As a grid of video feeds blinks into view, attendees across the country prepare for an ideological collision. All have signed up for a virtual forum billed as an “empathy cafe,” held to spark dialogue between police and community members. Among the participants are officers as well as people who’ve been burned in encounters with law enforcement.
The setup seems like a guaranteed powder keg. But as moderator Lou Zweier explains, this forum has some strict rules of engagement. “We’re going to do four-minute speaking turns,” he explains to the group, which will be separated into smaller breakout rooms. After one person in each breakout room gets a chance to speak about what’s on their mind, someone else in the room—a person chosen as the “reflector”—will sum up the speaker’s opinions and concerns as best they can, whether or not they agree. The reflector then becomes the next speaker and chooses a new reflector, and the process continues. “The listening and reflecting go around the circle,” Zweier says. “Everyone gets a chance to speak and be heard.”
This event, led by empathy educator Edwin Rutsch, offers a chance for minds to meet across the kind of yawning divide that’s grown commonplace in the U.S. Such forums have popped up in part because trying to understand someone else’s perspective doesn’t always seem like a social bet that pays off.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Empathy is often defined as the capacity to understand what someone else is thinking and feeling. It is distinct from sympathy, which may imply pity (you might feel sympathy for someone in pain without grasping what they’re going through), and from compassion, which involves a desire to ease someone’s plight.
Because empathy can allow people to connect across political, racial and economic divides, it lays a foundation for acts of cooperation and caring that allow diverse societies to flourish. Higher levels of empathy are tied to both individual well-being and broader social cohesion.
When psychologist Sara Konrath set out to investigate empathy in the U.S., she found that it had been in decline for decades. She tracked Americans’ self-reported empathy levels between 1979 and 2009 and found that people were increasingly less likely to agree with statements such as, “I sometimes try to understand my friends better by imagining how things look from their perspective.”
Konrath’s follow-up analysis, which tracked empathy levels between 1979 and 2018, did show rebounds in young people’s willingness to take others’ viewpoints and understand their feelings. But research highlights social and biological factors that continue to make empathy daunting. Polarization has been increasing, meaning that people see the world in fundamentally different ways and trust one another less. What’s more, recent studies show that people shrink back from the mental effort it takes to understand what someone else is thinking and feeling. Meanwhile rates of loneliness, resentment and depression in the U.S. are high.
To promote empathy as a collective good, researchers have rolled out a smorgasbord of education programs. There are classroom programs for elementary schoolers, training seminars for employees and even immersive empathy retreats. New research shows that empathy instruction can boost people’s ability to engage reflectively across political divides. Yet fully grasping someone else’s experience is a heavy cognitive lift.
Empathy has long been an adaptive way of ensuring social cohesion. It also exacts a steep cognitive price.
Increasing empathy, says Stanford University social psychologist Jamil Zaki, will take more than teaching skills such as listening actively to others. Empathy is a socially motivated process, Zaki and other researchers say, meaning that people won’t necessarily empathize just because they know how. Instead—much as kids with athletic peers often want to excel at sports—people want to understand others when they enter into communities where empathy is the established norm.
Neuroscientists are beginning to piece together a clearer picture of empathy’s neural origins. When researchers elicit empathy by, say, showing people a film clip about what someone else is going through, a series of interconnected brain regions activate on functional MRI scans. Among these regions are the dorsomedial prefrontal cortex, which helps to gauge other people’s emotional states, and the anterior insula, which is involved in processing pain. Similarly, researchers have identified single neurons in the dorsomedial prefrontal cortex that encode information about others’ thoughts.
Our ability to perceive other people’s inner states most likely evolved because it helps to forge the kinds of strong social ties that promote survival. In human ancestral environments, nomadic groups understood one another’s emotions and state of mind; the bonds among them deepened, helping the group to function as a resilient unit.
Whereas empathy has long been an adaptive way of ensuring social cohesion, it also exacts a steep cognitive price. Taking someone else’s perspective is a complex, challenging operation for the brain, in part because it requires a sophisticated assessment of what the other person may be thinking and feeling. In one 2020 study at the University of Liverpool in England, researchers found that empathy for others’ pain requires a host of different brain networks to interact, including some responsible for inferring others’ mental state. (When people perceive their own pain, on the other hand, their brain activity related to understanding others diminishes.)
A series of experiments led by Pennsylvania State University psychologist C. Daryl Cameron found that most individuals prefer to opt out of the cognitive effort empathy requires, especially if they don’t know the other person well. In multiple rounds of game play, Cameron offered participants in a study two card games to choose from: an “objective deck” game, which asked them to describe the appearance of people on the cards, or an “empathy deck” game, which asked them to describe the people’s possible experiences and feelings based on their expressions. Most people stated that they preferred the objective deck.
