Attribution Theory in Psychology: Definition & Examples

Attribution theory explains how we infer the causes of our own and other people’s behavior, so that we can understand and predict the social world.

For example, is someone angry because they are bad-tempered, or because something bad happened to them?

Dispositional attribution assigns the cause of behavior to some internal characteristic of a person rather than to outside forces.

When we explain the behavior of others, we look for enduring internal attributions, such as personality traits. This is known as the fundamental attribution error.

For example, we attribute the behavior of a person to their personality, motives, or beliefs.

2. External Attributions (Situational Causes)

The process of assigning the cause of behavior to some situation or event outside a person’s control rather than to some internal characteristic.

When we try to explain our behavior, we tend to make external attributions, such as situational or environmental features.

Heider’s Naive Psychologist

Fritz Heider (1958) asked how ordinary people, not psychologists, make sense of behavior. His answer was that we all act as naive psychologists.

We link the behavior we can see to causes we cannot see. Those inferred causes, rather than the raw behavior, are what give an action its meaning.

Three of Heider’s ideas run through every theory that followed:

• Enduring dispositions: We search behavior for stable traits, because traits make other people predictable.
• Intention: We separate intended acts from accidents, since only intended acts reveal what someone is really like.
• Covariation: We credit causes that are present when the behavior appears and absent when it does not.

Heider also gave us the metaphor that explains the biases below. Behavior is the vivid “figure” in our field of view, while the situation is the receding “ground.” The actor’s personality is therefore what catches the eye.

Heider and Simmel (1944): The Moving Shapes Study

• Aim: To test whether people spontaneously interpret moving geometric shapes as characters with intentions rather than as mere motion.
• Procedure: Women college students watched a short animation of a large triangle, a small triangle and a small disc moving around a rectangle. The main group was simply asked to write down what happened.
• Findings: Of the 34 observers in that group, all but one described a social drama rather than geometry. The large triangle was read as a bully, the smaller shapes as a frightened pair.
• Conclusions: People impose intention and personality even on objectively meaningless movement. Seeing persons-with-causes is a default, not an effortful inference.

The demonstration has been replicated many times and later became foundational for research on theory of mind and the perception of animacy.

Its limits are the artificial stimuli and the interpretive coding of open-ended reports. It shows that people animate what they see, not the rules by which they attribute causes to real behavior.

---

Jones & Davis Correspondent Inference Theory

Jones and Davis (1965) thought that people pay particular attention to intentional behavior (as opposed to accidental or unthinking behavior).

Correspondent Inference Theory (CIT) explains how we link observable actions to stable personality traits.

Edward Jones and Keith Davis developed it in 1965. Their theory holds that observers look for a direct “correspondence” between a behavior and a disposition.

This process gives us a predictable, stable view of the people around us.

Suppose we watch someone act aggressively. CIT explains how we decide whether that person has a “hostile personality” or was simply reacting to provocation.

Preconditions for Attribution

Before making a correspondent inference, an observer must first establish the presence of intentionality.

To link an action to a person’s character, we must believe the actor was not forced by circumstances.

• Capability: The observer must conclude that the actor could physically and mentally perform the action and achieve its effects.

• Knowledge: The observer must believe the actor knew the likely consequences beforehand. An act seen as accidental or forced draws a situational attribution instead.

The Covariation Model is the most comprehensive framework for how we assign causality to behavior.

Harold Kelley developed it in 1967, building on the idea of the “naive scientist.”

Social perceivers work like detectives. They gather data across many observations to decide whether an action comes from disposition or situation.

The core principle is covariation. We attribute an effect to a cause that is present when the behavior occurs and absent when it does not.

---

Three Dimensions of Causal Information

To reach a logical conclusion, an observer systematically evaluates three specific types of information: consensus, distinctiveness, and consistency.

1. Consensus

Consensus refers to how far other people behave the same way toward the same stimulus.

If a large group behaves that way, consensus is high. If the behavior is unique to one individual, consensus is low.

For example, if everyone in a theater laughs at a comedian, consensus is high.

