And this, they argue, means that people cannot get fat just because of the environment, since the environment should have equally influenced everyone, but this did not happen.

The targeting effect seems to be very significant, and this fact, in turn, supports the case for the existence of social infection.

In fact, the work of Christakis and Fowler offers a new perspective on public health. If they're right, public health initiatives that focus only on victim assistance are doomed to fail. To truly combat pervasive social bad behavior, you must simultaneously focus on people who are so distant that they don't even realize they are influencing each other.

It is tempting to think, faced with the work of Christakis and Fowler, that the best way to improve your life is simply to cut ties with bad behavior. And it is obvious that this is possible, because people change friends often, sometimes abruptly. But changing our social network can be more difficult than changing our behavior: there is strong evidence in research that we do not have as much control as we might think over how we relate to other people. For example, our location on a social network or how many of our friends know each other are relatively stable patterns of our lives.

Christakis and Fowler first noticed this effect when they examined their data on happiness. They found that people deeply entangled in circles of friendship tended to be much happier than "isolated" people with few connections. But if the “isolated” girl did manage to find happiness, she did not have sudden new connections and did not migrate to a position in which she would be more closely connected with others.

The converse is also true: if a well-connected person became unhappy, he did not lose his connections and did not become "isolated." In other words, your online place affects your happiness, but your happiness does not affect your online place.

Social media science ultimately offers a new perspective on the age-old question: To what extent are we independent individuals?

Looking at society as a social network and not as a collection of people can lead to some thorny conclusions. In a column published in The British Medical Journal, Christakis wrote that a strictly utilitarian view suggests that we should provide better medical care to well-connected people because they are more likely to pass on those benefits to others. "This conclusion," wrote Christakis, "worries me."

However, there is something inspiring about the idea that we are so closely connected, two scientists argue. “Even if we are influenced by others, we can influence others,” Christakis told me when we first met. “And so it becomes more important to take actions that benefit others. Thus, the network can act in both directions, undermining our ability to have free will, but increasing, if you will, the importance of having free will."

As Fowler pointed out, if you want to improve the world with your good behavior, math is on your side. Most of us, within the three steps, are associated with over 1000 people - all those who we can theoretically help to become healthier, more cheerful and happier simply by our own amazing example.