ONE WILD WORLD — NEWSLETTER

Issue #023  |  5 Myths You Still Believe  |  May 2026

It is one of the most repeated findings in popular psychology. Women feel more. Women understand others better. Women are wired for empathy. The trouble is that the entire claim rests on a single type of evidence — questionnaires asking people how empathetic they think they are — and the moment researchers swap the questionnaires for actual tests, the gap shrinks dramatically. In brain scans, it disappears entirely. The most-cited gender difference in social psychology is, in significant part, a measurement artefact.

BY THE NUMBERS

In 1983, two American psychologists named Nancy Eisenberg and Randy Lennon published a landmark meta-analysis in the journal Psychological Bulletin. They had pulled together every study they could find on gender differences in empathy. They sorted those studies by what kind of measurement each one had used. And they found something interesting.

On self-report questionnaires — where you sit a participant down and ask them how empathetic they are, how often they feel for other people, how moved they are by suffering — women scored markedly higher than men. The effect was large and consistent across studies.

On more objective tests — physiological measurements of arousal in response to others' distress, behavioural observations of how people actually react when someone is upset, tasks where participants don't know empathy is being measured — the gap shrank to almost nothing. In some studies, it disappeared entirely.

Eisenberg and Lennon's conclusion, published over forty years ago, was careful: there might be a real difference in empathic capacity between men and women, but the size of the difference depended enormously on how you chose to measure it. The further you got from people simply telling you about themselves, the smaller the gap became.

If you want to measure empathy without a participant's social filter in the way, you can skip the questionnaire entirely and look at the brain itself. When a person sees another person in pain — a needle going into a hand, a face contorted in distress — a specific network of brain regions lights up. The anterior insula. The mid-cingulate cortex. The temporoparietal junction. These are the neural signatures of empathic response, and they have been mapped in hundreds of fMRI and EEG studies.

So researchers asked the obvious next question. If women really are more empathetic at a deep biological level, their brains should show it. The empathy-related neural response in women should be larger, faster, or more sustained than in men.

It isn't. A series of well-controlled neuroimaging studies, going back to the early 2000s, have failed to find any reliable gender difference in the basic empathic brain response. When men and women watch someone in pain, their anterior insula activates by roughly the same amount. Their mid-cingulate cortex shows the same pattern. The neural circuitry that produces empathy is, as far as the imaging can tell us, identical between the sexes.

On a questionnaire, the empathy gap is enormous. On a brain scan, the empathy gap is gone.

If the difference is in self-report rather than in the underlying capacity, then it should be possible to make the gap appear and disappear by manipulating the social context. And it is.

In a series of clever studies starting in the 1980s and continuing into the 2020s, researchers have shown that the gender empathy gap depends almost entirely on whether participants know the test is about empathy. When participants are told they are being assessed on emotional sensitivity — the explicit empathy framing — women score much higher than men. When the same task is presented as a measurement of, say, attention or perception, the gap collapses.

This is the 'motivation' explanation, first proposed by William Ickes and his colleagues in 2000. Both men and women have roughly equal empathic ability. But women are more motivated to demonstrate it, and men are more motivated to suppress or downplay it, because of what each gender has been taught about how empathy fits with their identity. Take away the social cue, and the gap takes itself away with it.

A 2025 study at the University of St Andrews tested this directly. Researchers gave participants the same empathy questionnaire under two conditions: one where the form was explicitly labelled as measuring 'empathy', and one where it was labelled with a neutral cover term. Both men and women scored higher when the word 'empathy' appeared on the form — but the women's scores went up more. The effect of the label was about half as strong as the effect of being female. The label was doing almost as much work as the gender.

Here is the part of the literature that almost nobody quotes, because it cuts directly against the popular story.

If empathy translates into prosocial behaviour — actually helping people in real life — then the more empathetic gender should help more often. So researchers have run dozens of studies measuring helping behaviour in real-world or naturalistic settings. Picking up dropped objects. Stopping for someone in distress. Offering assistance to a stranger.

In 1986, the psychologist Alice Eagly and her colleague Maureen Crowley published a meta-analysis of 172 such studies, covering tens of thousands of participants. The results were striking. In the majority of helping situations studied, men were significantly more likely to help than women. Not slightly. Significantly.

