The limits of empathy

Diana Kwon on when walking in another’s shoes is not enough.

Snapshots of the horrifying aftermath of terrorist attacks, refugees fleeing their war-torn homes, and families mourning a victim of police brutality can be gut-wrenching. Many people consider this ability to understand and feel what others are feeling, or empathy, as a primary source of morality and the glue that holds societies together. President Barack Obama has described empathy as the ‘heart of my moral code’ and has suggested that an empathy deficit is at the heart of many of our society’s problems.

Empathy is a key component in our relationships, and in many situations, it does motivate people to help others in need. However, this is not always the case. Recent psychological studies suggest that empathy is not quite the societal cure-all we often believe it to be. While it can help promote cooperation and motivate prosocial behavior, in some cases, empathy can actually deepen divisions been groups and inspire aggression against others.

A force for good?
Most people see empathy as a good thing. Thinking about the absence of empathy conjures up an image of a cold-blooded killer or ruthless con-artist with no regard for other’s emotions or well-being. And indeed, a long line of studies has shown that empathy can evoke prosocial behavior. Some of the earliest experiments were conducted in the 90s by the social psychologist C. Daniel Batson, one of the leaders of empathy research, and his colleagues. In one study, they asked participants to imagine how a person from a stigmatised group – such as a person with AIDS, a homeless man, and even a convicted murderer – was feeling, finding that this experimental manipulation could improve attitudes towards such individuals. Other groups have also shown that feeling empathy can help reduce the will to harm others and improve intergroup relations.

‘Considerable evidence supports the idea that empathic concern motivates helping directed toward reducing the empathy-inducing need,’ says Batson. In fact, he points out, many novels were written with the goal of inducing concern for outgroup members by putting the reader in their shoes. ‘Often what [the writers] are trying to do is create this caring for, this valuing of the other’s welfare to induce concern,’ says Batson. ‘The reader knows this is a fictitious character, but those feelings can then generalise [to others].’

One notable example he points to is Uncle Tom’s Cabin, a widely-read abolitionist novel by Harriet Beecher Stowe published in 1852. ‘[This book] is probably considered to be the work of fiction written in English that has had more impact in changing public policy than any other,’ Batson told me. Stowe’s highly influential book helped raise awareness about the harms of slavery, and some historians have even argued that it acted as a catalyst for the Civil War that came less than a decade later.

In the same way, contemporary artists use various forms of media to cultivate awareness and concern for mistreated groups of people. For example, the popular television show Orange is the New Black allows viewers to delve deeply into the lives of trans, homosexual and minority inmates while shedding light on the real issues facing these groups in the American criminal justice system.

While empathy can be a strong motivator for morally good or altruistic behaviour, its influence can also go in the opposite direction. The notion that empathy is not always a force for good was recently popularised by psychologist Paul Bloom at Yale University, with widely discussed pieces in popular media outlets like the New Yorker and the Boston Review as well as a recent book, Against Empathy. Bloom’s central argument is that empathy, which he defines as ‘feeling what other people feel’, is not the best guide for making moral decisions.

Bloom is not the first to take this stance on empathy. The philosopher Jesse Prinz made a similar argument in a 2011 essay, where he contends that empathy is not necessary for moral judgements. Even before them, famous thinkers like Immanuel Kant argued more generally that when it comes to making moral decisions, rational considerations trump emotional reactions.

Recent evidence supports this notion. In certain conditions, rather than motivating prosocial behaviour, empathy fosters hostility and aggression. In one 2014 study, psychologists at the University of Buffalo led by Anneke Buffone found that when participants felt o inflict pain on that person’s competitor in a math test – a relatively non-threatening environment – even when the competitor posed no direct threat. ‘People are punishing emotionally rather than to restore the victim,’ Mina Cikara, a psychologist investigating intergroup neuroscience at Harvard University, told me.

So does empathy make us do good or bad? Some studies suggest neither. One meta-analysis of empathy and aggression studies, led by psychologist David Vachon at the University of Minnesota, revealed that there is virtually no relationship between having low empathy and being malicious across various types of aggression, including verbal, physical and sexual attacks. ‘It turns out that if I want to know how likely you are to help people or give to charity or be a good person, knowing how empathic you are will tell me very little,’ says Bloom.

Even stories, which are powerful methods to induce empathy for oppressed or mistreated groups, are not always used for good. This is evident in political rhetoric, where politicians like Donald Trump use empathy to manipulate. Trump harnesses the strong emotional responses evoked by drawing attention to victims of terrorist attacks in Western countries to encourage people to support anti-immigration policies and turn away refugees. ‘Donald Trump talks a lot about people who are assaulted by illegal immigrants, raped or murdered,’ Bloom says. ‘I wish to some extent that the population could become more immunised against that sort of emotional appeal.’

