Skip to main content

Verified by Psychology Today

Warming to Each Other

Humans have always relied on one another for warmth, both physical and emotional. New research hints at how we do it, and points toward a future when we might enlist temperature to help boost our mental health and relationships.

Illustrations by Christiana Spano
Illustrations by Christiana Spano

Imagine a male emperor penguin, about 10 years old. Let’s name him Harry. It’s July, the middle of the Antarctic winter, and Harry is into his fourth month on an empty stomach. Male emperor penguins routinely fast when they pair with females, and continue without food for as long as 115 days. After mating, the females lay eggs and then waddle back to sea in search of food. They won’t be back for months, having left each stay-at-home penguin dad with an egg to care for.

So, the wait begins. Harry places his egg on his feet and envelops it in his brood pouch, a thick, skirtlike layer of naked skin. To ensure incubation, the egg is kept at a cozy 96.8°F (36°C), not easy in Antarctica, where winds can blow at 110 miles per hour and temperatures may dip below−49°F (−45°C). If Harry wants his egg to survive—and to go on living himself—he must huddle with others, perhaps thousands of others, more closely than rock fans at a sold-out concert, leaving each bird no more than a square foot to himself. The pressure inside the group can become so great that some penguins are pushed up on top of the others like crowd surfers.

As a means of saving energy, all of this apparent discomfort makes perfect sense. When penguins huddle, they drastically limit the amount of body surface each exposes to the elements, making it far easier to conserve heat. What’s more, each warms up the microclimate of the pack, which can rise to as high as 99.5°F (37.5°C).

For years, in places like Pointe Géologie, the rocky archipelago off the coast of Antarctica, scientists have conducted studies of huddling penguins. To better understand why and how the animals huddle, the researchers glue temperature-measuring instruments and transmitters to the feathers of selected birds. Their studies have shown, for example, that the penguins may spend as much as 38 percent of their time packed tightly with others, sometimes huddling for just a few minutes at a time and sometimes for hours on end. The wind-whipped group may appear perfectly still, but time-lapse videos show that the structure of the group constantly fluctuates as penguins move within it, swapping places with one another.

Penguins are not the only animals that rely on others for thermoregulation, or the maintenance of their core body temperature. We do it as well, although not as much as we used to. “Warmth was created by bodily contact. Comfort was a matter of bodies,” French historian Daniel Roche wrote in 1981 in Le Peuple de Paris, and he was correct: A human body radiates about 330 Btu (British thermal units) of energy per hour—roughly equivalent to that radiated by a classic tungsten-filament 100-watt lightbulb. Keeping a small bedroom of about 100 square feet warm during a relatively mild Parisian winter requires about 1,100 watts per hour, and so, in 18th-century Paris, few kept a bed to themselves: Their radiated body heat could cut space-heating requirements by over 60 percent. The physical connection between warmth and sociability, between coziness and trust, made perfect sense, as indeed it has for generations. Yet, in modern times, that physical link has been largely broken, at least for adults. The reason for this is the triumph of central heating.

By reducing the physical-survival need for huddling, central heating may pose a challenge to how human beings use social thermoregulation to create, strengthen, and maintain social and emotional bonds. The proliferation of even-more-modern technologies, such as electronic and digital communication, has allowed us to connect with one another in the total absence of physical proximity, let alone contact. We can hear and see one another in real time despite separation by great distance. And yet this ability, remarkable though it is, serves to remind us of all that we lose when physical proximity is absent. Even today, we can be dramatically reminded of how, throughout evolution, the reliance of social warmth on physiological warmth became hardwired into our bodies, just as it is in other animals such as penguins, naked mole rats, or the Barbary macaques of Morocco.

Optimizing and Economizing

Apart from getting enough oxygen, temperature regulation is the most vital and energetically expensive thing any animal must do. But unlike oxygen, temperature fluctuates in the environment all the time, so survival requires continual vigilance. Fortunately, animals are canny economists or, perhaps better, “optimizing agents.” They continually evaluate the cost versus the benefit of different behaviors to figure out which will demand the least energy and thereby help conserve precious body fat.

