Like most scientists, Steve Cole didn’t believe things happened without reason. But no matter which variable he tested—and he tried all the usual suspects, from sleep to sex—nothing seemed to explain why the gay men were dying.

It was the late 1980s, the age that gave us the personal computer, the disposable contact lens, and The Simpsons — but also the Challenger explosion, the global stock market crash, and “new” Coke. AIDS was raging in the United States.

Cole was a young psychology researcher who’d landed on a study of 988 HIV-positive, AIDS-free gay men¹, and he was trying to crack the workings of the lethal virus. Over nine years, Cole tracked 80 of these men. Every six months, they gave blood, sat for interviews, and filled in questionnaires. Every time, there were fewer and fewer of them. A significant number fell sick, many perished. And as they did, Cole wondered why some subjects succumbed to HIV while others were able to resist it. He looked at the obvious—age, socioeconomic status, overall health, sleep quality, exercise habits, sex life, anxiety levels, depression history—but none of these predicted who would get AIDS or how soon they’d die.

At one point, Cole thought to compare the openly gay men to those who hid their sexual identity. It turned out that closeted men got AIDS faster and died sooner than out men did. What about being in the closet made those men more vulnerable to HIV? A strand of research at the time implicated repressed emotions as a possible factor—closeted men bottled up their feelings, which made them sick—but the evidence was thin and the biological mechanism unknown. After all, how do you convert emotion into disease? How do feelings turn into something real?

In a follow-up study, Cole decided to test yet another variable: subjects’ sensitivity to rejection. He found that, on average, closeted men were more sensitive to rejection than openly gay men. It made sense: If you cared too much about others’ judgment, you’d want to hide the parts of yourself likely to be stigmatized, such as your sexual identity.

More importantly, Cole found that sensitivity to social rejection predicted an AIDS diagnosis and early mortality even better than whether one was openly gay. In other words, the mysterious deaths of the HIV-infected men appeared to stem not from their failure to express themselves but from a fear of being rejected if they did. Indeed, sensitive men who hid their sexual identity did not suffer more than those in the open: the closet, Cole theorized, probably protected them from social rejection and its dire health effects. Suppressing their emotions perhaps saved their relationships, and their relationships, in turn, saved their lives.

At the time, the impact of social connections on health wasn’t entirely new. For a long time, researchers had observed a curious link between social isolation and a number of diseases, from heart disease to cancer to some neurodegenerative disorders. But no studies had been conducted to test this link directly, and no one could really explain it. According to one theory, social relationships exert a positive influence on health because of peer pressure. If, the theory went, there are healthy people around you whose opinions you care about, you’ll be more likely to swing by the gym, slug kale juice every morning, and skip the second or third glass of wine at the end of a tough workday.

But Cole’s research showed that bad habits weren’t actually wrecking his subjects’ health. The gay men who contracted AIDS were just as physically fit and mentally well as those who resisted the disease. There seemed to be something else, something more direct, that was stripping some participants’ immune defenses and leaving them fatally exposed to the virus.

For a long time, Cole couldn’t figure it out. The technology that now allows scientists to zoom into cells and read genetic code was new and prohibitively expensive. Then, in the early 2000s, Cole got his hands on a gene-sequencing machine. He also met John Cacioppo, a neuroscientist from the University of Chicago, who had pioneered the idea that social relationships shape our brains. Cacioppo believed who we know — and how well we know them — impacts our thoughts, behaviors, and physical health on the very level of biology.

Cacioppo never bought the peer pressure theory of social relationships. “What I knew was that no matter what social species you’re talking about, all the way down to fruit flies, if you isolate them, they die earlier,” he says. And since fruit flies are hardly the type to sink into debauchery unless other fruit flies keep an eye on them, there had to be another route for social isolation to get so deeply under the skin of humans and animals.

To find out what this route might be, Cole and Cacioppo analyzed data from 14 elderly Chicagoans of various ethnic and socioeconomic backgrounds. The data, collected over four years, included the subjects’ demographics, habits, behaviors, medical conditions, and self-reported feelings of loneliness. In addition, Cole and Cacioppo scraped the inside of each participant’s cheek with a swab, then gene sequenced and analyzed the DNA samples.

