HPC › Decisions › Glossary
Fundamental attribution error is the systematic tendency to over-attribute other people's behaviour to stable character traits while discounting situational forces that constrain that behaviour. When someone makes an error at work, the observer's first inference is incompetence rather than inadequate resources or an unrealistic deadline. Coined by Lee Ross in 1977, the bias operates below conscious awareness.
The term is often used interchangeably with correspondence bias, though some researchers reserve correspondence bias for observer judgements specifically.
How it works
Jones and Harris (1967) demonstrated in three experiments that observers attributed genuine pro-Castro attitudes to a speaker even when told the speech position had been assigned by coin toss, not chosen freely.1 The behaviour was treated as evidence of stable character regardless of the situational constraint. Gilbert and Malone (1995) identified four distinct mechanisms behind this correspondence bias: unawareness that a situation is constraining behaviour; unrealistic expectations about conduct under pressure; inflated categorisation of the act's extremity; and incomplete correction when an initial dispositional snap judgement is interrupted before situational adjustment occurs.2
The two-step model clarifies where the bias enters. Characterisation (extracting a trait from a behaviour) is automatic, capacity-free, and occurs regardless of context. Correction (adjusting that trait inference to account for situational forces) is effortful and resource-dependent.2 Because correction is the step that fails, the error is especially pronounced in high-cognitive-load environments such as workplaces under deadline pressure, where effortful situational adjustment is perpetually interrupted.
Malle's (2006) meta-analysis of 173 published attribution studies challenged the assumption that the effect is universal.3 The actor-observer asymmetry (explaining one's own behaviour situationally while explaining others' dispositionally) showed an average effect size of approximately d = 0.01 to 0.09, near zero overall. The asymmetry held reliably only for negative events, hypothetical scenarios, and interactions with unfamiliar actors. The error is real and consequential in those conditions; the textbook claim that it is pervasive across all social judgement is not supported.
Example
A project misses its deadline. The project director's peers conclude she is disorganised and poor at managing complexity. The evaluators do not ask about the three scope changes the client inserted two weeks before delivery, the reduced headcount from a mid-cycle hiring freeze, or the technical dependency that arrived three weeks late. The behaviour is visible; the constraints are not.
The assessors penalised character because the situational forces were invisible, not because they were absent.
Why it matters
In organisational contexts, the correspondence bias produces systematically inequitable evaluations. When a project fails or a team member underperforms, evaluators attribute the outcome to that person's character (laziness, poor judgement, limited ability) rather than to structural constraints such as inadequate resourcing or unrealistic timelines.2 The consequence is not merely an unfair individual assessment; it is a misallocation of corrective action. An organisation that diagnoses a structural problem as a personnel problem will replace the person, leave the structure intact, and encounter the same failure again.
The bias is correctable. Luong and Butler (2023) found that a single brief reading passage explaining the fundamental attribution error produced measurably lower attributional bias on a subsequent task, compared to a control group who read an unrelated text.4 Targeted awareness, even without formal training, can interrupt the incomplete-correction pattern that Gilbert and Malone identified as the error's proximate mechanism.2
What is the difference between the fundamental attribution error and correspondence bias?+
Fundamental attribution error and correspondence bias describe the same core tendency: over-weighting character and under-weighting situation when explaining others' behaviour. The terms are often used interchangeably. Some researchers reserve correspondence bias for observer judgements only, while fundamental attribution error can include self-directed attributional errors in principle, though the evidence for the self-directed form is weak.
Why do people attribute behaviour to personality instead of situation?+
Characterisation (extracting a trait from a behaviour) happens automatically, without effort or conscious intent. What does not happen automatically is the second step: adjusting that first impression to account for situational constraints. That adjustment requires cognitive resources and is the step most easily interrupted by distraction, time pressure, or cognitive load. When it fails, the dispositional label sticks.
Does the fundamental attribution error apply to how we judge ourselves?+
Less than the textbook version suggests. Malle's 2006 meta-analysis of 173 attribution studies found the actor-observer asymmetry had an average effect size near zero overall. The asymmetry appeared reliably only for negative events and interactions with unfamiliar others, not as a pervasive gap between self-judgement and judgement of others.
How can the fundamental attribution error be reduced?+
Before judging another person's conduct, list two or three situational factors that could produce the same behaviour. This deliberate step simulates the effortful correction that the error bypasses automatically. Even light awareness helps: a single brief explanation of the error measurably reduced attributional bias in a controlled study of 244 adults.
