The Psychology of Empathy — How It Works and Whether It Can Grow
Last reviewed by the Men Women Psychology editorial team.
The evidence
What the research actually shows
Mark Davis's influential work (1983) treats empathy as multidimensional rather than a single capacity. His Interpersonal Reactivity Index separates perspective-taking (understanding another's viewpoint), empathic concern (warm, other-oriented feeling), personal distress (one's own anxiety at another's suffering), and fantasy (identifying with characters). These components do not always move together, which is why someone can excel at one and struggle with another.
Judith Hall's meta-analysis of nonverbal decoding (1978) found that women on average were somewhat better at reading emotional cues from faces, voices, and gestures, a small but consistent advantage. Yet the size of this gap depends heavily on how empathy is measured: it tends to be larger on self-report questionnaires, where expectations about gender may shape answers, and smaller or negligible on objective performance and physiological measures.
Researchers increasingly distinguish accuracy from motivation. Some evidence suggests measured differences in empathy reflect not only ability but how much people are motivated to engage it in a given context — empathy can be turned up or down depending on goals and social expectations. This fits the broader finding that empathy is partly trainable, a set of skills that responds to practice and intention rather than a fixed quantity.
The mechanism
Why this happens
The components of empathy draw on different processes. Cognitive empathy relies on perspective-taking — deliberately modeling what another person knows and feels — while affective empathy involves a more automatic resonance with their emotional state. Because these run on partly separate systems, training one does not automatically strengthen the other, and people develop uneven profiles.
Average gender differences, where they appear, likely reflect a mix of socialization and motivation more than any large fixed gap. Many girls are encouraged from early on to attend to others' feelings and to talk about emotions, building practice in reading and naming them. When empathy is measured by how people describe themselves, gender expectations can further widen the apparent difference beyond actual performance.
Context strongly shapes empathy. Fatigue, stress, and feeling threatened all reduce the capacity to take another's perspective, while safety, similarity, and a clear cue that someone is struggling tend to increase it. This responsiveness to circumstance is part of why empathy is better understood as a skill exercised in conditions than as a fixed personal trait.
In practice
What this looks like in real life
One partner may be excellent at understanding why the other is upset yet feel little emotional pull to do something about it, while another feels the distress intensely but struggles to figure out its cause. These are different components of empathy — cognitive and affective — and a person can be strong in one and weaker in the other.
Someone who is exhausted or stressed may seem uncharacteristically unempathetic, missing cues they would normally catch. This usually reflects depleted capacity in the moment rather than a lack of caring, since perspective-taking is effortful and the first thing to fade under load.
People often rate themselves as more or less empathic than they actually perform when tested, because self-report is colored by self-image and expectations. The gap between how empathic someone believes they are and how accurately they read others is a recurring finding in the research.
Myth vs. evidence
What most people get wrong about this
A widespread misconception is that empathy is a single fixed trait you either have or lack. Research suggests it is multidimensional and partly learnable: perspective-taking and emotional attunement both improve with practice and intention. Framing empathy as a skill rather than a personality verdict opens the door to developing it.
Another error is overstating gender differences. The advantage women show on average is modest, appears largest on self-report rather than objective tests, and likely reflects socialization and motivation as much as ability. Treating empathy as something women have and men lack is not supported by the evidence and tends to discourage men from developing it.
Why it matters
What this means for relationships
Because empathy is partly skill, couples are not stuck with whatever level they start with. Practices like deliberately taking the other's perspective, reflecting back what you hear before responding, and asking rather than assuming tend to increase felt understanding, which research links closely to relationship satisfaction.
It also helps to recognize which component is missing in a given moment. A partner who understands but seems unmoved may need a cue to act, while one who is overwhelmed by your distress may need help thinking clearly. Naming the gap kindly, rather than concluding the other person simply does not care, tends to lead to better repair.
Where it varies
The nuance
These are averages with large overlap. Janet Hyde's gender similarities hypothesis (2005) shows that on most psychological measures, including many related to empathy, men and women are far more alike than different. Plenty of men are highly empathic and plenty of women less so; individual variation dwarfs the average gap.
Empathy is not uniformly good in unlimited amounts. Davis's distinction between empathic concern and personal distress matters here: feeling another's pain so intensely that it becomes your own distress can lead to withdrawal rather than help. The most useful empathy pairs understanding and warmth with enough regulation to stay present and act.
Questions people ask about this
What are the different types of empathy?
Research, particularly Davis's work, distinguishes cognitive empathy (understanding what another feels), affective empathy (sharing the feeling), and compassionate concern that motivates helping. There is also personal distress — one's own anxiety at another's suffering. These run on partly separate processes, so a person can be strong in one and weaker in another.
Are women more empathic than men?
On average, women score somewhat higher on some self-reported empathy measures and are slightly better at reading nonverbal cues, per Hall's meta-analysis. But the gaps are modest, largest on self-report rather than objective tests, and likely reflect socialization and motivation as much as ability. The overlap between genders is large.
Can empathy be learned or is it fixed?
Research suggests empathy is partly a skill that responds to practice and intention rather than a fixed trait. Perspective-taking and emotional attunement both improve with deliberate effort. Empathy also varies by context — it drops under stress and rises with safety — which further supports treating it as a developable capacity.
Why does someone understand my feelings but not seem to care?
This often reflects the gap between cognitive and affective empathy. A person can accurately grasp why you are upset yet feel little emotional pull to act, since these draw on different processes. It usually is not coldness; a clear cue about what you need can bridge the gap better than assuming indifference.
Can someone be too empathic?
In a sense, yes. Davis distinguishes empathic concern from personal distress — feeling another's pain so intensely it becomes your own overwhelm, which can lead to withdrawal rather than help. The most useful empathy pairs understanding and warmth with enough emotional regulation to stay present and actually support the other person.
How can I become more empathic?
Research points toward deliberately taking the other person's perspective, reflecting back what you hear before responding, and asking rather than assuming what they feel. Because empathy drops under stress, managing your own state helps too. These practices tend to increase felt understanding, which is closely tied to relationship satisfaction.
Research sources
These references point to the published research and established frameworks behind this page. They are provided for further reading; see our research methodology for how sources are selected.
- Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44(1), 113–126.
- Hall, J. A. (1978). Gender effects in decoding nonverbal cues. Psychological Bulletin, 85(4), 845–857.
- Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85(2), 348–362.
- Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60(6), 581–592.