Isolation, it appeared, could fool the body into thinking it’s in mortal danger.

The results were remarkable: The immune systems of the lonely subjects behaved differently from those of the non-lonely—and dramatically so. For instance, the lonely subjects’ antiviral defenses were partially shut down, which suggested a possible reason for why socially sensitive gay men were so vulnerable to HIV. But that wasn’t where the damage ended. In addition to suppressing the body’s innate defenses against viruses, loneliness also seemed to crank up inflammation to dangerous levels, raising the risk of myriad diseases.

Inflammation, in normal doses, is not a foe. In fact, it’s part of the body’s defense against malicious microbes. It’s also a marvelous example of nature’s flair in designing human beings. Circulating through your bloodstream are thousands of immune cells. These cells come equipped with receptors that not only detect invading pathogens but also tell them apart and mount a defense tailored to each type of bug.

The immune system’s default mode is antiviral. That’s because, as a social species, we’re constantly exposed to viruses that spread from one person to another. Bacteria, on the other hand, can infiltrate through open wounds. During injury, immune cells squeeze out of blood vessels and travel to fight bacteria wherever they are, triggering a cascade of healing processes. One of these processes is inflammation. It’s the redness, swelling, and heat you experience when you injure yourself, and it’s what keeps infection from spreading throughout your body.

Once the intruder is no longer a threat, inflammation subsides. If it doesn’t, it can become extremely dangerous. Prolonged inflammation has long been known as a risk factor for infectious diseases, but around the time that Cole and Cacioppo began their collaboration, other researchers started to link inflammation to a host of non-infectious illnesses, from asthma to arthritis, diabetes, cancer, Alzheimer’s, and depression. A 2012 review of the clinical literature suggested that chronic inflammation may, in fact, lie at the root of many more diseases than initially thought. “Perhaps,” the researchers concluded, “even all of them.”

In their own study, Cole and Cacioppo showed that loneliness might trigger inflammation the same way a physical injury would—but without activating the built-in break that normally keeps inflammation in check. The lonely subjects’ immune systems had switched from their default antiviral mode to a chronic injury-fighting mode, leaving them exposed to both viral infections and inflammation-related diseases. This switch provided a clue to how our social world interacts with our biology, and how loneliness can wreak havoc on our brains and bodies.

It also explained why both the closeted and the rejection-sensitive gay men fell prey to HIV more easily: The former were more likely to be socially isolated while the latter felt social isolation more strongly. And isolation, it appeared, could fool the body into thinking it’s in mortal danger. It tinkered with one of our most fundamental survival mechanisms—the immune system—and could even turn it against us.

Though astonishing, the link between loneliness and immune activity hinged on just one study of only 14 people. Correlations like this come up constantly in scientific research, and many of these findings are never replicated. Genetics at the time had a particularly embarrassing replication rate: alleged “breakthroughs” would arrive hot off the press, only to be demolished and discredited the next day.

Naturally, Cole was cautious about drawing conclusions. But as more studies followed, a growing body of evidence emerged to support his initial findings. In 2011, he and Cacioppo extended their original sample from 14 to 93 people and got the same results: reduced viral defenses and elevated inflammation in the lonely subjects. This pattern cropped up in various other studies: people suffering from PTSD and breast cancer, people grieving or caring for dying spouses, people dealing with relationship difficulties, even people being evaluated while performing a task.

“Social isolation is the best-established, most robust social or psychological risk factor for disease out there. Nothing can compare.”

So far, research suggests that the situations that most reliably predict a dangerous surge of inflammation involve social rejection and loss. The health impact of divorce, for example, depends largely on who initiates it: If it’s your spouse, you not only lose an important relationship but feel rejected, which will derail your immune system more than if the decision to end the marriage came from you.

A 2010 meta-analysis of 148 studies concluded that being lonely is not only bad for your health, it’s in fact significantly worse than drinking or not exercising. Loneliness emerged as a risk factor for early death with an impact as significant as smoking and three times more significant than obesity. Yet even this may be an understatement because, according to Cacioppo, obesity “does not make you as miserable as loneliness.” Not even stress can measure up. “We think of stress as a risk factor for disease,” says Cole. “And it is, somewhat. But if you actually measure stress using our best available instruments, it can’t hold a candle to social isolation. Social isolation is the best-established, most robust social or psychological risk factor for disease out there. Nothing can compare.”

