John Cacioppo, Ph.D., is the Tiffany & Margaret Blake Distinguished Service Professor at the University of Chicago, where he directs the Center for Cognitive and Social Neuroscience.


Sandra Aamodt: How do you define loneliness?

John Cacioppo: Perceived social isolation. You can feel isolated in a crowd. You can also choose to be alone and feel blissful solitude. When people feel others around them are threats rather than sources of cooperation and compassion, they feel socially isolated, lonely.

Sandra Aamodt: The US has a very independent culture, but most people live in interdependent cultures that emphasize relatedness in defining the self. How does that difference affect loneliness?

John Cacioppo: Our culture emphasizes going from childhood dependence to adult independence. What it means to be an adult in social species, however, is not to be independent of others but to be a member on whom others in the group can depend. I think some of our society’s problems relate to that misconception of what it means to be an adult in a social species.

There’s some evidence to suggest that loneliness is higher in interdependent than independent cultures. That might sound surprising, but people in independent cultures feel most lonely on the holidays, when the cultural norm is to enjoy the company and comfort of friends and family. In interdependent cultures, that holiday expectation is true all the time, and so disconnected people in those cultures may experience more pain.

Sandra Aamodt: It’s clear why evolution would select for hunger or pain. Why does connection matter so much?

John Cacioppo: Loneliness has a lot in common with pain, hunger, and thirst. You would not want to be in these states, at least not for very long, but each has evolved as an aversive biological signal that motivates us to do something that’s good for us as individuals and as a species. Physical pain motivates us to take care of our physical body. Loneliness motivates us to take care of our social body, and in doing so, it fosters caring about others and being willing to work to stay together. We’re a fundamentally social species, and a social animal that is isolated is almost certain to live a shorter, more miserable life. Many physiological effects of isolation in invertebrates and simpler mammals are found in humans as well.

Sandra Aamodt: Is loneliness bad for health because lonely people don’t take care of themselves?

JOHN: In a meta-analysis of over 100,000 subjects published in 2012 by Julianne Holt-Lunstead and colleagues, the effects of social connection on health were comparable to smoking and about three times larger than obesity. The social science explanation is that friends and family encourage good health behaviors, but we haven’t found that to be a satisfactory explanation. For instance, in a nationally representative sample, we showed that loneliness in 2002 predicts who will die by 2008. We found health behaviors predict mortality as well, but these health behaviors did not explain the effect of loneliness on mortality. Interestingly, experimentally isolated nonhuman social animals also die earlier. This is not because the cockroach, mouse, or fruit fly’s friends aren’t encouraging good health behaviors. We’ve looked at factors like social support, hostility, personality, objective and perceived stress, and depression. The effects of loneliness are not attributable to these other factors but instead appear to operate directly on neurobiological mechanisms ranging from gene expression and inflammatory biology to neuroendocrine functioning.

Sandra Aamodt: What’s your explanation?

John Cacioppo: A higher vigilance for social threats. Lonely people are often completely unaware that their brain has gone on alert. An isolated rat put in an open field will walk around the walls and avoid the middle, which is called predator evasion. We find lonely people are hypersensitive to social threats. If I show the word “vomit” or “reject,” it takes longer to say the color in which “reject” appears. The lonelier you are, the longer it takes.

We got supporting results in a functional imaging study. When we show a negative social picture and contrast a nonsocial picture, the lonelier people are, the more likely they are to show visual cortical activation. That’s like being hungry and having the golden arches grab more attention. The more lonely you are when you see a negative social picture, the less likely you are to activate the temporoparietal junction. Lonely people, focused on self-preservation, take other people’s dire circumstances less seriously. Consistent with that, we’ve done an experience sampling study, where we beep people nine times a day for seven days. We find lonely and connected people in the same circumstances, but lonely people interpret the stressors as more stressful.

With positive pictures, we see a different pattern. Reward-related activity in the ventral striatum depends on the individual’s loneliness. In the imaging study, connected people who see a happy couple, for instance, show greater ventral striatal activity than to an equally positive nonsocial picture. Lonely people rate the scene as equally positive but show less ventral striatal activity than in response to the nonsocial picture. Similarly, in the sampling study, positive social interactions were as frequent but rated as less positive by lonely individuals.

Sandra Aamodt: Lonely people often seem to have difficulty making new connections. Why is that?

John Cacioppo: It’s because of their hypervigilance for social threats, which they don’t correct for because they’re unaware of it. Lonely individuals have higher vascular resistance at baseline, across the course of their day, which is what happens when you threaten somebody. If I’m unaware that my brain is hypervigilant for social threats, and you act ambiguously, I’m likely to respond in self-preservation mode. Of course that makes me a less likeable person. We found in the Framingham data that loneliness spreads over a ten-year period because lonely people are more likely to have negative interactions with others. If I’m negative toward my friends, we’re less likely to stay friends. So loneliness is actually contagious.

