Someone cuts you off in traffic and you know, instantly, what kind of person they are. You cut someone off last Tuesday and you know, just as instantly, that you were late for something that mattered and the lane ended sooner than it looked. Hold those two certainties side by side for a moment. They cannot both be the whole truth. Yet most of us live inside exactly this asymmetry all day long: we explain ourselves by our circumstances and everyone else by their character. It is quiet, it is automatic, and it is the reason so many good relationships end up feeling like a long list of small disappointments that nobody can quite name.

The mistake has a name

In 1967, the psychologists Edward Jones and Victor Harris asked people to read essays either defending or attacking Fidel Castro, then to guess the writer's real opinion. When readers were told the writer had chosen the position freely, they assumed the essay reflected a genuine belief. That part is reasonable. The strange part came next. When readers were told the writer had been assigned the position by a coin flip, with no choice at all, they still rated the pro-Castro writers as more sympathetic to Castro. The situation fully explained the essay. People read the person anyway.

A decade later, Lee Ross gave this pattern its name: the fundamental attribution error. Daniel Gilbert and Patrick Malone later preferred the gentler term correspondence bias, but the idea is the same. When we watch other people act, we lean hard toward explanations about who they are, and we lean away from explanations about what they are facing. The quiet coworker is cold, not overwhelmed. The friend who cancelled is flaky, not drowning. The partner who snapped is unkind, not frightened about money.

Why your brain does it anyway

This is not a character flaw. It is a feature of how perception is built. Gilbert's work suggests that judging someone's disposition happens almost automatically, in the same breath as seeing the behavior. Correcting for their situation is a second, slower step that takes mental effort. When you are busy, tired, distracted, or stressed, that second step is the first thing to go. People under cognitive load in his studies made the dispositional judgment and simply never got around to the correction.

There is a perceptual reason too. When you watch someone else, they are the figure and their circumstances are the background. Your attention lands on the person. When you act, you are looking outward, so what fills your field of view is the situation: the deadline, the traffic, the thing your mother said on the phone. Jones and Richard Nisbett called this the actor-observer asymmetry. Later analysis by Bertram Malle found the asymmetry is less universal than textbooks once claimed, but it shows up most reliably in one place: negative events. We are especially generous with ourselves when things go badly, and especially stingy with everyone else.

A famous study from Princeton makes the point uncomfortably clear. John Darley and Daniel Batson asked seminary students to walk across campus to give a short talk, some of them on the parable of the Good Samaritan. Along the way, each passed a man slumped in a doorway, apparently in distress. The strongest predictor of who stopped to help was not personality, not faith, not even what they were about to preach. It was whether they had been told they were running late. Hurry, a situation, turned future ministers into people who stepped over a stranger. If you had watched from a window, you would have drawn conclusions about their hearts.

What assuming the worst costs you

The price is paid mostly in close relationships, because that is where the sample size is largest. Frank Fincham and Thomas Bradbury spent years studying how couples explain each other's behavior, and they found a signature of unhappy marriages. Distressed partners make what the researchers called distress-maintaining attributions. A negative act gets explained as internal, stable, and global: you did that because you are selfish, you are always like this, and it touches everything. A positive act gets explained away: you were only nice because your friends were watching. Satisfied couples tend to run the pattern in reverse. Crucially, these attribution habits predicted how satisfaction changed over time. The story you tell about why your partner did something is not a side effect of the relationship. It is part of the machinery.

There is a subtler cost, and it is about you. Every time you reach for a verdict instead of a question, you close a door that curiosity could have walked through. The fast explanation feels like knowing. It is actually the end of finding out.

Curiosity is the correction, not kindness

The usual advice here is to be more charitable, to imagine how the other person feels, to walk a mile in their shoes. It turns out that advice is weaker than it sounds. In a series of studies published in 2018, Tal Eyal, Mary Steffel, and Nicholas Epley tested whether perspective-taking, deliberately imagining another person's point of view, made people more accurate about what that person actually thought and felt. It did not. In several experiments it made people more confident without making them more correct. What did improve accuracy was something far less glamorous. The researchers called it perspective-getting: simply asking the other person and listening to the answer.

This is the deep idea worth carrying out of this article. The antidote to assuming the worst is not generating a nicer assumption. Your imagination is still your imagination, built from your own background and your own mood. The antidote is treating your first explanation as a hypothesis instead of a finding, and then going to get data. That shift, from verdict to question, is what curiosity is at the level of a single moment. It does not require you to think well of anyone. It only requires you to admit you do not yet know.

The practical question that does most of the work is this: What would have to be true for a reasonable person to do what they just did? Not a saint. Not a villain. A reasonable person, with a body that gets tired and a life you cannot see. Almost every behavior has a reasonable-person story, and the exercise of finding it does something the moral instruction to be charitable never quite manages. It makes you interested.

Your next moves

  • Catch one verdict today and write it down as a sentence. The form is always the same: They did X because they are Y. Then write three situational explanations underneath, however unlikely. You are not choosing the right one. You are proving to yourself that the first one was not the only one.
  • Run the swap test. When someone's behavior annoys you, ask what you would say if you had done the identical thing last week. If your version of the story involves traffic, sleep, or a bad morning, give the other person the same raw material.
  • Replace one conclusion with one question. Pick a person you have already quietly judged this week and ask them something real and open: "What was going on for you yesterday?" Then stop talking. Perspective-getting beats perspective-taking, but only if you actually get it.
  • Declare a no-verdicts window. Gilbert's load research says your correction step fails when you are depleted. Make a rule: no character judgments after a certain hour, or in the car, or before coffee. Write the behavior down and decide what it means tomorrow.
  • Keep a record of the times you were wrong. When a first impression turns out to have a situational explanation you never saw, note it. A list of your own misreads is the most persuasive argument your future self will ever get for staying curious a little longer.

A place to keep the question open

The hardest part of all this is not the insight. It is remembering, three hours later, that the verdict you reached on the stairs was supposed to be a question. That is what Curiokit is built for: a small, quiet place to capture the questions you have not answered yet, including the ones about the people in your life, and to come back to them when you are rested enough to actually wonder. If you want a home for your "what would have to be true" questions, you can start one at curiokit.lumenlabs.works. Either way, the next time someone disappoints you, try the question before the verdict. You might be surprised who they turn out to be.