The interpretation matters. Most of those studies involved offering help to strangers — often male strangers — in public settings, sometimes involving physical risk or assertive action. These are scenarios where men have been historically socialised to act, and where women may reasonably feel unsafe. The finding does not mean men are 'more empathetic' than women. It means that the relationship between empathy and behaviour is heavily mediated by context, by safety, by social roles, and by what each gender has been told helping looks like.

Empathy and helping are not the same thing. The popular story collapses them. The data does not.

It is worth being precise here, because this is an area where any oversimplification immediately becomes either flattery or insult.

The honest summary, from forty years of accumulated research, looks something like this. Women score consistently higher than men on questionnaires asking about empathic tendencies. Some of this is real — there are small differences in how readily men and women report and discuss emotional experiences. Some of it is identity — women have been taught that empathy is part of who they are, and they answer questionnaires accordingly. Some of it is what social psychologists call 'demand characteristics' — participants giving the experimenter the answer they think is expected.

On more objective tests, a smaller difference remains. On the most rigorous measures — neuroimaging, naturalistic behaviour, tasks where participants don't know empathy is being assessed — the gap shrinks further or disappears altogether. Across the board, the size of the female empathy advantage is roughly proportional to how much the test depends on a person's beliefs about themselves rather than on what they actually do.

Real-world helping behaviour, the closest behavioural correlate of empathy, shows men helping more often in stranger-helping scenarios. Women probably help more often in close-relationship or care-taking scenarios — though that has been less rigorously studied, partly because it is harder to set up in a laboratory and partly because researchers, until recently, did not think to ask.

This is not a story about whether women are 'really' more empathetic than men. The honest answer to that question is 'sometimes, slightly, depending on the measurement and the situation, and the difference is much smaller than the cultural conversation implies.' That is not a satisfying answer. It does not fit on a magazine cover. It does not power a self-help book. It is, however, what the evidence actually says.

The reason it matters is what happens when the simpler version of the story gets imported into other domains. If empathy is gendered, then jobs that supposedly require empathy — caring, teaching, social work, medicine — get gendered along with it. Men who do well in those jobs are treated as exceptions; men who don't are treated as confirming the rule. Women who are not particularly empathetic are seen as somehow failing at being women; women who are highly empathetic find themselves with empathy-shaped roles whether they want them or not. A measurement artefact becomes a labour market. A questionnaire effect becomes a stereotype. A research finding that depended on what label was at the top of the form ends up shaping who gets hired for what.

The gap is real on the questionnaire. The questionnaire is not measuring empathy. It is measuring a story about empathy that we have all been told.

A 1983 meta-analysis. Forty years of follow-up studies. Brain scans that show the same response in both sexes. Real-world helping experiments where men out-help women more often than not. And a popular story that survives all of it because it is easier to remember than the truth.

Women are not more empathetic than men. Women report more empathy than men. The space between those two sentences is where almost the entire popular conversation lives.

SOURCES

Eisenberg, N. & Lennon, R. — "Sex Differences in Empathy and Related Capacities", Psychological Bulletin, 94(1), 100–131 (1983)

Eagly, A.H. & Crowley, M. — "Gender and Helping Behavior: A Meta-Analytic Review of the Social Psychological Literature", Psychological Bulletin (1986)

Ickes, W., Gesn, P.R. & Graham, T. — "Gender Differences in Empathic Accuracy: Differential Ability or Differential Motivation?", Personal Relationships (2000)

Han, S. et al. — "Empathic neural responses are modulated by the perceived fairness of others", Nature (2010)

Lamm, C., Decety, J. & Singer, T. — "Meta-analytic evidence for common and distinct neural networks associated with directly experienced pain and empathy for pain", NeuroImage (2011)

Christov-Moore, L. et al. — "Empathy: Gender effects in brain and behavior", Neuroscience & Biobehavioral Reviews (2014)

Yang, Y. et al. — "Are women more empathetic than men? Questionnaire and EEG estimations of sex/gender differences in empathic ability", Social Cognitive and Affective Neuroscience (2023)

Brown, G.R. et al. — "Gender difference in self-reported empathy: Effects of task instructions and exposure to gender essentialism primes", PLOS One (2025)

Heck, P.R., Brown, M.I. & Chabris, C.F. — "A robust negative relationship between self-reports of social skills and performance measures of social intelligence", Social Psychology Quarterly (2024)

Murphy, B.A. & Lilienfeld, S.O. — "Are self-report cognitive empathy ratings valid proxies for cognitive empathy ability? Negligible meta-analytic relations with behavioral task performance", Psychological Assessment (2019)