If you’ve ever seen a group of cyclists racing on the road, you’ve noticed that they tend to follow others in an elongated pack, called a peloton. It’s not that the cyclists are lonely. They are relying on one another, “drafting” to conserve energy by exploiting the lower air resistance behind other racers. Skilled drafters reduce their own power requirements by as much as 39 percent at a speed of 24 miles per hour, especially if they are in the middle of the peloton. This is also why fish swim in schools and bacteria travel faster as a group. Like huddling, drafting is all about the economy of action—choosing the most cost-efficient behavior to save resources. It’s based on a stunningly simple principle: Animals need to take in more energy than they spend. If they don’t, they die.

In penguins, the size of the social network provides highly economical protection from the cold. Researchers have, however, also measured the body temperature of vervet monkeys in large social networks as well as the temperature of more solitary vervets. Psychologists like to observe vervet monkeys not just because they are exceptionally cute but because they share with humans a number of less attractive characteristics, including tendencies toward hypertension, anxiety, and both social and dependent alcohol use. Vervets also understand the importance of forming large social networks. Early in life, much like that of human infants who cling to the caring warmth of their mothers, some part of the vervet brain becomes a “prediction machine” by which it comes to know that it can rely on others to keep it warm through a combination of proximity and grooming. When researchers compared the body temperature of vervets in large social networks with the temperature of more solitary ones, they found that a highly attached vervet within a large network has a lower body temperature than a lonely one.

Illustrations by Christiana Spano
Illustrations by Christiana Spano

Finding Warmth in Diversity

What drives vervet society drives human society as well—with an important difference. In vervets (and in penguins), the size of the network provides highly economical protection from the cold. In humans, social diversity provides this protection, not only giving members the emotional feeling of warmth but actually correlating with measurably higher oral temperatures, as my colleagues and I found in the studies of our Human Penguin Project (HPP).

We sought to identify known correlates of core body temperature, including social-relationship variables such as nostalgia and attachment to homes. We also included variables typically thought to affect body temperature, including stress and medication use. Finally, we identified variables relating to metabolism and network quality, such as daily diet and consumption of sugary beverages, acting on the common hypothesis that relative social isolation leads people to increased levels of sugar intake.

We were purposefully overinclusive in selecting variables because we wanted to ensure that we identified many of those that have figured most prominently in past research as predictors of core body temperature. We addressed such issues as self-control and attachment as well as alexithymia (the inability to identify and describe one's own emotions), all of which bear directly upon the regulation of stress and, therefore, might relate to body temperature.

For our studies, participants were asked to complete a survey between 9 and 11 in the morning, and to refrain from eating or drinking anything warm or cold for 10 minutes beforehand and from exercising for an hour beforehand. They also used an oral thermometer to measure their own temperature before and after taking the survey. In our main study, we collected data from a dozen countries, covering 1,523 participants. To test the social thermoregulation principles we had identified, we needed to test the basics, which required using supervised machine learning to identify both the social and nonsocial factors that bear upon core body temperature. The complexity involved in predicting people’s core body temperature lies in the range of variables that must be examined. We found that network diversity, defined in terms of the number of high-contact social roles in which one engages, is one key predictor of core body temperature.

We used another procedure to demonstrate that colder climates relate to higher levels of complex social integration and that a higher level of social integration relates, in turn, to higher core body temperature. Despite the availability of central heating, people continue to rely on social warmth—associated with complex social integration—to counter the physical cold of their environment. We don’t exactly understand how humans do it, but our working hypothesis is that emotions are tied to temperature changes; people also help regulate one another’s temperature via social emotion-regulation processes. And if you perceive that your interaction partner is emotionally close to you, you will likely be willing to spend more energy to help thermoregulate him or her.

