Fundamental attribution error is stranger than it sounds.

Watch someone behave badly once and you will assume they are a bad person, barely registering the situation that cornered them.

The tendency has a name, the fundamental attribution error, and it may be the most socially expensive bias ever documented. When we explain other people’s behavior, we overweight personality and underweight circumstance. She snapped because she is rude, not because she just got terrible news. He failed because he is incompetent, not because the task was impossible. The situation, which is often doing most of the causal work, fades from the explanation like background scenery.

The term was coined by Lee Ross in 1977, but the classic demonstration came a decade earlier. In 1967, Edward Jones and Victor Harris had participants read essays about Fidel Castro, some pro, some anti. Some participants were told the writers had chosen their position freely; others were told the position was assigned by a coin flip. Even the participants who knew the position was assigned rated the pro-Castro writers as genuinely pro-Castro. They could not fully discount the behavior, even when the situation explained it completely. Knowing about the coin flip was not enough to correct for it.

Ross and colleagues sharpened the point with the quizmaster study. Participants were randomly assigned to be questioners, who invented difficult trivia questions, or contestants, who tried to answer them. Both the contestants and the observers rated the questioners as more knowledgeable, even though everyone knew the roles were random and the questioners’ advantage came entirely from getting to choose the questions. The situational advantage was visible, acknowledged, and still underweighted.

The mechanism is partly perceptual: people are vivid and salient, situations are background, and attention flows to the figure, not the ground. It is partly a matter of cognitive effort: dispositional explanations are simple and fast, situational ones require work. And it is partly cultural: individualistic societies, which tell stories about character, show the error more strongly than collectivist ones, which tell stories about context.

The error runs the world’s judgments. Hiring managers read resumes as character and miss the circumstances behind the gaps. Voters blame leaders for economies and praise them for luck. Internet mobs judge strangers from seconds of video, with no access to the hours around it. Every one of these judgments feels like perception. Most of them are the error.

The error shrinks, but does not vanish, across cultures. Joan Miller showed in 1984 that Indians, from a more collectivist culture, made more situational attributions than Americans, attending to context where Americans saw character. The fundamental attribution error is real everywhere it has been tested, but its size varies with the stories a culture tells. Societies that narrate life as character produce stronger dispositionalism. Societies that narrate life as context produce less. The bias is fundamental. Its volume is cultural.

Modern life mass-produces the error’s favorite conditions. Social media serves up strangers’ worst seconds with none of the surrounding hours, and the platform’s design, vivid person, invisible situation, is a fundamental-attribution-error machine. The quiz-show host looks brilliant next to flustered contestants, though the host holds the answer cards. The resume’s gap looks like character until you learn about the illness, the layoff, the pandemic. Every thin slice of a person invites the error. Thickness is the only cure.

Ross named it fundamental in 1977 because it underlies so much else. The correction is a habit, not an insight: when you catch yourself judging a person, ask what situation would make a reasonable person do the same thing. Usually, you can find one. That is the point.