Perspective-Taking Is a Skill, Not a Personality Trait

I have sat across the table from people who describe themselves as "just not a people person," as if that settled the matter. And I have sat across from people who call themselves natural empaths, as if that settled it too. Both descriptions do the same quiet damage. They treat perspective-taking, the ability to genuinely register how someone else sees the world, as a fixed trait you either got or didn't.

In my view, that framing is not just wrong. It is the main reason so many well-intentioned people never get better at it.

The Trait Myth and Why It Persists

We tend to sort social abilities into two bins: skills, which anyone can build with practice, and traits, which are just part of who you are. Public speaking, negotiation, and cooking go in the first bin. Warmth, humor, and "reading people" tend to land in the second, even though there's no good reason they should.

Part of the problem is that perspective-taking feels effortless when it works. You just seem to know what your old friend is feeling, or what your teenager is actually upset about, and it doesn't feel like a technique you applied. It feels like intuition. When something feels automatic, we assume it can't also be trained. But automatic and trained are not opposites. A jazz musician's improvisation feels spontaneous to the listener and is the product of ten thousand hours of scales. The same logic applies here.

The psychologist Jamil Zaki, who has spent much of his career studying empathy at Stanford, describes it using a gym metaphor: empathy behaves less like a fixed trait and more like a muscle that responds to how you use it. People who believe empathy can grow tend to work harder at understanding others in difficult moments, and that effort compounds. People who believe it's fixed tend to give up the moment it gets hard, which then confirms their belief that they were never good at it to begin with. The belief becomes the ceiling.

What Happens When We Stop Practicing

If perspective-taking is a trained capacity, it should be able to atrophy the same way any unused skill does. There's reason to think it has.

A 2010 meta-analysis led by Sara Konrath, reviewing decades of data on American college students, found that self-reported empathy measures dropped by roughly 40 percent between 1979 and 2009, with most of that decline concentrated after 2000. Researchers have offered a range of explanations, from more time spent in mediated communication to increased social comparison, but the finding itself is hard to argue with: whatever produces the capacity to take another person's perspective, it moved in the wrong direction over three decades, and it moved fastest exactly when more of our social contact started happening through screens instead of faces.

I don't think this means people got worse human beings. I think it means the specific muscle got less exercise. That's actually the more hopeful reading, because muscles that atrophy from disuse can be rebuilt from use.

Why "Just Imagine Their Perspective" Doesn't Work

Here is where the research gets genuinely surprising, and where I had to revise my own assumptions.

The intuitive advice, the thing you've heard in every conflict-resolution workshop, is to imagine yourself in the other person's shoes. Close your eyes, picture their situation, feel what they feel. It sounds right. It is also, according to the data, close to useless on its own.

In a 2018 study published in the Journal of Personality and Social Psychology, researchers Tal Eyal, Mary Steffel, and Nicholas Epley compared what they called perspective-taking (imagining another person's viewpoint) against what they called perspective-getting (actually asking that person and listening to the answer). Across a series of experiments, imagining someone else's perspective did not reliably improve accuracy about what that person was actually thinking or feeling. Asking them directly did, and by a wide margin.

That finding reframes the whole project. Perspective-taking, as most of us were taught it, is an act of imagination performed alone in our own heads. But the thing we actually want, accurate understanding of another person's inner experience, doesn't come from imagination. It comes from information. And information requires a question, asked out loud, followed by real listening to the answer.

This is worth sitting with, because it explains a specific kind of failure a lot of thoughtful people experience: they work hard at imagining the other side, feel genuinely compassionate while doing it, and are still wrong about what that person needed. The empathy was real. The accuracy wasn't there because imagination was never going to supply it.

There's a second finding that complicates the trait story even further. In a series of studies led by Adam Galinsky at Kellogg, people who were primed to feel powerful, even through something as small as recalling a memory of having authority over someone, became measurably less likely to adopt another person's visual or emotional perspective than people primed to feel powerless. Power didn't just fail to help perspective-taking. It actively suppressed it. If perspective-taking were a stable trait, a temporary psychological prime shouldn't move it at all. The fact that it does is strong evidence the capacity is situational and trainable, not fixed.

Trait View vs. Skill View

The two framings lead to genuinely different behavior when perspective-taking breaks down, which is exactly when it matters most.

I find the middle rows the most useful ones. If you know your own perspective-taking gets worse under stress or when you hold more power in a conversation, that's not a character flaw to feel bad about. It's a known variable you can build a workaround for, the same way a pilot builds a checklist around the known fact that humans get worse at vigilance after the first hour of a flight.