Someone to Hold On To

We started suspecting that there are links between emotional closeness, the willingness to invest in one’s partner, and thermoregulation thanks to a project on social thermoregulation in relationships, in which we presented people with 36 statements about their feelings of attachment in close relationships, such as “My romantic partner makes me doubt myself,” “I feel comfortable sharing my private thoughts and feelings with my partner,” and “I find it easy to depend on romantic partners.” Participants responded to the statements on a scale from 1 (completely disagree) to 7 (completely agree). They then held a warm—or cold—cup, and we asked them to name five people who spontaneously came to mind. After this, we asked them how close they felt to each individual they had named. Based on the results of earlier, similar studies conducted by others, we predicted that holding a warm cup would elicit thoughts about people to whom participants felt close. We were very wrong: Consistently, people who had held cold cups thought mostly about loved ones.

We took our analysis a step further to explore whether and how attachment to others affected participants’ responses to temperature. Previous attempts to do this had yielded inconsistent effects. By presenting our 36 statements followed by contact with a warm or cold cup, a request to think of five people, and our questions about how close participants felt to each of the five, we were able to detect a consistent pattern: People who believed they could depend on their partner, and felt comfortable sharing their innermost thoughts with him or her, tended to think more of those loved ones when cold. The reverse occurred for those who did not feel comfortable depending on romantic partners.

Illustrations by Christiana Spano
Illustrations by Christiana Spano

Safe Havens

It is not just the people close to us that protect us from the cold and to whom we form attachments. History teaches us that humans have consistently sought demarcations of space—from caves, alcoves, and grottos to huts and, finally, houses—to shelter themselves from predators and the elements, especially the potentially lethal cold. The evidence of material culture supports the idea that houses can fulfill a need for affiliation or belonging.

To explore this, two of my former students, Bram van Acker and Jennifer Pantophlet, one of their peers, Kayleigh Kerselaers, and I devised studies to test whether houses, as pictured in real estate ads, became more attractive when temperatures dropped. We wanted to test an aspect of the house-as-a-home idea that was less obvious and more specifically a function of social thermoregulation.

Our pilot study was based on hypotheses in line with the social-cognition principle known as “priming.” We all have a concept of a warm house, so touching something physically warm could prime us to think that a house we are asked to rate is socially warmer—that is, homier. Based on previous findings, this prediction seemed reasonable. But we were wrong, again: Holding a cold cup tended to prompt people to judge an advertised house as homier. So, we updated our hypothesis and changed our procedure, moving outside the laboratory and seeking participants more likely to have actual experience with buying a house—in other words, not students. We also decided not to try to prime participants by having them hold cups, but instead made use of prevailing warm indoor and cold outdoor temperatures.

What we found was not quite what we hypothesized: Actual temperature, exclusively, predicted the degree to which people assessed a house as more communal, or a “safe haven.” This, in turn, predicted both how attractive they found a house and their willingness to pay more for it.

As we anthropomorphize our homes, so do we anthropomorphize many products and objects as thermoregulating tools. In some cases, we even project onto objects that seem to have nothing to do with keeping us warm. In 2014, I conducted a study with my student Janneke Janssen (whose idea was the impetus for the study) and colleague and friend Jim Coan, whose work inspires much of my own. Together, we explored how consumers maintain “warm, trusting relationships” with product brands.

“Communal brands” are those associated with what marketers call a “brand community,” a community that is based on a shared attachment to a product, logo, or brand. Marketers believe that communal brands are connected with individual identity and culture. Our group created a set of studies to test the hypothesis that just thinking about communal brands would increase participants’ perceived temperature. We tested a pretty large sample (2,552 people) from Amazon Mechanical Turk, a website popular among psychologists doing research. In five studies, we found that thinking about positively perceived communal brands leads people to estimate ambient temperature as higher. What is perhaps more curious (and very much in line with the studies on houses) is that some exploratory analyses we conducted in this set of studies suggested that the increased temperature perception drives people’s willingness to purchase the brand as well as the amount they are willing to pay for it.

It Must Be Love

Although there are many things left unexplained, the picture that emerges is clear: Temperature regulation plays a crucial role in our lives, giving me the confidence to predict that we can use it to modernize relationship therapy in a manner my colleagues and I call Social Thermoregulation Therapy (STT). Although we are by no means at the stage of this translation process to application, advances in electronic health are creating digital wearables, such as bracelets with temperature sensors, as a means of assessing the thermoregulatory dynamics in a particular relationship. STT can enhance existing therapies by integrating sensor and actuator technologies—wearable tech—into the well-established approaches of Emotionally Focused Therapy to help people use thermoregulation to improve their close relationships.

The stakes are high. A large body of work demonstrates that successful relationships are among the strongest predictors of physical and mental health. To date, research on how relationship quality influences physical and mental health has focused on what we may call the “higher-order” levels, defined for couples as experiencing fewer marital problems, having generally better health, and enjoying greater satisfaction with their relationship.

My colleagues and I turned from such higher-level results to lower-order thermoregulatory issues: health problems caused by dysregulated body temperature, the role of temperature regulation in sociality, the reliance of social warmth on physical warmth, and the lower-level dynamic of coregulation, which describes an individual’s continuous action or behavior as modified by his or her partner’s changing actions or behavior. We believe that thermoregulation is crucial to physiological coregulation in the close relationships of couples. That elicits the question: Can therapies be developed to improve physiological coregulation in couples?

Illustrations by Christiana Spano
Illustrations by Christiana Spano

Temperature-Based Therapy

One promising direction can be to integrate such a therapy into an evidence-based therapy that already takes to heart emotional coregulation, like Emotionally Focused Therapy (EFT). Sometimes one partner responds to the other’s anger in a way that is not conducive to feeling safe for either of them. EFT helps partners identify those aggravating patterns and improve well-being. We suspect that, by assessing how couples’ temperature changes, we can begin to understand the interactions underlying destructive patterns of behavior and expression. Thus, a chunk of why people behave the way they do in relationships can be studied by focusing on temperature regulation. This makes it possible to use STT to complement EFT by adjusting a relationship’s peripheral temperatures in ways that enhance the individuals’ perception of social predictability within it.

I often express reservations and doubts about the state of psychological science. But I believe there are steps we are taking to make our insights far more practically useful. The replication crisis, which emerged after 2011, has revealed that many of our findings are insufficiently precise and insufficiently generalizable. Yet we do know that there are some general patterns that we find time and again. To me, this implies that we are capable of identifying general phenomena but that we don’t yet have the capacity to apply psychological science with a high degree of accuracy outside of the laboratory, in the “real world.”

Digital wearables such as bracelets with temperature sensors will enable us to assess the thermoregulatory dynamics of coregulation in a relationship. To know whether social thermoregulation couples therapy can work, we need to study temperature-regulation patterns in couples. If we are able to distinguish via temperature the patterns present in high-quality relationships from those in low-quality relationships, we believe we can also improve relationships via temperature.

By using a smart algorithm to connect sensors with digital devices that manipulate skin-level temperature, could it be possible to make responsive temperature adjustments in couples to push low-quality relationships toward high-quality ones? Overall, we hope that doing this will beneficially change couples’ behavior, even if just a little bit. Ideally, this manipulation will enhance both partners’ feelings of predictability and safety—but we must note that changing temperature will not change an abusive relationship into a high-functioning, high-quality relationship.

This research is in an early stage, but it shows strong potential. Commercially available bracelets already allow users to adjust their skin-level temperature by electronically cooling or warming one spot on the body to improve overall comfort and mood. This is a significant step toward integrating wearables into the
ever-growing Internet of Things, technology that has revolutionized temperature regulation in homes and businesses.