Friends and foes
Empathy’s limitations become most apparent in the context of conflict and competition. Empathy is biased – we are more likely to empathise with those who are from similar social, racial and political circles. Engendering a strong empathic response for atrocities towards in-groups is a potent tool to mobilise people to a cause. ‘Empathy has been historically used as a major tool to spur people to war,’ says Bloom.

In competitive situations, rather than feeling sadness or distress at the sight of a suffering out-group member, people tend to feel pleasure at another’s pain, or schadenfreude, and will not feel motivated to aid them. There are even separate neural circuits that determine how we react to another group’s suffering. One 2010 study led by psychologist Grit Hein, who was then at the University of Zurich, found that distinct neural responses in brain areas associated with empathy predicted whether football fans were willing to endure pain to help supporters of the same team or fans of a rival team.

According to Cikara, it is not simply the dangers of low empathy towards outgroups, but the risk of extreme empathy for in-groups that can lead people to take extreme measures, such as sacrificing themselves and hurting others in the process. ‘This is interesting because it suggests something counterintuitive, which is that […] maybe one way of attenuating bias between groups is actually to get people to be a little bit less responsive to in-group suffering,’ Cikara told me.

Recent investigations by Cikara and her colleagues support the notion that intergroup conflicts could be mitigated by reducing the gap between empathy felt for one’s own group and those they are in conflict with. For example, they found that shifting people’s focus away from their group membership using short descriptions of the individuals in both groups successfully reduced this bias.

Regulating empathy, cultivating compassion  
Empathy is a powerful tool, so how can we harness its power for good? Most of us think of empathy as an automatic, uncontrollable response to the pain and distress of those around us. Experimental evidence from infant and animal studies suggest that empathy is innate: babies will cry when they hear another baby crying and rats will help free a fellow rat trapped in a cage without training or the promise of a reward.

Recent evidence, however, suggests that we can regulate how much empathy we feel. Jamil Zaki, a social psychologist at Stanford University, and his colleagues have found that when people believe empathy is under their control, they are more likely to empathise even in difficult situations, such as responding to someone with opposing sociological views or listening to emotional stories by someone from a racial outgroup. ‘It turns out that simply believing that empathy is something that you can change seems to get people to put more effort into empathising, especially in cases […] when empathy might not naturally help people do the right thing,’ Zaki told me.

According to Zaki, people already regulate empathy all the time. Doctors, he says, may tune down their empathy to avoid burning out from feeling too much of their patients’ pain. Like other emotions, being able to tune our empathic responses in certain situations might help harness its potential benefits. Regulating emotions can help improve political attitudes in conflicts – one group of psychologists found that training Israelis in emotion regulation made them more likely to support conciliatory rather than aggressive strategies in Israeli-Palestinian policies.

In a similar way, learning to control our empathic reactions may help improve intergroup relations. If you recognise a politician using emotionally manipulative stories, you might want to turn your empathy down. On the other hand, you would want to vamp it up in situations where you are faced with people from different groups or backgrounds.  

Alternatively, psychologists like Bloom and Tania Singer, a neuroscientist at the Max Plank Institute for Human Cognitive and Brain Sciences in Germany, believe in a different strategy altogether. Rather than working with empathy, they believe that cultivating compassion – a more distanced form of care and concern for others’ well-being – is a more effective solution.

Singer and her colleagues have been working on compassion training techniques by drawing on the Buddhist practice of loving-kindness mediation, which involves quietly concentrating on extending caring feelings from loved ones to strangers and eventually to all living beings. Studies reveal that compassion, unlike empathy, does not suffer from the same type of limitations. It also activates a completely different network of brains areas and increases prosocial behavior while improving emotional well-being.

Finding empathy in the brain
Psychologists define empathy in myriad ways, and some see compassion as a component of empathy rather than separate from it. ‘You can find almost as many definitions of empathy as you can find people writing about empathy,’ says Cikara.

Empathy is often described as a combination of three factors: cognitive empathy (thinking about another’s emotions), emotional empathy (sharing another’s emotions), and motivational empathy (caring about another’s emotions – or compassion). ‘Empathy is not just one thing, but rather it’s an umbrella term that describes the different ways that people respond to each other’s emotions,’ Zaki explains.

