“Try to see it from their side.”
You’ve heard it a hundred times. Probably said it, too. And it sounds so simple that nobody bothers to explain how you’re supposed to do it.
The awkward part is that most of us believe we’re already decent at reading other people. We nod, we imagine, we fill in the blanks, and then we act on a version of the other person that lives mostly in our own head. Sometimes that version is close. Often it isn’t, and that’s where arguments, cold silences and “that’s not what I meant” come from.
Understanding someone else’s point of view is a skill. Like any skill, it has a few moves that work better than others, and some of the popular ones work a lot worse than you’d expect. This guide walks you through nine of them, each backed by research or a documented real-world example, plus what to do when you catch yourself slipping back into your own head.
If you’ve read about NLP’s perceptual positions (the “me”, “them” and “observer” points of view), you’ll recognise a few ideas here. The difference is that we’re going to check which parts hold up when someone actually measures them.
1. Ask instead of guessing
This one surprised the researchers who tested it.
In 2018, Tal Eyal, Mary Steffel and Nicholas Epley published 25 experiments in the Journal of Personality and Social Psychology asking a blunt question: does telling people to “imagine the other person’s perspective” make them more accurate about what that person thinks and feels? They tested it on reading emotions from faces, spotting fake smiles, detecting lies and predicting a spouse’s preferences. Most people expected perspective taking to help. It didn’t. It often made people less accurate while leaving them just as sure of themselves, or surer. In one task described by Nautilus, couples who had been together around a decade expected to get about 13 of 20 questions about their partner right. They got five.
What did work, in the final experiment, was simply asking the other person and listening to the answer. The authors call this “perspective getting”, and the lesson is uncomfortable but freeing. Your imagination can only rearrange what you already know. New information has to come from them.
And asking doesn’t cost you socially. Karen Huang and colleagues at Harvard (Journal of Personality and Social Psychology, 2017) found that people who asked more questions in conversation, especially follow-up questions, were liked more by their partners. In speed-dating data, the people who asked more follow-ups were more likely to get a second date. Participants didn’t predict that effect, which is probably why most of us under-ask.
Start small: the next time you catch yourself thinking “she’s probably upset because…”, stop at the “because” and turn it into a question. “You seemed quiet after the meeting. What was going on for you?” Then, the hard part, wait. If you want more ways to keep a conversation open like that, our guide on how to have a good conversation with anyone goes deeper.
2. Accept that you’re harder to read than you think
In 1990, a Stanford psychology graduate student named Elizabeth Newton ran a deceptively silly experiment. One person tapped out the rhythm of a well-known song, like “Happy Birthday”, on a table. Another person listened and tried to name it.
Tappers expected listeners to get it right at least half the time. Out of 120 songs, listeners named three.
That’s 2.5%.
The tappers weren’t stupid. They could hear the melody in their heads, so the taps sounded obvious to them. The listener only heard knocking. Chip and Dan Heath later popularised the study in their book Made to Stick as an example of the “curse of knowledge”: once you know something, it’s very hard to imagine not knowing it.
The same gap shows up in everyday messages. Justin Kruger, Nicholas Epley, Jason Parker and Zhi-Wen Ng (Journal of Personality and Social Psychology, 2005) had people send serious and sarcastic statements by email or voice recording, then predict whether the receiver would catch the tone. A summary of the first experiment, with 30 pairs of Cornell students, reports these figures:
| Situation | What senders expected | What actually happened |
|---|---|---|
| Tapping a song (Newton, 1990) | At least 50% recognised | 2.5% recognised |
| Tone sent by email (Kruger et al., 2005) | About 78% understood | 56% understood |
| Tone sent by voice (Kruger et al., 2005) | About 78% understood | 73% understood |
Email, it turns out, was barely better than a coin flip. And knowing each other well didn’t fix it.
Why does this matter for understanding other people? Because half of every misunderstanding is you assuming they got what you meant. When someone reacts “strangely”, check first whether they received the message you think you sent. Next time, do this: before you hit send on anything with an edge to it, read it out loud in a flat voice. Kruger’s team found that senders who did exactly that got better at predicting whether the sarcasm would land.