What Actual Training Looks Like

If imagining someone's perspective doesn't reliably work, and asking does, then a real training method has to be built around structured questioning and checking, not internal visualization. A few components show up again and again in the research on this.

Ask before you assume. This sounds almost too simple to write down, and yet it is the single largest lever in the Eyal, Steffel, and Epley findings. Before constructing a story in your head about why someone did what they did, ask them. Not a leading question that confirms what you already believe, but a genuine one: "What was going on for you there?"

Reflect back what you heard, and let them correct it. This is the core move of reflective listening, and it functions as a built-in accuracy check. You state your understanding of their position, out loud, and treat it as a draft rather than a conclusion. The correction they offer is usually where the real learning happens. I have had conversations where my reflection was maybe 70 percent right, and the 30 percent gap the other person filled in taught me more than the whole rest of the conversation.

Practice with people who see the world differently than you do. Empathic accuracy research using what's called the Ickes paradigm, where strangers watch recordings of themselves and try to identify each other's actual thoughts moment by moment, consistently finds accuracy in the range of only 20 to 30 percent above chance guessing, even between people who like each other. That number gets worse, not better, across lines of real difference: political, cultural, generational. Which tells you something useful. If you only ever practice perspective-taking with people who already think like you, you are training on the easy version of the skill and will be unprepared for the hard version, which is the version that actually matters in a divided room.

Treat being wrong as data, not as a verdict on your character. This is the piece the trait framing destroys. If perspective-taking is a fixed gift, being wrong about someone means you lack the gift. If it's a trained skill, being wrong means you have a specific, fixable data point about where your model of that person broke down. Same event. Completely different next move.

Repeat it enough times that it stops requiring willpower. Deliberate practice research across domains, from chess to surgery to music, consistently shows that skills become more automatic with volume of repetition, not with talent alone. There is no reason to think social perception is exempt from this. The people who seem like natural perspective-takers are very often just people with an enormous number of repetitions behind them, often from professions or life circumstances that forced the reps: teachers, nurses, hostage negotiators, parents of several very different children.

Where the Skill Breaks Down Even With Practice

I want to be honest about the limits here, because overselling this would be its own kind of dishonesty. Even trained, deliberate perspective-taking has a ceiling. The empathic accuracy numbers I cited above, in the 20 to 30 percent range above chance, come from studies of people actively trying. Nobody gets to perfect mind-reading. Culture, power, fatigue, and simple unfamiliarity all impose real limits on how well one person can model another's inner life, no matter how much they've practiced.

What training changes is not whether you'll ever be wrong. It's how often, how badly, and what you do next when you find out. A trained perspective-taker still misreads people. The difference is they built in a way to find out quickly, and a habit of updating instead of defending the original guess.

A Practice to Start This Week

Pick one relationship where you've noticed friction, a coworker, a family member, someone whose recent decision confused or frustrated you. Instead of building your usual internal theory about why they did what they did, go ask them one direct question about it. Then say back what you heard, in your own words, and ask if you got it right. Let them correct you if you didn't.

Do this five times over five different conversations this week. Notice how often your first internal guess, the one you would have quietly assumed and acted on, turns out to be at least partly wrong. For most people, that number is higher than they expect. That gap between assumed and actual is the whole skill, made visible.

Frequently Asked Questions

Is perspective-taking a personality trait or a skill?

The research points toward skill. Perspective-taking measurably declines with disuse, drops under conditions like power or stress, and improves with specific practices like direct questioning and reflective listening, all of which are hallmarks of a trainable ability rather than a fixed trait.

Can adults actually improve at reading other people?

Yes. Studies on empathy training interventions, along with deliberate-practice research more broadly, show meaningful improvement with structured effort. The gains tend to come from behavioral changes, like asking and checking, rather than from trying harder to imagine or empathize internally.

Why doesn't just imagining someone's perspective work?

Research by Eyal, Steffel, and Epley found that imagining another person's viewpoint did not reliably improve accuracy about what they were actually thinking, while directly asking them did. Imagination draws on your own assumptions; asking draws on their actual experience.

Does having power over someone affect how well you take their perspective?

It does, and not in a good way. Studies led by Adam Galinsky found that people primed to feel powerful became less likely to adopt another person's perspective, even in small, temporary ways. Anyone in a position of authority, at work or at home, is working against this current by default.

What's the difference between perspective-taking and perspective-getting?

Perspective-taking is imagining someone else's viewpoint on your own. Perspective-getting is asking them directly and listening to the answer. The research consistently favors perspective-getting for actual accuracy, which is why direct questions and reflective listening outperform quiet imagination.

Last updated: 2026-08-08