In part, that’s because it’s harder to empathize with someone who feels distant or unknown than with a close loved one. “The more shared experiences you have with someone, the more of a rich, nuanced representation you can draw on,” Cameron says. But empathy for someone whose experience feels alien—the person who disagrees with you online, the man in a tent outside the subway or even a cousin who spouts extremist views—is a different matter. A host of disquieting unknowns arises: Is identifying with this person going to put you in danger? Will it compel you to sacrifice something important, such as time, money, tranquility?
When such anticipated costs overwhelm people, they’re more prone to withdraw altogether rather than trying to understand where the other person is coming from. “We are quite adept at learning how to manage our emotional environments to cultivate what we want to feel,” Cameron says. “Empathizing with a stranger, taking on their experiences—either negative or positive experiences—people find it difficult, they find it costly. And the more they feel that way, the less they opt in.” In a 2020 study at the University of Lübeck in Germany, fMRI scans of people who’d just heard stories about mass tragedies showed less activation in their brains’ core empathy networks compared with those who had not heard the disturbing stories.
But larger structural forces are likely at play, too. Wealth inequality in the U.S. has steadily risen since the 1980s, and people in rarefied income brackets often have little motivation to understand the struggles of those at the poverty line. “We’re much more segregated economically nowadays,” Konrath says. “That can impair our ability to see and to care and to have those people be our neighbors and friends that we naturally want to help.”
The impulse to sidestep empathy’s complications also leads people into polarized online echo chambers—many of which also persist in the physical world—where we’re less and less likely to maintain friendships with those whose views differ from our own. It’s easier than ever to be a water strider, gliding away from others with frictionless ease.
That suspicion and detachment are what Rutsch, a former computer systems administrator turned empathy educator, aims to dissolve through empathy cafes and other similar events. He founded the Center for Building a Culture of Empathy in 2010, with an aim to create a headquarters for the global empathy movement. Rutsch based his approach on that of humanistic psychologist Carl Rogers, who used reflective listening techniques to build trust and rapport with clients.
Rutsch has traveled around the country with pop-up “empathy tents,” which he pitches near demonstrations and protests. Once he pitched a tent just beyond a 1,000-person rally in Los Angeles for former president Donald Trump, which was also attended by large numbers of counterprotesters. Rutsch invited in six people from each side. Then he mediated six pairs in listening carefully to one another, then stating their own understanding of the other person’s thinking. “Of those six pairs, five of them ended up giving each other hugs afterward,” Rutsch says. “On the other side of the street, they were screaming and yelling at each other, and the police were having to keep them apart.”
Rutsch attributes such outcomes to the intentional structure of these chats. Even if the other person says a false or off-the-wall statement—for example, that the 2020 election was stolen—“you reflect back your understanding of what they have said,” he says. “When it is your turn to speak, you can challenge what they have said. They have to take it into their consciousness to be able to reflect it. That means they have to listen to you. They cannot live just in their own worldview.”
In these dialogues, listening and being carefully listened to in return often begins to soften conspiracists’ armor. “Knowing that you were willing to listen to them, they often drop their judgments after a while and get more real,” Rutsch says. “You get to a deeper understanding of each other and see each other’s humanity. You may get to a deeper fear, perhaps, that is the reason for the lie.”
Many researchers have devised programs that similarly help participants hone specific empathy skills. Some, for adults, focus on how to empathize in work interactions; others, for elementary-aged students, teach the nuts and bolts of how to take another person’s perspective.
In Roots of Empathy, now offered in hundreds of schools across the U.S. and around the world, a local family brings their young baby into the classroom once a month, and trained instructors guide students to practice “perspective taking” by identifying what the baby might be thinking and feeling at different times. In studies, elementary schoolers enrolled in Roots of Empathy were better than control students at understanding others’ emotions. They also proved more likely to help others in the classroom by the end of the school year, based on reports from their peers.
Yet established empathy programs such as these often rest on the assumption that once people have empathetic skills in their arsenal, they’ll be more apt to put them to use. That’s not always the case, says Harvard social psychologist Erika Weisz. Studies show that even when people know how to empathize intellectually, they may not exercise that ability unless they truly feel the desire to do so. If they expect empathy to be costly or unpleasant, for instance, some people will refrain from it no matter their training or skill level.