2. Distinctiveness

Distinctiveness measures how the individual responds to different stimuli or situations.

If the person only performs the behavior in one specific context, distinctiveness is high. If they perform the behavior across many different environments, distinctiveness is low.

If Tom only laughs at one specific comedian, his behavior is highly distinctive.

3. Consistency

Consistency evaluates the frequency with which the behavior occurs between the same person and the same stimulus over time.

If the behavior happens reliably every time the situation arises, consistency is high.

If it is a rare, one-off event, consistency is low. If Tom laughs every time he sees this specific comedian, his consistency is high.

---

Patterns of Attribution

Observers combine these three data points into specific patterns to establish the root cause of an action.

Internal (Person) Attribution

We conclude that a behavior is driven by an individual’s personality when consensus is low, distinctiveness is low, and consistency is high.

• Example: Only Tom laughs at the comedian (low consensus), Tom laughs at all comedians (low distinctiveness), and Tom always laughs at this comedian (high consistency). We attribute the laughter to Tom’s jolly disposition.

External (Stimulus) Attribution

We attribute the behavior to the external object or person when consensus, distinctiveness, and consistency are all high.

• Example: Everyone laughs at the comedian (high consensus), Tom only laughs at this one (high distinctiveness), and he always does (high consistency). We conclude the comedian is genuinely funny.

External (Circumstance) Attribution

If consistency is low, we ignore both the person and the stimulus; instead, we attribute the behavior to a “fluke” or unique set of circumstances.

If Tom has never laughed at this comedian before, we assume something peculiar happened that day, such as Tom being in an unusually good mood.

---

Decision-Making with Incomplete Information

When observers lack the historical data required for a full covariation analysis, they rely on causal schemata.

These are mental shortcuts or preconceptions about how certain causes interact to produce effects.

Kelley identified two primary types of schemata:

1. Multiple Necessary Causes: For some outcomes, we assume several factors must all be present simultaneously. To win a marathon, an athlete must possess fitness, motivation, and proper training.

2. Multiple Sufficient Causes: For other behaviors, we assume any single cause could be responsible. If an athlete fails a drug test, they might be cheating, or they might have taken a banned substance by accident. Either explanation is sufficient to explain the outcome.

---

McArthur (1972): Testing the Covariation Model

Kelley’s model was an abstract framework until Leslie McArthur put its rules to an experimental test.

• Aim: To test whether causal attributions really follow Kelley’s rules, and to establish the relative influence of consensus, distinctiveness and consistency.
• Procedure: Participants read one-sentence descriptions of a person responding to a stimulus, such as “John laughs at the comedian.” Each was paired with systematically varied high or low consensus, distinctiveness and consistency information.
• Findings: Attributions broadly matched Kelley’s predictions, but the three information types were not weighted equally. Consistency exerted the strongest influence, distinctiveness a moderate one, and consensus by far the weakest.
• Conclusions: People can reason covariationally when the data are laid out for them. Their systematic under-use of consensus, however, shows that real perceivers fall short of the ideal.

The study turned an abstract model into quantified predictions and confirmed its core structure. Its weakness is that participants were handed tidy, pre-packaged information they rarely possess in everyday life.

Evaluations and Limitations

While the model provides a systematic “ideal” for judgment, it is often viewed as a normative model rather than a descriptive one. Human thought is rarely perfectly rational.

• Information Weighting: Observers do not use all three information types equally. They rely most on consistency and often ignore consensus entirely (McArthur, 1972).

• The Abnormality Focus: We focus on whatever seems “abnormal” against the background of the event. Unusual behavior draws attention and becomes the focus of our attribution.

• Complementary Frameworks: The model overlaps with Correspondent Inference Theory. CIT labels traits from a single act, while Kelley’s model decides between internal and external causes over time.

---

Configuration Model (Kelley)

The human attribution process frequently deviates from the logical “naive scientist” models proposed by Kelley and Jones.

While these theories describe how we should ideally process information, everyday cognition is heavily influenced by cognitive shortcuts known as heuristics.

These mental shortcuts allow for rapid social judgments; however, they often lead to systemic errors that prioritize internal traits over environmental realities.