Each of these, though closely connected, are actually independent psychological and neurological processes. In recent years, as researchers have started to probe the brain to better understand how empathy works at the neural level, studies are beginning to show how these three components can be teased apart in the brain.

At the very basic level, neuroscientists have found that empathy stimulates shared representations in the brain – participants activate the same neural areas in response to feeling pain and observing others in pain. Social neuroscientists like Claus Lamm at the University of Vienna have been using a variety of neuroimaging techniques such as functional MRI to study the underlying brain mechanisms of empathy. The classical approach for studying this, according to Lamm, is the pain paradigm, where researchers place participants in an MRI scanner and record their neural activity as they receive painful shocks or observe others getting a painful shock.

In recent studies, Lamm and his colleagues have discovered that these activations can be artificially manipulated. One functional MRI study revealed that giving participants placebo painkillers decreased activation in brain areas associated with pain and empathy for pain. Further, they found that opioid-blockers could block the placebo’s reduction for both one’s own pain responses as well as the empathic response for another’s suffering.

Studies have also shown that cognitive empathy, emotional empathy and compassion all activate unique networks in the brain, and that this activation predicts different behavioral outcomes. ‘What the neuroscientific investigation do is that they basically confirm what the social psychologists have suggested for quite some time already,’ Lamm told me.

A force for good and evil
Perhaps the best way to think about empathy is as an entity separate from morality. Batson himself suggested this in a 2009 article where he wrote: ‘Empathy-induced altruism is, we suggest, best thought of as neither moral nor immoral, but amoral.’

Empathy is a powerful force, capable of doing good and harm. Some psychologists believe that humanity would best thrive if we avoided it all together and relied instead on rational, reasoned thought. Other see a more delicate balance – that both are necessary to make the world a better place. ‘What Paul’s saying is that if you’re going to get fairness or justice in the society, empathy-induced altruism is not necessarily the best way,’ says Batson. ‘My own bias is that reasoned moral thought alone isn’t the best way either. The kind of change Paul’s talking about takes the two working together.’   

Overall, it is important to know when to empathise and to assess the motives of people who try to stir our emotions with a critical light. Learning to numb our reactions to stories some politicians tell may ultimately help make our world a better place.

Box: Rethinking disorders of empathy 

Society often stereotypes people with autism and Asperger’s syndrome as antisocial, unemotional and lacking in empathy. The scientific evidence, however, reveals that while people with autism may have difficulties with cognitive empathy (understanding what another person is feeling), their emotional empathy (the ability to feel what another feels) is intact. 

A growing body of research suggests that the reduced ability to empathise in some individuals with autism is actually due to alexithymia, a separate condition that impairs emotional processing. Alexithymia is present in around 10 per cent of the general population and approximately half the people with autism. 

One 2011 study, for example, led by psychologist Geoffrey Bird at King’s College London, revealed that gaze avoidance – the tendency to spend less time scanning parts of the face that display emotion, such as the mouth and eyes – was a feature of alexithymia rather than autism. 

In another recent study a group of neuroscientists in Italy and Austria found that participants with autism displayed similar empathic responses to moral dilemmas as those without the condition. In fact, those with autism displayed stronger emotional distress when faced with a utilitarian dilemma (sacrificing one to save many) and were less likely to endorse options that that caused direct harm to another person. 


People with psychopathy are also often defined as being callous and without empathy. Popular depictions of psychopaths include serial killers like Ted Bundy or the fictional Hannibal Lector.

In recent years psychologists have started investigating the question of whether individuals with psychopathy lack the ability to understand and feel what others are feeling completely or are just less likely to do so in certain situations. Evidence suggests the latter – recent neuroimaging studies show that while individuals with higher levels of psychopathy are less perturbed by emotional stimuli, when primed to attend to emotional cues, the differences between psychopaths and non-psychopaths largely disappear.

In one 2013 Brain study, for example, Christian Keyser’s group at the Netherlands Institute for Neuroscience found significantly less neural activity in participants with psychopathy who passively viewed videos of emotional hand interactions compared to non-psychopathic subjects. However, when the experimenters asked them empathise with the actors in the clips, these differences are significantly reduced. 

Psychopaths make up a disproportionate amount of the incarcerated population. Understanding the neurobiological origins of psychopaths may help rehabilitate offenders and create early prevention systems. If psychopathy is the result of the reduced propensity rather than the lack of ability to empathise, training these individuals to attend to emotional stimuli may prevent antisocial behaviour later in life.

- Diana Kwon is a science reporter. 

Illustration: Ana Louis.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber