The idea that heat is related to violence is as old as Aristotle. The Greek philosopher believed that anger was a boiling of blood and heat around the heart. A link between the two has been smuggled across the eons into modern languages. Our metaphors are thick with this connection: Rage is often described as incandescent. Tempers “flare.” People are “boiling mad.” A hothead is someone who loses their cool. In Spanish, you might call such a person a calentón. In fact, heat stress can trigger hormones like adrenaline, making you more irritable, and numerous studies have found that, at a societal level, as temperatures rise, conflict increases.
But behavioral and experimental economist Alessandra Cassar says the assumption that high temperatures lead to violence on an interpersonal level doesn’t hold up to scrutiny. Cassar and her colleagues recently published a study in PNAS Nexus that found that even when heat makes us cranky, it doesn’t make us behave in less prosocial ways toward one another. This was true across numerous cultures: The team studied people from the United States, Mexico, Colombia, Kenya, and India. Gender, on the other hand, made an enormous difference.
I spoke with Cassar, who studies cooperation and the ways disaster and conflict shape altruism, about what surprising variable makes women more competitive, how something called the “tend and befriend” instinct might be a stronger response to threats than “fight or flight,” and how we can prevent climate change from increasing violence around the world.
Your findings seem to contradict the general consensus in the research about the relationship between heat and violence. How far back does the research go?
The idea that heat drives violence has a long history in the research. In the psychology literature, it goes back at least 40 years. Some experiments do show an effect, and some of them are really clever. For example, my favorite was in Arizona. It was maybe about 15 years ago. In the scorching heat at 1 p.m., they put graduate students in a car at a traffic light. The instruction was to stop at the traffic light, and when it turned green, not to move. There was another graduate student on the side checking how long it would take for people in the back to honk. Depending on whether it was super scorching hot or somewhat less hot, there was much more honking and agitated behavior. They studied baseball, too, and found that when games are held on really hot days, there’s a lot more retaliatory behavior. But overall, the results were hit or miss, and the standards of evidence were much lower in the early days of this research than now.
What about at the level of entire societies?
More crime happens in the summer. Women get raped a lot more in the summer. There is more domestic violence. You might say, “Oh it must be that heat has an effect on human behavior.” But there are other hypotheses. In sub-Saharan Africa, for example, rising temperatures are linked to higher rates of conflict, while this link seems to be absent in northern Europe, but only if you exclude Russia. It could have more to do with resources. If you’re rich, it’s easy. You can turn on the air conditioning or plant different crops. But if you have few resources, there isn’t much you can do to adapt.
We wanted to understand what happens at the individual level so we said, “Let’s do a bunch of games,” because we can’t just give them guns and machetes. In our experiment, we made people play a game with cake called the Envy Game. In this game, you’re offered a way to split a cake with someone. The twist is you can choose to shrink the cake—meaning everyone gets less—just to prevent the other person from having more than you. It’s petty, but people do it in real life. To test what makes people more likely to behave this way, they ran the game under different conditions. Some participants were in a very hot room, and some participants played a rigged game where they thought they’d win but were then made to lose on purpose. Both the heat and the rigged game made people more irritated. But it didn’t make them less generous or cooperative.
Did this outcome surprise you?
If you step back a second, you wonder, is it even realistic to think that heat by itself may have such a powerful effect on human behavior? I don’t think it stands even from a theoretical point of view. How did humans manage to conquer every single part of the world—every niche, from the scorching hot desert, to the super cold that we have in Siberia or even in North America? It’s through cooperation. If anything, harsh outside conditions have taught people that the key to survival and to manage the stress and the risk of the environment is actually to cooperate. It’s not to attack each other and be nasty to each other.
The idea of Lord of the Flies is very, very old. It’s fabulous, maybe from a literature point of view, and I wouldn’t deny that it might happen occasionally. But in practice, when you put a bunch of humans in really difficult conditions, like in a postwar situation or after a natural disaster, what do they do? They cooperate. They get together with each other. They share the risk. They become more prosocial, more egalitarian. That’s a survivor strategy. It’s not to be even more competitive,violent, and less prosocial with each other because then both people are going to have less and less. It’s not an adaptive strategy. The adaptive strategy is always to cooperate, get together, and join forces.
You studied people from five different countries: the United States, Mexico, Colombia, Kenya, and India. What were the most surprising differences at the country level?
First of all, and this may come as no surprise, culture matters a lot. We sorted people into types based on how they think about fairness. Egalitarians want an even split no matter what, and they will even sacrifice their own resources to keep things balanced in either direction, while maximizers just want the total pot to be as large as possible, regardless of how it’s divided. Neither type is more virtuous than the other. It mirrors a tension that runs through the entire economic system, where capitalism excels at growing the pie but not sharing it, while planned economies prioritize equal slices at the cost of a smaller pie overall. And when they looked at the data across countries, maximizers won out. Most people, globally, care more about the total size of the reward than how fairly it’s split.
But across the five countries, almost nobody was purely selfish or spiteful. People genuinely care how others are doing, not just themselves. The biggest cross-country difference was in the maximizer tendency: The U.S. led the pack in not caring about equal splits, just wanting the overall pie as large as possible, while the poorer countries—India and Kenya—leaned more egalitarian. And after all that, when they finally checked whether heat made any difference to these preferences, the answer was essentially no—hot room or not, people’s fundamental attitudes toward fairness and sharing stayed the same.
Read more: “Are Some People Addicted to Revenge?”
You also found some pretty strong gender differences.
Yes, whereas heat barely moved the needle, gender turned out to be the sharpest dividing line in the data. Men were consistently more “competitive,” preferring the bigger-pie outcome across nearly all five countries, while women leaned more toward equal splits, a pattern that held even in the most egalitarian nations.
The really striking finding, though, came from a separate series of experiments: When the competitive game was tweaked to allow winners to share their gains afterward, women’s competitiveness shot up dramatically and matched men’s. This suggests women aren’t less competitive by nature, but are more sensitive to how competition is structured and what it costs the people around them. I’m speculating here, but I believe that this isn’t about women being morally better, but about a deeply ingrained difference in social strategy. Men are wired toward building large coalitions that require hierarchy and coordination, while women tend to prioritize the tighter, more reciprocal relationships that depend on fairness.
You mention in the paper this idea that “tend and befriend” is an ancient and innate way of dealing with conflict among women that’s in opposition to the popular “fight or flight.” This “tend and befriend” hypothesis was new to me. Is there strong evidence that this pattern is part of female biology, or is that more of a hypothesis?
This hypothesis came out about 25 years ago, but there isn’t that much research about it. There is some, though, that shows that even men tend and befriend more than fight or flight. As I mentioned earlier, when there’s a particular stressor such as heat, the best way to overcome it is to form a group, to find strength in social support. Maybe that’s even more important if there’s an enemy. That’s when you withdraw into your family, your friends, your group to receive help. Fight or flight may have been overblown in terms of our understanding of how frequent it is. When I measure these tendencies, I always find that tend and befriend is more common than fight or flight.
What do your findings ultimately tell us about what climate change might do to the incidence of violence around the world?
Climate change unfortunately will have an effect on conflict and violence, but not because it’s going to change the nature of people and make them automatically more violent. Heat could increase conflict because it’s going to reduce the economic opportunities of farmers. It’s going to reduce the economic opportunities for so many people for whom land and natural resources are important. And with poverty and reduced economic opportunities, we’re going to see more violence. But at the individual level, there is no direct link.
Knowing this at least gives us some idea about what we should do: Focus on economics more than on air conditioning, focus on helping poor countries develop. They have to help themselves. You cannot impose it from the outside, but it’s really focusing on economic opportunity, on pivoting toward jobs that aren’t so affected by agriculture and climate change, and helping people get together to become more prosocial and develop good,strong institutions. It’s only by becoming more prosocial and united that we’re going to be able to overcome the threat that is climate change. We need to increase our social capital. Social capital is a nice term that means a lot of things. It means having trust in the institutions of your country, trust in each other, and believing in your group.
This is worrisome because in the United States you aren’t going in that direction. Some people are getting richer and richer and richer. The top 10 percent earn half of all income in the country. The bottom 50 percent get about 2 percent of total income and are barely scraping by. There’s something really nasty happening. I’m very worried. Egalitarianism could be our safety net, but it doesn’t seem like there’s much political will. Maybe we could blame the heat.
If cooperation is so much a part of human nature, why is it so hard for humans to pull it off?
It’s very hard in the United States because it’s very big. If you’re in a smaller group, you know that Kristen, “She’s very nice. She always shared her fish.” It’s very easy to cooperate when you can put a name to a person and remember what the person did because reputation matters and social norms matter. You don’t want to see your best friend going hungry. You’re going to offer a bed and some food. But when you’re in a big group where it’s impossible to keep track of everyone, and you’re facing a random other person, and you see how nasty the average Joe is on social media, it’s very easy to mistrust. When you’re part of a huge group, you need solid institutions you can trust. It’s not easy. It’s so hard to support cooperation in a huge place like the United States. 
Enjoying Nautilus? Subscribe to our free newsletter.
Lead image: BRO.vector / Shutterstock