3. Catch your first guess before it sets
Why do we get other people wrong so reliably? The mechanism is worth understanding, because once you see it you start noticing it everywhere.
Nicholas Epley, Boaz Keysar, Leaf Van Boven and Thomas Gilovich proposed in 2004 (Journal of Personality and Social Psychology) that perspective taking works like anchoring and adjustment. Step one: you automatically start from your own view. Step two: you nudge away from it, toward what you think the other person sees. Step three: you stop as soon as you reach an answer that feels plausible. The problem lives in step three. Because you stop at “plausible” rather than “correct”, you usually end up closer to yourself than to them.
Their experiments backed up each link in that chain. Judging that someone saw things differently took people longer than judging that they saw things the same way, which fits the idea that difference requires extra mental work. When people were rushed, their judgments got more egocentric. When they were paid for accuracy, less so. And people who tended to accept the first plausible answer that came to mind stayed closer to their own view.
So the cause and effect runs roughly like this: speed and stress shorten the adjustment, short adjustment leaves you anchored on yourself, and being anchored on yourself makes the other person look unreasonable. Note that this is a lab finding about how judgments are formed, mostly with strangers and simple tasks. It doesn’t prove every argument you’ve had was caused by anchoring. It does tell you where the leak is.
The practical fix is almost boringly simple. When the stakes are high and you feel the urge to reply instantly, slow the adjustment down on purpose. Take one breath and ask yourself, “What would I believe if I knew only what they know?” If your body is too wound up to do that, a couple of quick resets for your nervous system can buy you the few seconds you need.
4. Look for the need under the position
People rarely argue about what they actually need. They argue about the solution they’ve already picked.
One of the clearest documented examples comes from the 1978 Camp David talks between Egypt and Israel, mediated by US President Jimmy Carter. Roger Fisher and William Ury used it in their negotiation classic Getting to Yes to show the difference between positions and interests. On paper, the positions were incompatible. Israel wanted to keep part of the Sinai Peninsula, which it had occupied since 1967. Egypt wanted all of it back. Any line drawn on a map would leave someone unhappy.
Underneath, though, the interests didn’t clash. Israel’s concern was security: it didn’t want Egyptian tanks massed on its border. Egypt’s concern was sovereignty: the land had been Egyptian, and getting it back mattered for national dignity. The agreement that grew out of Camp David returned the Sinai to Egypt while keeping large parts of it free of military forces. Both needs met. Neither position “won”.
Lab research points the same way. Adam Galinsky, William Maddux, Debra Gilin and Judith White (Psychological Science, 2008) ran negotiation studies in which a simple price deal was impossible but a creative one was available. People told to imagine what their opponent was thinking reached more deals than a control group. People told to imagine what their opponent was feeling made their counterparts happier but reached fewer deals. Thinking about someone’s goals and feeling for them are related, but they aren’t the same tool.
Your move: in your next disagreement, write two short lines. “They’re asking for…” and “They might need that because…”. If you can’t fill in the second line, you haven’t understood them yet, and the conversation in section 1 is where you go next.
5. Listen more than you watch
We tend to treat the face as the window to someone’s inner life. So it’s strange to learn that the face might be getting in the way.
Michael Kraus, then at Yale, ran five experiments with more than 1,800 adults in the US, published in American Psychologist in 2017. People tried to identify others’ emotions under different conditions: voice only, video only, or voice and video together. The APA’s summary of Kraus’s voice-only study reports that participants who only listened were, on average, more accurate than those who only watched or who did both. The one exception was a computer voice reading a transcript, which produced the worst accuracy of all. So it isn’t the words alone. It’s the human voice carrying them.
Kraus offered two possible explanations. People get a lot of practice masking emotion with their faces, so faces can mislead. And processing two rich streams of information at once may hurt performance on both, which means more input isn’t automatically better. Keep in mind these were controlled experiments with mostly short interactions, not long relationships, so treat the effect as a nudge rather than a law.
Even so, the practical side is easy to use. When a conversation really matters, give the voice your full attention. Notice pace, pauses, the little rise when someone is unsure, the flatness when they’re tired of explaining. Test it this week: next time a friend sounds “fine” in a text, call instead. Or, in a face-to-face chat, try looking slightly away for a moment and just listening. It also does wonders for reading people with real emotional intelligence, because tone often tells you what words are covering for.