Like Zaki and Cameron, Weisz frames empathy as a socially motivated process—one that’s dependent not just on what someone knows about empathy but on how compelled they feel to show it. She’s found that another way to nudge people toward empathy—and keep them there—is to embed them in communities where empathy is a baseline expectation. “People want to increase their empathy if you tell them, essentially, it will help them socially,” Weisz says. “That is a perfectly reasonable leverage.” Unlike empathy skills training, which teaches specific methods of relating to others, Weisz’s approach involves building communities that value and reward empathetic behavior. It draws on a kind of constructive peer pressure.
In a pilot program at four California middle schools, Weisz tracked the effects of establishing empathy as a social norm among students. She held three virtual workshops where seventh graders completed activities such as reading stories their classmates wrote about why empathy was important to them. Several weeks after the workshops, students in the program’s social-norm group proved more motivated to show empathy toward others.
Weisz attributes these results to the relative ease and simplicity of following a social norm—as opposed to, say, practicing a just-learned empathy skill every day. “My enthusiasm about motivated empathy interventions comes from the fact that they complement people’s existing daily lives,” Weisz says. “You don’t need to completely add a new variable. It’s just like riding a wave that’s already cresting.”
This approach informs the empathy program at Third Street Elementary School in Los Angeles, which I visited last spring. Drawing on a curriculum from Harvard’s Caring Schools Network (CSN), Third Street teaches students empathetic skills—and crucially, the students learn to exercise those skills within a community that models empathy at every level. “Character education programs are in a sense about literacy, that kids know right from wrong,” says CSN psychologist Rick Weissbourd. “We’re more focused on identity or moral motivation: What makes someone want to be a good person in the world, or what motivates someone to care for other people?”
During a perspective-taking exercise, Laura, a parent volunteer in one of Third Street’s fourth-grade classrooms, asked students: “What would you do if you saw a student tease another student because of what they’re eating?” The kids then sorted themselves into groups around the room based on how they’d answer: ask an adult for help, ignore the situation, tell the unkind kid to stop, or check in with the person who was teased.
“The person who teases the other person might keep doing it,” said one boy, arguing for telling the offender to stop.
Not everyone agreed. “It really honestly depends on who it is,” one girl says. “If it was my friend, I would probably go over there.” If it’s not her friend, she added, she would just get out of the way.
“You don’t want to make it a bigger problem than it already is,” another girl adds.
As each student chimed in, the rest listened attentively, taking in the conflicting opinions. Through this kind of habitual, focused listening, Weissbourd explains, students learn to better appreciate where others are coming from. And it’s not just adults leading this dynamic. If a student is having a beef with someone else, they can approach any kid on campus wearing an orange hat. These peer counselors will listen carefully to their concerns and help them solve the problem.
In surveys at U.S. schools, students who participate in CSN curricula report being more helpful than control-group students, and their listening and perspective-taking skills improve. Third Street is a cocoon of sorts, a chance for kids to marinate in empathetic community. That could help prepare them for more challenging encounters later on, to cultivate empathy for those with whom they disagree profoundly.
“The kind of empathy we see in everyday life, a lot of it is you embedding empathy within valuable social relationships,” Cameron says. “One approach might be to think about using that relational value as a starting point and then going to the harder places—extending that out to someone you don’t know or even someone who’s an enemy.”
Socially modeling empathy affects not just how community members behave but the way their brains work, says neuroscientist Grit Hein of Germany’s University Hospital Würzburg. In a 2024 study, Hein’s adult subjects watched videos of people getting hit with an intense burst of air and reacting in pain. People who watched a person respond empathically to the blast videos were likely to follow suit, whereas those who watched another person shrug off the videos acted equally blasé.
Hein found that people who witnessed an empathetic reaction were more likely to rate the recipient’s pain as high, whereas those who saw the low-empathy video rated the pain as low. Those who saw the social example of concern also had more activity in their brain’s anterior insula, which governs empathy processing, than people who saw the indifferent example.
“If you’re surrounded by empathic individuals, it really has an influence. It increases how your brain responds to the pain of another person,” Hein says. “The bad news is, it also works the other way around.” In other words, if you’re surrounded by people who are indifferent or hostile, you’re apt to mirror their social example as well. One 2023 study shows that people who identify as politically liberal have stronger empathetic brain activation than conservatives, raising questions about whether social norms within each political group might drive empathy differences between them.
Compared with Third Street students, attendees at Rutsch’s online cafe are scattered across the country and navigating political divides. The forum’s clear rules of engagement, however, create their own kind of fast-forwarded cultural norm. Only one person speaks at a time; the “reflector” must refrain from passing judgment; each person gets to choose their own fresh topic to discuss.
