Attribution is the everyday detective work of social life. Someone cuts you off in traffic, a friend snaps at you, a stranger is rude to a waiter — and almost instantly your mind hands you a cause. The catch is that the cause your mind hands you is predictably lopsided. This topic is really a continuation of a foundational big idea, Lewin's B = f(P, S): behavior is a function of the Person and the Situation. Attribution research is the story of how, when we explain other people, we reach for P and quietly forget S.
Bodhi says
The whole topic hangs on one distinction, so nail it first. An internal (dispositional) attribution says the cause is in the person — their personality, attitude, character. An external (situational) attribution says the cause is in the situation — the pressures, roles, and circumstances anyone would have felt. Heider (1958), the father of attribution theory, called this "naïve psychology" because we all do it without training. The rest of the topic is about the tilt in how we do it.
Two ways to explain a single act
Fritz Heider (1958) noticed that ordinary people are amateur scientists of behavior, and that every explanation we produce falls on one side of a fault line. Consider a classmate who bombs a presentation. You can explain it dispositionally — "he's just not a strong speaker" — or situationally — "the projector died and he'd been up all night with a sick kid." Same event, two radically different stories, two radically different judgments of the person. Which story you reach for isn't random, and it isn't fair.
Internal / dispositional
The cause lives in the person — traits, attitudes, ability, character. "She's ambitious." "He's careless." Assumes the behavior reveals who they are.
External / situational
The cause lives in the situation — pressures, roles, incentives, constraints. "Anyone would've done that under those conditions." Assumes most people would respond the same way.
Two mid-century theories tried to describe how we ought to move between these boxes. Jones and Davis's (1965) correspondent inference theory asked when we treat an action as truly "corresponding" to a disposition — we infer character most confidently when the behavior was freely chosen, produced unusual (non-common) effects, and ran against social expectations. A polite word from a diplomat tells you little; a rude one tells you a lot. Kelley's covariation model, next, is the more famous cousin.
Kelley's covariation model: the mind as scientist
Harold Kelley (1967) proposed that when we have information across time and people, we behave like little statisticians. We ask how the behavior covaries — does it appear and disappear along with some candidate cause? Kelley said we gather three specific kinds of information, and the pattern across them steers us toward the person or the situation.
1 · Consensus
Do other people act this way toward the same thing? Sarah loves the movie — is everyone else leaving the theater raving too? High consensus (everyone reacts this way) points to the situation/stimulus; low consensus (only she reacts this way) points to her.
2 · Distinctiveness
Does this person act this way toward other things too? Does Sarah rave about every movie, or is this rare for her? High distinctiveness (only toward this one) points to the stimulus; low distinctiveness (she raves about everything) points to her.
3 · Consistency
Does the person act this way toward this thing across time and settings? Will Sarah still love it next week, at home, in six months? High consistency makes any attribution firmer; low consistency ("she was just in a weird mood") blocks a clean attribution altogether.
Put them together and two clean recipes fall out. Low consensus + low distinctiveness + high consistency → an internal attribution ("that's just Sarah — she loves everything"). High consensus + high distinctiveness + high consistency → an external attribution ("the movie is genuinely great — everyone loves it, she rarely raves, and it holds up"). It's an elegant model. The problem, as Kelley half-admitted, is that it describes the rational observer we rarely are. We seldom have all three streams of information, and even when we do, we usually skip the audit and go with the gut. And that shortcut is where the errors live.
The fundamental attribution error: the load-bearing idea
Here is the finding the whole topic is built to deliver. The fundamental attribution error (FAE), also called the correspondence bias, is our pervasive habit of overestimating dispositional causes and underestimating situational ones when we explain other people's behavior. It's "fundamental" because it's not one bias among many — it's the default setting that most of the other biases here are variations on.
The classic demonstration is Jones and Harris (1967). Participants read an essay either supporting or attacking Fidel Castro, and — this is the key — were told the writer had been assigned the position, no choice involved. Rationally, an assigned essay reveals nothing about the writer's real attitude. Yet participants still inferred that pro-Castro essayists were personally more pro-Castro. They knew about the situational constraint and dispositionalized anyway. Ross, Amabile, and Steinmetz (1977) sharpened it into the famous "quiz game" study: participants watched a randomly assigned "questioner" invent hard trivia questions and a "contestant" struggle to answer. Everyone knew the roles were assigned by a coin flip, yet observers rated the questioner as far more knowledgeable — fooled by a situation that had simply handed one person all the advantages.
