Read this topic

Every time you watch someone act, your mind automatically asks: why did they do that? Attribution theory (the study of how people explain the causes of behaviour, their own and other people's) is the framework psychology uses to describe the predictable, and often biased, answers people reach for. For example, imagine a driver who cuts you off in traffic: within a split second, most people have already silently decided why, and that split-second explanation shapes everything that follows, from your mood to how you react.

An internal, or dispositional, attribution (explaining behaviour by the person's own character or traits) blames the person: they cut you off because they're a rude, inconsiderate driver. An external, or situational, attribution (explaining behaviour by circumstances outside the person) blames the context: they cut you off because they were rushing a sick relative to hospital. Every attribution question in this topic reduces to a version of this same split, asked about different people and different outcomes.

The trap almost everyone falls into, automatically and without noticing, is the fundamental attribution error (the tendency to over-attribute other people's behaviour to their disposition while underweighting the situation they were in). Going back to the driver: the automatic assumption is usually the internal one, they're simply careless, while the powerful situational explanation, an emergency, gets ignored entirely. This error is called "fundamental" precisely because it isn't an occasional lapse; it's the default, near-automatic way people judge others' behaviour, which is exactly why it needs to be actively corrected for rather than assumed away.

Explaining behaviour serves a real cognitive purpose: it lets us predict what someone will do next and decide how to respond to them, which matters for social coordination and safety. Dispositional explanations are cognitively cheap and stable: "she's like that" requires no further updating, while situational explanations require tracking context, which is more effortful. This cognitive shortcut-seeking is a large part of why the fundamental attribution error is so persistent: a dispositional explanation is simply easier to reach for and easier to hold onto than a situational one, even when the situational explanation is more accurate.

The actor-observer bias (the tendency to explain our own behaviour situationally but explain other people's behaviour dispositionally) asks a different question from the fundamental attribution error: whose behaviour is being judged? If you were the one who cut someone off, you would likely explain it by pointing to the situation, being late, not seeing the other car, rather than concluding you are a careless person; the bias is about a shift depending on perspective, self versus other. Self-serving bias (crediting our own successes to internal, personal causes, while blaming our own failures on external, situational causes) protects self-esteem rather than describing a general perceptual habit: a medical student who attributes an exam pass to their own hard work and ability, but an exam failure to an unfairly difficult paper, is showing self-serving bias. The just-world hypothesis (the belief that people generally get the outcomes they deserve, that the world is fundamentally fair) sits slightly apart, because it is a belief about outcomes in general rather than an explanation of one specific action: someone who hears about a robbery victim and responds that the victim "shouldn't have walked there at night" is reasoning from a just-world belief. Bernard Weiner extended the internal-external split by adding a second dimension, stability, producing four combinations: ability is internal and stable, effort is internal and unstable, task difficulty is external and stable, and luck is external and unstable. A patient who fails to improve with treatment and attributes this to being "simply a weak person" is making an internal, stable attribution to low ability, while a patient attributing the same outcome to "not trying hard enough that week" is making an internal, unstable attribution to effort, and these two attributions carry very different implications for future motivation, even though both are internal.

These four ideas get blurred together constantly, and the exam rewards untangling them precisely. Fundamental attribution error versus actor-observer bias: the FAE is specifically about judging OTHERS' behaviour dispositionally in general, while the actor-observer bias is about the SHIFT that happens depending on whether you're explaining your own behaviour or someone else's: the FAE is really just one half of the actor-observer pattern (the "observer" half). Actor-observer bias versus self-serving bias: the actor-observer split depends on WHO is being judged (self versus other), while the self-serving split depends on the OUTCOME (success versus failure): a person can be self-serving about their own success and failure without any other person being involved at all. Just-world hypothesis versus the other three: the first three all explain a specific behaviour or action, while the just-world hypothesis explains a life OUTCOME as deserved, making it a belief system rather than a moment-to-moment attribution bias.

Attribution biases aren't diagnosed in an individual patient the way an illness is, but attribution STYLE, a person's habitual pattern of explanation, is measured using structured tools, most notably the Attributional Style Questionnaire (ASQ), which presents hypothetical positive and negative events and asks respondents to rate their likely explanation along three axes: internal versus external (locus), stable versus unstable (stability), and global versus specific (does the explanation affect just this situation or everything). Researchers and clinicians code a person's answers along these three axes to characterise their explanatory style, most importantly to identify the depressive attributional pattern described below.

Attribution style has direct clinical relevance, especially in depression. A depressive attributional style tends to explain negative events as internal (my fault), stable (this will always be true), and global (this affects everything), while explaining positive events as external, unstable, and specific: a patient who, after one setback at work, concludes that they always fail at everything and it is entirely their fault is showing exactly this internal-stable-global pattern for a single negative event. Recognising this attributional pattern is a core part of cognitive models of depression, and cognitive behavioural therapy (CBT) directly targets it: attribution retraining, a structured technique within CBT, helps patients practise reattributing negative events to more specific, unstable, and appropriately external causes, and positive events to more internal, stable causes, which is associated with improved mood and reduced hopelessness. Beyond the therapy room, awareness of the fundamental attribution error and self-serving bias has a second practical application: it should make clinicians pause before dispositionally blaming a patient (labelling them "non-compliant" or "difficult") when a situational explanation (cost of medication, lack of transport, side effects) may better explain the same behaviour.

Examiners often build vignettes that test whether a candidate can classify an attribution along all three of Weiner's-plus-globality axes at once, not just internal-versus-external. A patient who says "I failed because I'm useless and I always mess everything up" is making an internal, stable, AND global attribution, the depressive triad, while a patient who says "I failed because that exam was unusually hard" is making an external, and likely unstable and specific, attribution. The trap is stopping at "internal or external" when the question is really testing whether you can also spot the stability and globality dimensions layered on top.

Every attribution reduces to internal-versus-external, but the biases that distort it, fundamental attribution error, actor-observer bias, self-serving bias, and the just-world hypothesis, are told apart by asking whose behaviour is being judged and whether the outcome was a success or a failure.