John Cacioppo didn’t set out to study loneliness because of a harrowing personal experience. “People are disappointed when they hear that,” he says in an interview with University of Chicago Magazine. He was a social psychologist at a time when social psychology tried to explain human behavior through the social and cultural experiences we are aware of and can verbalize. But most of human behavior is unconscious. The solution, for Cacioppo, was not a return to digging into repressed desires in dreams and fantasies—by that time, psychoanalysis was seen as little more than quackery—but a better understanding of the brain, the grand originator of all thought, emotion, and motivation. There was no way, he thought, “to comprehend the full scope and processes of mental existence without delving into biological reality.”

But neuroscientists at that time paid no attention to the social world that Cacioppo wanted to marry with “biological reality.” They saw human behavior as little more than electrical currents zipping inside our heads. This made no sense to Cacioppo. He knew that the brain evolved in an intensely social context. Throughout history, relationships with other humans were far too important for survival to have left no mark on our biology and evolutionary development.

In 1992, Cacioppo and his colleague Gary Berntson founded social neuroscience, which aims to study the neural, hormonal, genetic, and molecular mechanisms underlying social behavior. In one of the field’s founding documents—a hefty 1,357-page tome published by MIT Press in 2002—Cacioppo includes a quote by David Spiegel at Stanford University that marvelously sums up the main tenets of social neuroscience:

We are fundamentally social organisms, our mythic rugged individualism notwithstanding. We are born to the most prolonged period of abject dependency of any mammal. For the species to survive, human infants must instantly engage their parents in protective behaviour, and the parents must care enough about their offspring to nurture and protect them. Even once grown we are not particularly splendid physical specimens. Other animals can run faster, see and smell better, and fight more effectively than we can. Our major evolutionary advantage is our brain and ability to communicate, remember, plan, and work together. Our survival depends on our collective abilities, not our individual might. Thus, it makes sense that our health may also depend on our interactions with one another.

The breakdown of these interactions seemed like a good starting point for studying their health impact; hence Cacioppo’s focus on loneliness. To be sure, he never expected that his research would take over two decades, but the more he learned about social isolation, the more there was still to learn. “This just continues to change how I think about us as a species.”

From an evolutionary point of view, loneliness appears to serve a purpose, which Cacioppo likens to that of hunger. Hunger prods you to look for food before you run out of fuel and starve to death. It signals a survival risk. And to make sure you heed that signal, hunger feels unpleasant and bothers you until you feed yourself. Loneliness acts in a similar way. Historically, we depended on others for protection and nurturing, and social isolation could cost us our lives. Loneliness left us vulnerable to lurking predators. Feeling lonely, just like feeling hungry, is a danger signal. When you stray too far from the tribe, the pang of loneliness drives you back to repair alliances, invest in new relationships, and thus avoid dying.

This fact — that during our evolution loneliness cued physical threat — can help explain the unusual immune activity of people facing loss and social rejection. For a long time, researchers saw inflammation as an automatic physiological response to bacterial invasion: Immune cell receptors would sniff out a foreign agent, and off would go a built-in sequence of defense reactions and healing processes. But now we know that these same mechanisms can kick in preemptively, before bugs have infiltrated the body, and before an injury has actually occurred.

How does this work? Contrary to what we’ve long thought, it turns out that immune cells do not passively execute prewired defense mechanisms. Instead, they actively listen to outside signals—especially those coming from the brain—and act on them. When your brain sends a message that you are lonely, the immune cells hear danger! and turn off some of your antiviral defenses: You don’t need those now, after all, since viruses spread among people. Being lonely, you run a higher risk of injury and infection by bacteria, which is why your immune system diverts resources from antiviral to antibacterial defenses. In other words, it starts pumping out inflammation.

Your immune system is stuck in prehistoric cave-dweller times.

In the modern world, however, loneliness seldom equals mortal danger. You get lonely and feel rejected for all kinds of non-life-threatening reasons: you break up with your partner, get passed up for promotion, give a talk and it tanks, call a friend to have a good cry and she’s distracted. You begin to think your friendship doesn’t mean a whole lot to her. Perhaps she used to envy you and now she gloats internally at your spectacular fall. Before you know it, you’ve spun a drama of epic proportions out of the tiniest smidgeon of evidence.