The single most important idea in this topic
The FAE is what happens when we forget the S in Lewin's B = f(P, S). When you explain someone else, your gut computes behavior as if it were a function of the person alone — B = f(P) — and drops the situation out of the equation. That's it. That's the error. The real test of the idea is whether you'll put the S back in when your intuition wants to blame the person.
Why does the FAE happen? Two reasons, stacked
It's easy to learn that the FAE happens and skip the mechanism — but the mechanism is the interesting part. There are two reasons, and they stack on top of each other.
The first is perceptual salience. Borrowing from Gestalt psychology, think of any social scene as a figure against a ground. The person moves, talks, has a face — the person is vivid figure. The situation is static, invisible, taken for granted — it's the ground that recedes. Whatever captures our attention feels causally important, so the person, being figure, hogs the causal credit while the situation, being ground, gets none. Storms (1973) proved this was really about the camera angle of perception: when he literally reversed people's visual perspective with video — showing actors a tape of themselves — their attributions flipped to match whatever was now salient. Salience is attribution.
The second reason is effort, and it's Daniel Gilbert's two-step process of attribution (Gilbert, 1989; Gilbert & Malone, 1995). Step one is automatic and nearly effortless: we characterize — we make a snap internal attribution ("he's rude"). Step two is effortful and optional: we correct — we adjust for the situation ("...though the flight was cancelled and he'd been on hold for an hour"). Step one runs on autopilot; step two runs only if we have the time, energy, and motivation to spend. Busy, tired, distracted, or cognitively loaded, we simply never take step two — and we're left stuck with the dispositional snap judgment. The FAE isn't stupidity; it's an unfinished computation.
Bodhi says
Notice how neatly the two explanations fit together. Salience says the person looks like the cause; Gilbert says correcting that first impression takes work you often won't do. So the initial reading is biased toward the person, and the fix is the step most likely to get skipped. No wonder the error is so sticky. The practical upshot: whenever you're rushed or drained, assume your snap judgments of others are running on step one only.
The culture caveat: how "fundamental" is it, really?
The FAE was named in individualist Western labs, and the name oversells it. When you take the same paradigms to collectivistic cultures, the error shrinks — and sometimes vanishes.
A subtlety worth knowing
Miyamoto and Kitayama (2002) ran a Jones-and-Harris-style essay study across cultures and found Americans committed the correspondence bias robustly while Japanese participants largely did not — East Asian perceivers gave far more weight to the situational constraint. Morris and Peng (1994), in the celebrated "fish study," showed animated fish moving in a group and asked why one fish swam ahead of the others. American participants tended toward internal explanations (that fish is a leader, an outlier); Chinese participants leaned situational (the group is pushing it, the water is pulling it). The lesson isn't that one culture is right — it's that the FAE is a Western default, not a hardwired human universal. It's "fundamental" only if your sample is WEIRD (Western, Educated, Industrialized, Rich, Democratic). Whenever you meet a "basic" bias, ask: basic for whom?
The actor–observer difference (with an asterisk)
Here's a twist: the FAE is about how we explain others. When it's our own behavior, the tilt reverses. The actor–observer difference is the tendency to attribute our own actions to the situation while attributing the same actions in others to disposition. You were late because traffic was brutal (situational); your coworker was late because he's disorganized (dispositional). Jones and Nisbett (1971) explained it partly through salience again — as the actor, you can't see your own face; what's vivid to you is the situation you're staring at, so your figure is the world, not yourself.
Where the science stands now
Textbooks often state the actor–observer difference like a law, but Malle's (2006) meta-analysis of over 170 studies found the effect is small on average and heavily moderated — it's much stronger for negative events (we excuse our own failures more than our successes) and for people we're close to, and it can even reverse. So don't memorize it as an iron rule; hold it as a tendency that depends on whether the behavior was good or bad and on who's involved. That nuance is exactly what separates a careful understanding from a rote one.