Sandra Aamodt: How far in the network does it spread?

John Cacioppo: Three people. If your friend is lonely, you have a 25 percent chance of being lonelier four years from now. When somebody interacts with you negatively, you’re more likely to interact negatively with somebody else. Negative interactions, though not the most frequent, are the most powerful interactions we have. Unfortunately they last longer and have a bigger impact than comparable positive interactions.

Sandra Aamodt: There’s a lot of talk that social media are actually making people lonelier. What do you think?

John Cacioppo: In the early 90s, there was a study, in which people were randomly assigned to get Internet access or not. To everyone’s surprise, the group with Internet became lonelier and more depressed. People say, “That was early, and now social media is associated with more connection.” Well, it depends what you mean by connection. Research shows that the less lonely you are, the more friends you have on social networking sites, but you’re making those friends offline and bringing them with you to Facebook.

Our own work suggests that what’s important is having friends on whom you can count. Popular people and billionaires have more than enough friends, but they can be very lonely because they can’t trust anyone. We did a study where we looked at the proportion of interactions with friends that were face-to-face, on social networking sites, in chat rooms, on gaming sites, or on dating sites. The greater the face-to-face percentage, the less lonely people were. Now most people use Facebook to leverage face-to-face interactions, but some use it as a substitute. Metaphorically, that’s like eating celery. If you’re hungry, it’s better than nothing, but it doesn’t provide enough nutrition.

Sandra Aamodt: Does video chat help?

John Cacioppo: I teach mindreading, being able to read facial expressions. You’d be surprised how much the structural features and coloration of someone’s face influences your ideas about traits and personality. A lot of mindreading is teaching people what not to believe about the face, what cues are misleading, about the need to verify your ideas, and how to test competing hypotheses.

I’m going into that because one of the most powerful ways to get information about someone’s state of mind is to see where they’re looking. When you’re on Skype, that’s a pretty rich connection, but you can’t tell where they’re looking or even if they’re looking at you. They’re looking at the screen, and the camera is somewhere else. That’s misinformation, not just a lack of information. What we’ve found, more by experience than experiment, is if we’re putting together a scientific collaboration, it’s important to meet face to face. After we get to know each other, we can go to Skype because we know how to interpret each other without the rich cues, but that has to be refreshed every three to six months.

Sandra Aamodt: Do introverts and extroverts differ in their tendency to feel loneliness?

John Cacioppo: The big difference is how many friends they need to feel connected. It’s not a big number in either case—one for an introvert, or three for an extrovert.

Sandra Aamodt: How much does loneliness vary between individuals?

John Cacioppo: It’s about 50 percent heritable in adults, in our study of 8500 twins or siblings. At ages seven and ten, heritability is about 30 percent. As they become twelve-year-olds, it dips down to the teens because hormones are raging in everybody then, which changes what people want from their relationships. The individual variability at that age is more due to social context than genetic predisposition, so heritability drops.

Sandra Aamodt: When you eat, you stop feeling hungry. But when a lonely person experiences social connection, it often doesn’t seem to help immediately.

John Cacioppo: It’s slower acting in adults, in part because of our culture. If you think about children, though, loneliness facilitates social learning, which is one of the reasons I think it evolved. We didn’t have time-outs historically and across cultures, but we had shunning and ostracism. Let me use time-out as a model. When a child is acting selfishly, all you do is to set the child aside. Then they act sad and moody, and that has a strong evolutionary function. If you turn around and push your way back into the group, that may be interpreted as aggressive, and you can be harmed. Acting sad and lethargic instead is a quiet call for someone to help and connect if they’re willing, and it does have that pull. I’ve put my child into time-out, and it’s hard to wait a few minutes when they look and feel so sad. Then after one minute per year of life, you hug them and bring them back. Their mood changes right away, and they become better social citizens. That’s a powerful input and fairly quick.

Sandra Aamodt: What’s the best way to end a period of loneliness?

John Cacioppo: There are four types of treatments: social support, social skills training, social engagement—bringing lonely people together—and social cognition. That last approach comes from our ideas about hypervigilance for social threats. It involves paying attention to how we think about other people and how to change that. The first three all are better than chance, but none is very effective. Social engagement seems to work quickly, but it breaks down because two people with hypervigilance for social threats are likely to have toxic interactions. Providing social support works much less well than people thought because it makes lonely people feel pathetic—and they still have no control. As for teaching social skills, what we find in population-based research is that lonely individuals are you and me when we’re in an isolating condition. Most of them have normal social skills. When something goes wrong, their brain puts them in a state of self-preservation, so they don’t use those skills appropriately.