6. Trade experiences, not only arguments
What happens when you stop debating someone and start comparing life stories? One real organisation built an entire method around that question, and it was tested carefully.
The Los Angeles LGBT Center developed a canvassing approach that later became known as deep canvassing. In 2015, from January to June, canvassers from the Center and SAVE, a South Florida LGBT organisation, went door to door in Miami-Dade County after the county had passed protections for transgender people. Instead of reciting a script, canvassers held conversations of about 10 to 15 minutes. They listened, and they asked voters to recall a time they themselves had been judged or treated differently, then to connect that memory to what transgender people experience. Psychologists call this analogic perspective taking.
David Broockman (then at Stanford) and Joshua Kalla (then at UC Berkeley) measured the results in a randomised trial published in Science in 2016. According to the Berkeley and Stanford summaries, about one in ten voters who had the conversation shifted their attitudes, a change the authors compared with the national drop in prejudice against gay and lesbian people between 1998 and 2012. The effect lasted at least three months and held up when voters were later exposed to counterarguments.
It’s one study on one issue, and the canvassers were trained and experienced, so don’t expect a ten-minute chat to transform every relationship. But the structure is useful anywhere. Arguments make people defend their position. A shared experience lets them walk into yours on their own legs.
A quick experiment: the next time a friend can’t understand why something upsets you, skip the case you’d make in court. Ask instead, “Have you ever felt like people decided who you were before you said anything?” Then listen to their story first. If you’re working on being warmer and kinder with people in general, this is one of the gentlest places to begin.
7. Give the other person a way to say yes
Sometimes understanding someone’s perspective means noticing what they can’t afford to say out loud.
On Friday, 26 October 1962, at the height of the Cuban Missile Crisis, Soviet leader Nikita Khrushchev sent President John F. Kennedy a message hinting he might remove the missiles from Cuba in exchange for a promise that the US would not invade the island. The next day, Saturday 27 October, a second, tougher message arrived demanding that the US also remove its missiles from Turkey. Kennedy’s advisers in the Executive Committee were split on how to respond.
Llewellyn “Tommy” Thompson, a former US ambassador to Moscow who knew Khrushchev personally, suggested replying to the first letter and setting the second aside. As Robert McNamara recounted in Arms Control Today in 2002, Thompson argued that Khrushchev needed to be able to say “I saved Cuba; I stopped an invasion”. In other words, Thompson wasn’t only asking what Khrushchev wanted. He was asking what Khrushchev needed to be able to tell his own people. Kennedy agreed, and McNamara later described the exchange as one of the most important discussions of the Cold War. Khrushchev announced the withdrawal of the missiles the following day.
Your stakes are lower, thankfully. The pattern isn’t. Picture this (hypothetically): your colleague refuses to drop a plan you know won’t work. You could prove it’s flawed in front of the team. Or you could find the part of it that’s genuinely good, and offer a version where they can say “I improved my idea” rather than “I lost”. People hold on to positions longer when letting go feels like losing face. Being a good boss or a good partner often comes down to this exact move.
Here’s a simple way in: before your next tense conversation, ask yourself, “If they agreed with me, what would they have to explain to someone else afterwards?” Then make that explanation easier for them.
8. Step into the observer’s seat
The NLP idea of a “third position”, watching a situation as a neutral outsider, sounds a bit like a workshop gimmick. It has some unexpectedly solid research behind it.
Eli Finkel and colleagues at Northwestern University followed 120 married couples for two years. Every four months, spouses reported on their marriage. In the first year, marital quality declined for everyone, which is depressingly normal. In the second year, half the couples did a short writing task three times: seven minutes each, 21 minutes in total. They had to think about their most recent disagreement from the point of view of a “neutral third party who wants the best for all involved”, and to consider how that person might see the good in it.
According to Finkel’s 2013 conflict-reappraisal study in Psychological Science, the decline continued for the control couples but disappeared for the couples who did the exercise. Interestingly, the intervention couples didn’t fight less often. They were just less distressed by their fights. The sample was married, heterosexual American couples, so it’s fair to wonder how far the result travels, but for 21 minutes of effort it’s a remarkable return.
