Attribution is the psych/soc topic that shows up most reliably in one specific form: a passage describes someone explaining a behavior, and the question asks which bias is operating. Almost all of these resolve if you can track two things — whose behavior is being explained, and whether the explanation points inward or outward.

The Basic Split

Dispositional (internal) attribution explains behavior by something about the person — personality, ability, effort, character. "He failed because he's lazy."

Situational (external) attribution explains behavior by circumstances — environment, luck, task difficulty, social pressure. "He failed because the exam was unfair."

That is the whole framework. Every bias below is a systematic tilt toward one of these two.

Kelley's Covariation Model

The formal account of how people should assign attributions, using three dimensions:

Consensus — do other people behave this way in this situation? High consensus means everyone does it.

Distinctiveness — does this person behave this way only in this situation, or everywhere? High distinctiveness means it is specific to this situation.

Consistency — does this person behave this way in this situation repeatedly over time?

The combinations that matter:

• High consensus + high distinctiveness + high consistency → situational. Everyone does it, only here, every time. Something about the situation is causing it.
• Low consensus + low distinctiveness + high consistency → dispositional. Only this person does it, and they do it everywhere, always. Something about them is causing it.

Low consistency generally points to a one-off circumstance rather than a stable cause either way.

If a question gives you three pieces of information about a behavior across people, situations, and time, it is testing Kelley. Sort them into the three categories and read off the answer.

Fundamental Attribution Error

Overweighting dispositional causes when explaining other people's behavior, and underweighting situational ones.

Someone cuts you off in traffic and you conclude they are a jerk, rather than that they might be rushing to a hospital. You see the person clearly and the situation poorly, so the person absorbs the explanation.

This is the most-tested item in the whole topic. Key restriction: it applies to others' behavior.

The standard explanation is perceptual salience. The actor is the figure; the situation is the ground. Whatever is in your attentional foreground gets the causal credit.

It is also culturally variable — the effect is stronger in individualist cultures than in collectivist ones, where situational explanations come more readily. This shows up in passages contrasting Western and East Asian samples.

Actor-Observer Bias

The extension: dispositional for others, situational for yourself.

You attribute your own lateness to traffic and someone else's lateness to disorganization.

Two mechanisms are usually cited. First, informational — you know your own history and can see that you are usually punctual, while you have no such data on the other person. Second, perceptual — when you act, your visual field is filled with the situation, not with yourself.

The distinction from FAE: FAE is only about explaining others. Actor-observer is the comparison between how you explain others and how you explain yourself.

Self-Serving Bias

Internal attributions for your successes, external attributions for your failures.

You aced the test because you're smart. You bombed it because the professor writes bad questions.

This is motivational rather than perceptual — it protects self-esteem. Note that it runs opposite to actor-observer for the success case: actor-observer predicts you explain your own behavior situationally, but self-serving bias predicts you take internal credit when things go well. The resolution is that self-serving bias is about valence. Good outcomes get pulled inward, bad outcomes pushed outward.

Depression is associated with a reversed pattern — internal, stable, global attributions for failure — which connects to learned helplessness.

Optimism Bias and Just-World

Optimism bias — believing you are less likely than others to experience negative events. Smokers underestimating their own cancer risk.

Just-world hypothesis — the belief that people get what they deserve. This produces victim-blaming: if something bad happened to someone, they must have done something to warrant it. It is dispositional attribution applied defensively, because the alternative (bad things happen randomly) is threatening.

Just-world shows up in passages about attitudes toward poverty, illness, or crime victims.

Related Concepts That Cluster With This

Locus of control (Rotter) — a stable personality trait rather than a situational attribution. Internal locus: outcomes result from your own actions. External locus: outcomes result from luck, fate, or powerful others. Internal locus correlates with better academic and health outcomes.

Do not confuse locus of control with attribution style. Attribution is about a specific event; locus of control is a general disposition.

Attribute substitution — when a question is hard, people unconsciously answer an easier one instead. This underlies several heuristics.

Halo effect — a global positive impression of someone bleeds into judgments of specific unrelated traits. Attractive people are rated as more intelligent and kinder. This is dispositional attribution generalized past the available evidence.

The Discrimination Questions

Passages usually give you a scenario and four bias options. The fast route:

1. Whose behavior? Own → self-serving or actor-observer. Someone else's → FAE.
2. Success or failure? If the explanation shifts based on outcome valence → self-serving.
3. Is it a comparison? If both self and other are being explained differently → actor-observer.
4. Is the target a victim? → just-world.
5. Three data points about consensus/distinctiveness/consistency? → Kelley.

The most common trap is FAE vs. actor-observer. If the scenario only involves explaining someone else, it is FAE. If it explicitly contrasts self-explanation with other-explanation, it is actor-observer.

The second trap is self-serving vs. actor-observer for your own behavior. Check whether the outcome was good or bad. Self-serving is outcome-dependent; actor-observer is not.