Social cognition works the best. The effect size is about three times larger than the other types of treatments. There aren’t many studies yet, and early reports often have the biggest effect sizes. But even if it’s one and a half or two times the size (and that’s conservative), there’s something better in that approach.

We’re now doing research with the Army, trying to make soldiers feel less isolated so that they’ll recover from adversities better. We’re training them to be interdependent in every aspect of life. Afterward, one master sergeant said, “I’ve been employing these principles at home, and my wife said ‘Why didn’t they give you this training earlier? This is great!’”

Sandra Aamodt: What are you teaching them?

John Cacioppo: We train how to connect, the importance of not leaving anyone behind, how to read when a soldier is starting to feel isolated and how to respond, and how to deal with conflict. We cover changing from blame to solving problems. We teach them how to read the eyes, but we also give examples of where the same expression can mean more than one thing. We teach about behavioral confirmation and its toxicity. The principle of pausing and backing off when they start to feel themselves running out of control. If they feel isolated, as long as they can talk about it, then they’re still in control. If they don’t, loneliness will have effects that they can’t control or recognize. We emphasize being the one on whom others can depend rather than being independent and ask them to imagine the difference in how you’d treat your family. Basically we’re trying to invalidate their implicit theories and get them to be better scientists. Because much of what happens in loneliness is not conscious. Lonely people don’t know it, but they lose the ability to control their impulses, which also happens in isolated nonhuman animals. It really is a brain state.

I thought if it’s true that loneliness is a brain state, then loneliness should penetrate the night. If it’s dangerous to fend off beasts by yourself with a stick, imagine how dangerous it is to put the stick down and go to sleep when you don’t have a safe social surround. Lonely individuals do have more microawakenings throughout the night and show greater sleep fragmentation. We also studied the Hutterites, a religious cult with overall the lowest levels of loneliness we’ve seen anywhere. Yet the lonelier they are—and nobody was really lonely—the greater the fragmentation of sleep. That’s above demographics, obesity, standard medical risk factors, depression, and stress.

Sandra Aamodt: What else have you learned from your longitudinal study of loneliness?

John Cacioppo: Loneliness leads to depression. If we include perceived threat, the effect of depression on loneliness disappears, but the effect of loneliness on depression doesn’t even decrease. It leads to higher blood pressure across time in older adults and higher vascular resistance in younger adults. Lonely people show higher rises in morning cortisol and a flatter cycle across the day, cross-sectionally and longitudinally. Again, that’s taking into account these other factors. I’ve never liked to just take one variable and claim that everything’s due to that, so we try to measure all plausible alternative interpretations.

There’s also an interesting change in gene transcription inside white blood cells as a function of whether you’re lonely or not. It seems that the immune system is changing based on the brain’s perception of the pathogens in your environment. If you feel lonely, your immune system changes to protect against bacteria. In social circumstances, your immunity changes to protect against viral pathogens. Since viruses are transmitted by other people, that makes perfect sense from an evolutionary perspective.

Sandra Aamodt: Is the chronic cortisol elevation in lonely people severe enough to damage the hippocampus and memory?

John Cacioppo: David Bennett published a longitudinal study over eight years showing that loneliness, above objective circumstance and depression, predicted cognitive decline and onset of dementia in at-risk individuals. The mechanism could be either hippocampal damage from high levels of cortisol or inflammatory biology. A study in 2010 showed that if you induce a stroke in a mouse that’s housed alone or in a pair for two weeks beforehand, the isolated mouse shows three times the neural damage and is more likely to die. If you block interleukin 6, the two groups look the same, suggesting a role of inflammatory processes in making neural damage worse.

Sandra Aamodt: How has what you’ve learned about science changed the way you live your life?

John Cacioppo: In just about every possible way. One of the most important is that before I got involved in science, I took failures as a bad thing. As a scientist, failure is exciting because we can expand, enrich, and correct our theories. If we’re never wrong, we never learn anything, so it’s exciting to fail for the right reasons.

Also, my research has led me to appreciate the importance of listening, understanding, and working collaboratively and synergistically with other people. If I change my thinking to recognize that my best friend in academics is my critic, that leads to more gratitude in how I view everybody.

We teach the soldiers that one of the best ways to build friendships is to enjoy positive times together. We soon realized that their idea of a positive event is beating a colleague and lording that success over them. So we explained that when they beat their battle buddy in the nine-mile run, that success is due to both people’s efforts because their buddy drove them to run that quickly.

Fair competition is the key. It may often sound like I believe cooperation is good and competition is bad. There’s another dimension, though, whether you excel for personal benefit at the expense of others or you and others are benefiting. On Wall Street in 2008, they were cooperating with each other, but they were exploiting everybody else. Then there’s competition like the Olympics that pulls people together.