Think of it like watching a football match from the stands instead of the pitch. On the pitch, every tackle feels personal. From the stands, you can see the shape of the whole game. The comparison breaks down in one place: in your life, you’re also one of the players, and you’ll have to walk back onto the pitch. The observer view doesn’t replace your own feelings. It gives you a break from them long enough to see the pattern.
Give it a go: set up three chairs, as NLP trainers often suggest. Sit in the first and describe the conflict as you see it. Move to the second and describe it as the other person would, in their words. Then sit in the third and describe both of you as a fair, kind stranger would. Write a sentence or two in each seat. Situations involving anger on one side or both benefit most from that third chair.
9. Say their side back until they agree you’ve got it
The late Stanford psychologist Lee Ross spent decades studying why people who disagree so often think the other side is biased. He called the underlying tendency naive realism. As described by Erika Weisz in Nautilus (2022), Ross argued that people “believe that their take on the world is the objective one”, a line from his 2016 TED talk. If that’s true, then anyone who sees it differently must be uninformed, irrational or biased. That’s the trap: once you assume you see reality as it is, understanding the other person starts to feel unnecessary.
Ross didn’t just describe the problem. In work published in 2018, he described a dialogue exercise used with people from both sides of the Northern Ireland conflict, in which each side had to restate the other side’s position accurately before arguing its own. He noted that it was difficult for everyone at first. That difficulty is the point. You find out quickly how little of their view you can actually reproduce.
This is also where perspective getting (section 1) and the observer’s view (section 8) meet. You ask, you listen, and then you prove you heard by handing their view back to them in a form they recognise. Not a strawman. Not “so what you’re saying is I’m terrible”. A version they’d sign.
It feels slow. It’s usually faster than a second round of the same fight. When you’re ready to try it, use one sentence: “Let me check I’ve got this right: you’re saying ____, and it matters to you because ____. Did I miss anything?” Keep correcting until they say yes, and only then share your side. Few habits do more for your everyday social skills.
What to do when you slip
You will slip. You’ll assume, you’ll snap, you’ll argue with a version of someone that turns out to be mostly you. Everyone does, including the researchers who study this for a living.
The question is what you do next, and here the research is kinder than your inner critic. Juliana Breines and Serena Chen at UC Berkeley ran four experiments, published in Personality and Social Psychology Bulletin in 2012, on how people respond to their own mistakes. In one, participants recalled something they’d done that they felt bad about. Those who then spent three minutes writing about it from a compassionate, understanding point of view were more motivated to make amends and to avoid repeating the mistake than people in the comparison groups. Beating yourself up felt productive. Treating yourself with understanding actually was. These were short lab studies, so the size of the effect in daily life isn’t clear, but the direction is.
There’s a nice symmetry there. The understanding you’re learning to offer other people works on you, too. If self-criticism tends to run the show, our guide on how to stop hating yourself is a good companion to this one.
It also helps to know that one bad moment doesn’t erase a habit you’re building. Phillippa Lally and colleagues at University College London followed 96 people forming new daily habits and published the results in the European Journal of Social Psychology in 2009. It took 66 days on average for a behaviour to feel automatic, with a huge range from 18 to 254 days. Missing a single opportunity didn’t meaningfully affect the outcome. Only being very inconsistent did.
So when you notice you’ve steamrolled someone’s point of view, do three things. Name it to yourself without drama (“I assumed again”). Go back to the person, even hours later, and ask the question you skipped. If you hurt them, apologise for the specific thing, and then keep the promise you make about doing better. Then carry on tomorrow. Listening well is one of the good habits worth having, and habits are built on the days after the slip as much as the good days.
The short version
Your imagination is a weak tool for knowing another person’s mind, and your confidence in it is usually higher than it deserves. The fixes are mostly humble: ask, listen to the voice, look for the need under the position, give people a way to say yes without losing face, step back to the observer’s chair, and say their view back until they recognise it.
None of this requires agreeing with everyone. It just means disagreeing with the real person in front of you rather than the one you invented.
Pick one technique from this list and use it in your next difficult conversation. Just one. And if you want more ideas for building closer, calmer connections with the people in your life, take a look through our Relationships section.