Pause, take a breath, and imagine the implications. If what I propose is at least 10 percent accurate, the proliferation of social thermoregulation wearables will remarkably extend current communications technologies by enabling us to send a bit of warmth along with our voice and video image. Of course, we will need to determine when this remote temperature-manipulation technology is appropriate—and when it is not. Skyping or FaceTiming a dose of warmth may work with your happily-ever-after partner, or perhaps even your partner for the night. It may work less well in an online job interview with strangers.

Humans are extremely complicated. The general principles of social thermoregulation are true and real, but how, in what situations, and with whom we can apply sensors to measure and actuators to manipulate temperature requires more study. I have great confidence that we will become proficient in this application within the next 5 to 10 years and that after that time we will be able to report on how therapies founded on the principles of social thermoregulation work, and work effectively.

Hans Rocha IJzerman, Ph.D., is an associate professor of social psychology at the Université Grenoble Alpes and the author of Heartwarming: How Our Inner Thermostat Made Us Human.

Reprinted from Heartwarming: How Our Inner Thermostat Made Us Human by Hans Rocha IJzerman. Copyright (c) 2021 by Hans Rocha IJzerman. With permission of the publisher, W. W. Norton & Company, Inc. All rights reserved.


We’re Getting Colder

by Walter Veit

A group of Stanford scientists led by Julie Parsonnet and Myroslava Protsiv recently explored the puzzling question of why average human body temperature has apparently decreased since the Industrial Revolution.

The common standard of 98.6°F dates back to mid-19th-century research by German physician Carl Reinhold August Wunderlich, but current measurements suggest that mean human body temperature is actually 97.9°F. Some researchers believe Wunderlich’s data must have been flawed, but others wonder if mean human body temperature has truly decreased. Why does it matter? The Stanford team believes the decrease may hold secrets regarding the improvement of human health and longevity.

Reviewing 677,423 body temperature measurements, the team was able to analyze 157 years of measurements spanning 197 birth years. The data showed that men born in the early 19th century had temperatures 1.06°F higher than men today. A similar effect was observed in women: a total decrease of 0.58°F. Overall, human body temperatures decreased about 0.05°F per decade.

The researchers maintain that these differences are unlikely due to changes in measurement methods: In Wunderlich’s time, temperatures were generally axillary—taken under the armpit—a method now known to deliver results about one degree Celsius lower on average than oral measurements. And yet, today’s oral measurements are the lower ones. Therefore, they argue, actual physiological change is more likely.

The researchers suggest this change is due to a decrease in our resting metabolic rate. Heat is a byproduct of basic metabolic processes. Humans have become larger in recent centuries, both in weight and height, which should have corresponded with an increase in temperature. Since it didn’t, the team argues, there must have been a decrease in our resting metabolic rate. Lower inflammation seems to be the best explanation, they believe: “Economic development, improved standards of living and sanitation, decreased chronic infections from war injuries, improved dental hygiene, the waning of tuberculosis and malaria infections, and the dawn of the antibiotic age together are likely to have decreased chronic inflammation since the 19th century,” with the effect of lowering average temperature.

The team also speculates that our temperature change may be evolutionary. More humans now live in temperate climates—88 percent of American households are equipped with air conditioning, and heating is almost universal—so there may be less need for a higher metabolic rate to ensure a stable body temperature with minimal energy expenditure. On the other hand, it may be a stretch to suggest that an actual evolutionary change has occurred in such a relatively short span of time.

Wunderlich’s classic finding, then, wasn't inaccurate. The new research shows that human body temperature can change, and has changed, with improvements in standard of living and access to health care. And so we have a new mean temperature: 97.9°F.

Walter Veit is a Ph.D. student in the history and philosophy of science at the University of Sydney.

Submit your response to this story to If you would like us to consider your letter for publication, please include your name, city, and state. Letters may be edited for length and clarity.

Pick up a copy of Psychology Today on newsstands now or subscribe to read the rest of the latest issue.