When attribution serves the self
Some attributional tilts aren't about salience or effort at all — they're about protecting the self. Recall the two master motives of social cognition: the need for accuracy and the need for self-esteem. Self-serving attributions are self-esteem winning.
The self-serving bias is the tendency to credit our successes to internal factors ("I aced the stats exam because I'm sharp") and blame our failures on external ones ("that test was unfair"). It runs on two engines: self-esteem maintenance — we distort reality to feel good about ourselves — and impression management — we present a flattering story to others. Tellingly, the bias is stronger in individualistic cultures and weaker (sometimes reversed into self-effacement) in collectivistic ones, and it's especially pronounced in solo sports like tennis, where the whole outcome sits on one person. This is the same self-esteem motive as before, just wearing an attributional costume.
A darker relative is defensive attribution, an explanation that fends off feelings of vulnerability and mortality. Its signature form is the belief in a just world (Lerner, 1980) — the assumption that good things happen to good people and bad things to bad people. It sounds comforting until you see its cost: victim blaming. If the world is fundamentally fair, then someone who suffered a mugging or an assault must have done something to invite it — a conclusion that lets the observer feel safe ("that won't happen to me, because I'm not like them") at the victim's expense. The just-world belief is the FAE aimed at suffering: it over-blames the person to protect the observer's sense of a controllable world.
The whole rogues' gallery, in one glance
It's easy to blur these together, so hold the distinctions crisp. The FAE over-blames other people's dispositions. The actor–observer difference flips that for the self — situation for me, disposition for you (small and moderated, per Malle). The self-serving bias splits the difference strategically — success is mine, failure is the world's. And belief in a just world reassures us that outcomes are deserved, which slides into blaming victims. Underneath all four sits the same pair of forces we keep meeting: the shortcuts of a cognitive miser and the self-protective pull of the self-esteem motive.
The one thing to carry out of this topic
When you explain a stranger, your mind runs a rigged experiment: the person is spotlit, the situation is in shadow, and the correction that would even things out is the step you're most likely to skip. That's the fundamental attribution error, and it is the beating heart of the topic. The trained move — the one worth carrying, and the one that makes you kinder in real life — is to put the S back into B = f(P, S). Before you decide someone is rude, lazy, or brilliant, ask what situation they were standing in. Half the time, that question changes the answer.
References
Gilbert, D. T. (1989). Thinking lightly about others: Automatic components of the social inference process. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 189–211). Guilford Press.
Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psychological Bulletin, 117(1), 21–38. https://doi.org/10.1037/0033-2909.117.1.21
Heider, F. (1958). The psychology of interpersonal relations. Wiley.
Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 219–266). Academic Press.
Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3(1), 1–24. https://doi.org/10.1016/0022-1031(67)90034-0
Jones, E. E., & Nisbett, R. E. (1971). The actor and the observer: Divergent perceptions of the causes of behavior. General Learning Press.
Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation (Vol. 15, pp. 192–238). University of Nebraska Press.
Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. Plenum Press.
Malle, B. F. (2006). The actor–observer asymmetry in attribution: A (surprising) meta-analysis. Psychological Bulletin, 132(6), 895–919. https://doi.org/10.1037/0033-2909.132.6.895
Miyamoto, Y., & Kitayama, S. (2002). Cultural variation in correspondence bias: The critical role of attitude diagnosticity of socially constrained behavior. Journal of Personality and Social Psychology, 83(5), 1239–1248. https://doi.org/10.1037/0022-3514.83.5.1239
Morris, M. W., & Peng, K. (1994). Culture and cause: American and Chinese attributions for social and physical events. Journal of Personality and Social Psychology, 67(6), 949–971. https://doi.org/10.1037/0022-3514.67.6.949
Ross, L. D., Amabile, T. M., & Steinmetz, J. L. (1977). Social roles, social control, and biases in social-perception processes. Journal of Personality and Social Psychology, 35(7), 485–494. https://doi.org/10.1037/0022-3514.35.7.485
Storms, M. D. (1973). Videotape and the attribution process: Reversing actors' and observers' points of view. Journal of Personality and Social Psychology, 27(2), 165–175. https://doi.org/10.1037/h0034782