Meanwhile, your immune system is stuck in prehistoric cave-dweller times. It equates social rejection, no matter how trivial, with getting mortally wounded or becoming a lion’s dinner. It links loneliness with physical danger. That’s why, says Cole, “purely symbolic or imagined stimuli—that is, situations that have not yet happened and may never actually occur—can engage the same ancestral programs that are triggered by actual social or physical threats.”

But unlike actual threats, which come and swiftly go (you kill or die, eat or get eaten), imagined—or perceived—threats can linger a long while, fed by our deepest fears, looming ever larger in our imaginations. When these threats take up residence in our minds, inflammation can turn from an acute emergency response into a chronic assault on the body. It can make us vulnerable to a battery of mental disorders and physical illnesses.

In one study, for example, the negative impact of feeling lonely on the immune system was twice as large as that of subjects’ marital status or their frequency of social contact (more objective measures of social integration and connectedness). This is not to say that objective circumstances don’t matter; of course they do. Losing a valued relationship, a job prospect, your self-esteem—these are not trivial matters, and the hurt is not simply in your head. Still, as far as loneliness goes, the research suggests that the factual reality of it impacts our immune system in a different way—and to a smaller extent—than our subjective experience.

Indeed when he speaks of loneliness, Cacioppo defines it as perceived, rather than objective, social isolation. The distinction makes sense. Intuitively, we know that feeling lonely does not simply equate to the number of people present in or absent from our lives. Who hasn’t felt alone in a crowd or been a stranger at a glitzy cocktail party? There’s a reason we make movies about angst-ridden holiday dinners and dreaded family gatherings: We often feel loneliest surrounded by those closest to us.

Through this lens, loneliness can be profoundly unsettling. It makes the old Milton quote resonate in a new way: “The mind is its own place, and in itself can make a heaven of hell, a hell of heaven.” Cole would say, slightly less poetically but no less forebodingly, that our cells have their own psychology, which makes “the world of the mind convert into the biochemistry of the body.”

Despite the fact that technology has connected us to a degree never before seen in history, we’re still as lonely as ever. Most researchers estimate that between 20 and 30 percent of people are chronically lonely. One study finds that, among Londoners, that number is closer to 50 percent. Loneliness is also universal. It can equally affect both introverts and extroverts, says Cacioppo, and the only difference between them is the number of close relationships required to feel connected (one for introverts versus three for extroverts).

What makes loneliness so pervasive? And why has technology done little to reduce it? Evolution, again, provides a clue. “It is not the case that historically other human beings were always necessarily a good thing to have around,” explains Cole in a 2016 interview:

There were occasions in which other human beings were raiding your camp or stealing your food and your spouse or infecting you with diseases. So, human beings are great assets to other human beings but they can also be great threats.

According to Cacioppo, this has hardwired into us a negative bias against other people that acts in tandem with our attraction to them. Our social existence emerges from this constant tug-of-war between a desperate need to connect with others and the very real threat they pose to our survival.

Sensitivity to social threats may, in fact, be even more deeply etched into biology than a desire for connection. Consider the evolution of our sense of taste. We are far more sensitive to bitter than we are to sweet, says Cacioppo, because throughout history bitter tastes often meant poison. A strong visceral reaction against bitter ensured that we stay away from foods that could kill us.

Similarly, a built-in negative bias against other humans would have protected us from miscalculated friendship. “If I make an error and detect a person as a foe who turns out to be a friend, that’s okay. I don’t make a friend as fast, but I survive,” says Cacioppo. “But if I mistakenly detect someone as a friend when they’re a foe, that can cost me my life.”

No matter how it arises, negative social bias can ultimately warp reality in an invisible yet dramatic way.

So two conflicting forces shape our social existence, amounting to what Cole refers to as the “paradox of loneliness.” Social motivation pulls us toward other people; social threat drives us away from them. This paradox can help explain why being alone is not the same as being lonely.