At first, the tension within the group is palpable. A community member named Sushila says she wants to know why police always dress like they’re going into battle. “If I were to see them in riot gear or carrying batons, that would make me very uncomfortable,” she says.
After Sushila speaks, Roger, now a lead official at Oakland’s Community Police Review Agency, tries to “reflect” what she’s saying. “You recognize that how you see them, specifically what you see them wearing, can potentially change that relationship,” he says. “You see the militarization as calling for a conversation, for the police to engage the community and explain why they see the equipment that they have as being necessary.”
When people feel heard, they tend to feel safer, allowing them to better process what others are saying.
What’s interesting about the discussions isn’t so much the reflectors’ input, which mostly mirrors what the speakers say. It’s how, over time, being intensively listened to—and intensively listening in return—seems to influence which new topics each speaker elects to bring up.
As the conversation continues, the participants’ stances shift toward curiosity—and even optimism. Sushila talks about planning events to help cops and community members establish a better relationship. She then recalls a time when a sheriff in her city encouraged this kind of rapport by acting in The Vagina Monologues. Sushila explains what that meant to her: “The fact that a police authority could show vulnerability and be so real ... I think she’s doing a great public service.”
For the remainder of the chat, people’s contributions center on how to forge relationships that benefit both police and community members—and how to keep confrontations between them from spiraling. “We need to get more law-enforcement officers to these events,” a security officer named John says toward the end, stressing that further similar exchanges could be valuable in bringing people together.
This dynamic parallels what political scientists Joshua Kalla of Yale University and David Broockman of the University of California, Berkeley, found in studies of more than 6,000 U.S. voters who chatted with canvassers about politicized topics, such as immigration and transgender rights. When canvassers engaged voters in typical back-and-forth arguments, few voters who had prejudicial opinions changed them. But when canvassers showed interest in understanding voters and asked them to share their perspectives, voters’ prejudiced views diminished for at least four months following the conversation. Likewise, Stanford psychologist Luiza A. Santos and her colleagues found that when people saw empathy as an asset in communicating with political opponents, they used more conciliatory language, and opponents were more likely to see their messages as persuasive.
The norms of polarized times, though, discourage such nuanced exchanges. Takedowns of opposing views get praised in activist circles and upvoted on social media, and civilly engaging with the other side can feel perilously close to endorsing harmful beliefs. But Kalla and Broockman’s research, as well as Rutsch’s forums, makes a surprising case for more empathetic, reflective social engagement: it’s being thoroughly heard, not condemned, that entices people to reject bigotry.
This style of listening, Rutsch emphasizes, does not mean absorbing others’ stances as your own. This kind of spongelike empathy is what Yale psychologist Paul Bloom rejects in his 2016 book Against Empathy.
When you take on someone else’s feelings, Bloom argues, those feelings rub off on you in ways that can interfere with logical decision-making—and even with helping. He also notes that too much empathy can be exhausting, draining people’s emotional resources in ways that put them off engaging with others. (Frontline health-care workers and others who witness trauma at close range may be especially vulnerable to this kind of fatigue.)
Bloom’s critics say empathy can coexist with this kind of deliberative reasoning—and doesn’t (or shouldn’t) involve identification with others to the point of exhaustion. Empathizing constructively means “sensing into the felt experience of someone else,” Rutsch says, “but it’s not like you’re taking it on to the point where you stop being present with them. It’s just about showing that you hear and understand the other person.” That understanding, in turn, can actually motivate the informed helping behavior Bloom calls for.
Cognitive science research helps to explain how such virtuous empathetic cycles can pick up speed. When people feel heard and understood, they tend to feel safer, and nervous system fight-or-flight responses recede, allowing them to better process what others are saying. A sense of safety may also help relieve the feeling of being overwhelmed and social angst that fuel the antiempathy bias Cameron describes.
Over time, reflective one-on-one dynamics feed into a broader environment where empathy starts to feel like its own reward. People grow compelled to understand one another in communities that model the practice—whether they’re groups of three or four, as in Rutsch’s online sessions, or entire workplaces and schools. Weisz hopes researchers can secure funding for future studies of how well social norm tweaks motivate empathy in settings like schools and workplaces.
After the empathy cafe breakout groups merge back into one, some attendees reflect on how to ease people into a practice that can feel, at first, like leaping into the abyss. “I think it’d be really valuable to do this as the first step in a longer process,” says a community member named Daniel. “Once you get used to hearing other people and knowing what different opinions are, knowing what different approaches are, then you can get to work on other things.”