Fundamental attribution error
Also known as: Correspondence bias
Blaming people's character while excusing the situation.
What it means
The fundamental attribution error, also called correspondence bias, is the tendency to over-attribute other people's behavior to stable internal dispositions — personality, character, intent — while underweighting the situational forces acting on them. Classic demonstrations show people inferring genuine attitudes from essays they know were assigned, as if the writer freely chose the position. It is often paired with the claim that we explain others by their traits but our own conduct by circumstances, yet this actor-observer asymmetry is weaker than once assumed and, by meta-analysis, close to zero overall, so the two effects are best kept separate. The bias also softens, and sometimes reverses, when situational information is made salient and in more collectivist cultural contexts, making it less a universal law than a default that strong context can override. It matters wherever we judge strangers, employees, defendants, or out-groups, because mistaking a situation for a character flaw fuels harsh judgment, poor management, and intergroup blame.
The original demonstration
The effect was first isolated by Edward Jones and Victor Harris in 1967. They asked participants to read a short essay either favoring or opposing a controversial position, then to estimate the true private attitude of the person who wrote it. In one condition participants were told the writer had chosen the side freely; in another they were told the position had been assigned by an experimenter, leaving the writer no choice. The rational inference is that an assigned essay reveals nothing about the author's real opinion. Participants did not behave that way. Even when they knew the stance had been dictated, they still inferred that a pro-position essay came from someone who genuinely held that view, and an anti-position essay from an opponent. The situational constraint was acknowledged and then largely discounted. Jones and Harris had expected the constraint to wipe out the inference and were surprised when it did not. The paradigm proved robust and easy to run, which is part of why it became a fixture of social psychology teaching. Lee Ross gave the phenomenon its enduring name a decade later, in a 1977 chapter on the everyday reasoner as a flawed intuitive psychologist, framing the over-reliance on dispositions as a systematic distortion in how people explain what others do.
Why it happens
No single mechanism explains the bias, and the leading accounts are complementary rather than competing. One is perceptual: in most encounters the person is vivid and moving while the situation is a static, easily ignored backdrop, so the actor captures attention and gets credited with causing the behavior. A second is a two-stage process of inference. Daniel Gilbert and Patrick Malone argued in their 1995 review that people first draw a quick dispositional inference almost automatically, then only sometimes correct it for situational pressure — and that correction is effortful, so it is skipped when the observer is distracted, hurried, or cognitively taxed. A third factor is that observers often lack or fail to imagine the situational information the actor takes for granted; unaware of the pressures in play, they read behavior as free expression of character. Gilbert and Malone also stressed that people hold unrealistic expectations about how others should behave and misjudge how strongly a given situation constrains action. Because several routes converge on the same output, the tendency is easy to produce but hard to attribute to one cause, and interventions that target only attention, or only effort, tend to reduce it partially rather than remove it.
What the evidence shows
The core correspondence-bias finding is one of the more durable results in social psychology: the tendency to read constrained behavior as revealing an attitude has replicated across many variations of the essay paradigm and related designs. What has not held up as well is a claim often folded into the same idea — that people explain others by disposition but themselves by situation. Bertram Malle's 2006 meta-analysis of the actor-observer asymmetry, drawing on more than a hundred studies, found the average effect close to zero and dependent on narrow conditions, a result the paper itself called surprising. This matters for how the bias is described: correspondence bias in judging others is well supported, but the neat mirror-image contrast between self and other is not. The two phenomena share intuition but not evidentiary standing, and careful writing keeps them apart. The mechanism story is also less settled than textbook summaries imply. The correction stage in Gilbert and Malone's model is inferred from moderators such as cognitive load rather than observed directly, and the size of the bias varies widely with how the situation is presented, how much choice the actor appears to have, and how the attitude question is framed.
Where it breaks down
The bias is a default, not a fixed law, and several conditions weaken or reverse it. The most studied is culture. Michael Morris and Kaiping Peng, in 1994, showed that when explaining the same social events, people from more individualist backgrounds leaned on the actor's dispositions while people from more collectivist backgrounds gave more weight to context and relationships. A 1999 review by Incheol Choi, Richard Nisbett and Ara Norenzayan synthesized this literature and reached a more careful conclusion than the headline suggests: the automatic first inference to disposition appears fairly general across cultures, but the willingness and ability to adjust for situation is what varies, being stronger in East Asian samples. So the difference is often one of correction, not of the initial reflex. The bias also shrinks within any group when situational information is made salient, concrete and hard to overlook, when observers are given time and motivation to think rather than judging under load, and when they are prompted to consider what pressures the actor faced. None of these fully eliminates the tendency, but each shows it is a starting point that stronger evidence can override rather than an unalterable feature of human judgment.
Where it shows up
The error does its damage wherever people judge others they do not know well and cannot see the constraints on. A manager who reads a missed deadline as laziness rather than an impossible workload will coach and staff the wrong problem. A hiring panel that treats one nervous interview as evidence of a nervous character discounts the pressure of the interview itself. In the justice system, jurors and officials can hear a coerced or scripted statement as sincere confession, the same discounting failure the essay studies produced. The bias is especially corrosive between groups: an out-group member's single bad act is read as proof of a group trait, while the situations that shaped it are ignored, feeding stereotypes and blame. Poverty, unemployment and ill health are routinely explained by the character of those affected rather than by circumstance, with direct consequences for policy attitudes. Knowing about the tendency helps only modestly, because it operates fast and feels like simply seeing what someone is. The more reliable correction is structural: build in the missing situational information, slow the judgment down, and ask explicitly what a reasonable person would have done under the same constraints before concluding anything about character.
Examples
A driver who cuts you off is 'a jerk'; when you do it, you were 'in a rush.'
A customer-service agent reads a script verbatim under company rules; a caller comes away convinced the agent personally does not care, attributing the flat tone to the person rather than the mandated script.
A commuter is cut off in traffic and mutters that the other driver is reckless and selfish; when the commuter does the same thing an hour later, it is because they were late for a meeting.
A hiring committee rates a candidate who stumbled over one question as lacking confidence, discounting that the candidate had just been given the hardest question with no preparation time.
Voters explain a neighborhood's high unemployment by the supposed work ethic of the people who live there, overlooking plant closures and the collapse of local transit.
First described in Term: Lee Ross (1977); demonstration: Jones & Harris (1967).
Key references
- Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3(1), 1-24. doi.org/10.1016/0022-1031(67)90034-0
- Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 10, pp. 173-220). Academic Press. doi.org/10.1016/S0065-2601(08)60357-3
- Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psychological Bulletin, 117(1), 21-38. doi.org/10.1037/0033-2909.117.1.21
- 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. doi.org/10.1037/0022-3514.67.6.949
- Choi, I., Nisbett, R. E., & Norenzayan, A. (1999). Causal attribution across cultures: Variation and universality. Psychological Bulletin, 125(1), 47-63. doi.org/10.1037/0033-2909.125.1.47
- Malle, B. F. (2006). The actor-observer asymmetry in attribution: A (surprising) meta-analysis. Psychological Bulletin, 132(6), 895-919. doi.org/10.1037/0033-2909.132.6.895