People low in social motivation, for example, tend to feel quite happy on their own or with just a few close friends. We call them introverts, and their experience is not so much loneliness as solitude.

For people high on social motivation, the reverse is true: They need a larger number of social ties in order to feel connected, and the absence of these ties can plunge them into deep loneliness. Then, there are those who desperately crave connection yet just as desperately want to be left alone. This is, in fact, one of the particularly cruel twists of depression: The more you need others to help you out of your hole, the more you withdraw into it, burying yourself deeper and deeper still.

Why do some people feel more threatened by others, thus risking social isolation? According to Cole, there are two main factors. The first is innate: You may have been born with a particularly sensitive nervous system that picks up the subtlest cues of social threats in your environment. A negative bias against other people can also be acquired through life experiences—especially those occurring in childhood, a critical developmental window and a sort of “dress rehearsal” for later life. Growing evidence suggests that early adversity can reprogram vital biological systems, adapting them to hostile conditions in the future. The nervous system can become more vigilant, the immune system more responsive to potential threats, and inflammation can be triggered more easily.

No matter how it arises, negative social bias can ultimately warp reality in an invisible yet dramatic way. Suppose you are wired (whether by nature or experience) to see the world as a dangerous place. In your daily interactions, you’ll treat others with suspicion or at least a mild reserve, and in response, they’ll be more ambivalent and reserved toward you. Their new attitude will reinforce your initial suspicions, which will make you more mistrustful of their motives and perhaps even hostile, which will make them even more ambivalent and reserved. And on and on.

“It becomes this vicious cycle,” says Cole, “where the behavior creates another loop from the social world outside me so then the social world really is kind of, in a self-fulfilling prophecy sense, proving my theory.” In other words, threats without any basis in reality can actually become real.

Hard to break, this self-perpetuating cycle can ultimately lead to chronic loneliness—and along for the ride, of course, is chronic inflammation. But it gets worse. Every loop between social threat and immune response further sensitizes the nervous system to danger signals while also grooving in the pathways that these signals use to travel from brain to body. Every burst of inflammation releases molecules that stimulate the same neuroimmune pipeline that produced it in the first place. This sets in motion yet another vicious cycle that continually reinforces inflammation in a process that scientists call “biological embedding.”

Things once considered merely trivial or a normal part of life can dramatically crank up inflammation and make someone sick.

The health effects of biologically embedded inflammation are varied and insidious. In the short term, it can lead to hypervigilance, social anxiety, and heightened sensitivity to pain. As the brain becomes increasingly sensitized to threat signals and the immune system more responsive to them, symptoms turn more severe. Disrupted sleep, chronic pain, and depression can arise. Over time, the whole mechanism can become so sensitive that it’s triggered by even lower stress levels. At this point, things once considered merely trivial or a normal part of life can dramatically crank up inflammation and make someone sick.

Most harrowing of all, it can reach a point where no external trigger is needed at all: Inflammation becomes so deeply embedded that it just keeps raging on even if life seems nothing short of miraculous. In the long term, this can lead to inflammation-related diseases, such as rheumatoid arthritis, cardiovascular disease, diabetes, Alzheimer’s, some cancers, as well as treatment-resistant depression, premature aging, and early mortality.

To call this bleak would be an understatement. It’s chilling—morbid even. But a number of researchers think there is reason for hope. The better we understand inflammation, the more we can do about it. Cole, for one, doesn’t believe that the human genome evolved to make us miserable. Quite the opposite, in fact: Our genes, he says, want us to be happy, to thrive. Just like he once set out to demystify the early deaths of gay men, Cole is now trying to pin down the biology of health, happiness, and connection.

But that’s another story altogether.

  1. In 1983, the National Institute of Allergy and Infectious Diseases in the US launched a massive study into the causes of AIDS. Known as The Multicenter AIDS Cohort Study (MACS), it included over 7,000 gay and bisexual men in Baltimore, Chicago, Pittsburgh and Los Angeles. The Los Angeles part of the study took place at UCLA, where 988 of the 1759 participants enrolled tested as HIV-positive but otherwise healthy. From this group, Cole drew a sample of 80 gay men, who had taken additional tests to capture the social, cognitive, behavioural and emotional aspects of AIDS.

This story was originally published on Mindrise.