---

Fundamental Attribution Error (FAE)

The Fundamental Attribution Error, also known as correspondence bias, represents the most pervasive distortion in social perception.

It describes the tendency to overestimate the influence of internal, dispositional factors while underestimating the impact of situational variables when judging others.

• Perceptual Salience: The person acting is the focus of our attention. The situational pressures around them are often invisible or hard to discern.

• The Two-Step Process: Gilbert (1989) proposed that attribution happens in two stages: a rapid internal attribution, then an effortful situational adjustment that is often skipped. The biased internal judgment is left intact.

---

Divergent Perspectives: Actor-Observer and Self-Serving Biases

Our explanations for behavior often shift depending on whose behavior is being evaluated and the outcome of the action.

Actor-Observer Effect (AOE)

The Actor-Observer Effect describes a discrepancy in how we explain our own actions versus the actions of others.

When we are the “actor,” our attention is directed outward toward the environment; therefore, we attribute our behavior to situational forces.

When we are the “observer,” our attention is fixed on the individual, leading us to attribute their behavior to internal traits.

Jones and Nisbett (1971) first described this asymmetry, and the observer’s half of it is simply the fundamental attribution error.

Storms (1973): Reversing the Point of View

If the effect really comes from where attention points, then changing the camera angle should change the attribution. Storms tested exactly that.

• Aim: To test the perceptual-salience explanation of the actor-observer effect by manipulating whom actors and observers could see.
• Procedure: Two participants held a brief, unstructured conversation while two observers watched. Some then re-watched a videotape that reversed the point of view, so actors saw themselves and observers saw the other actor.
• Findings: Control and same-orientation groups showed the usual pattern, with actors situational and observers dispositional. In the reversed group the effect flipped: actors now made more dispositional attributions than observers did.
• Conclusions: Moving the visual focus onto a different person moves the dispositional inference with it. The bias tracks what is perceptually salient and can be corrected by redirecting attention.

The effect is weaker or reversed for positive behaviors, which suggests it is entangled with the self-serving bias rather than being purely perceptual.

Self-Serving Bias

The self-serving bias is a motivational distortion used to protect and enhance self-esteem.

Individuals typically attribute their successes to internal, stable factors, such as high intelligence or hard work.

Conversely, they attribute failures to external, unstable factors, such as a “bad luck” or an unfair task.

A meta-analysis of 69 studies covering 10,515 athletes found a moderate self-serving bias for personal outcomes (Allen et al., 2020).

The size of the bias varied with the athlete’s sex, competitive level and world region. That makes it a flexible way of managing self-esteem rather than a fixed trait. For a fuller account, see self-serving bias.

---

Group-Level and Defensive Attributions

Attributions also serve to reinforce social hierarchies and provide a sense of psychological security in an unpredictable world.

Ultimate Attribution Error (UAE)

The Ultimate Attribution Error, named by Pettigrew (1979), is the group-level extension of the FAE.

Individuals tend to attribute the negative behaviors of “out-group” members to stable, internal traits, which reinforces negative stereotypes.

Conversely, positive behaviors by out-group members are dismissed as situational flukes or luck.

In-group members receive the opposite treatment; their failures are blamed on the situation, while their successes are seen as proof of superior character.

The evidence is thinner than the label suggests. Reviewing nineteen studies, Hewstone (1990) judged support for the error limited and preferred the more cautious term “intergroup attributional bias.”

Just-World Hypothesis and Victim Blaming

The Just-World Hypothesis is a defensive attribution where individuals believe that “people get what they deserve.”

This belief helps manage the anxiety caused by the unpredictability of life.

However, it leads to victim-blaming; this is the tendency to attribute a victim’s misfortune (such as poverty or assault) to their own dispositional flaws.

By blaming the victim, observers reassure themselves that they can avoid similar tragedies by simply making different choices.

---

The Bias Blind Spot

The “bias blind spot” is the tendency to spot biases in other people’s judgments while missing them in your own (Pronin et al., 2002).

Most people believe they are more objective and less susceptible to the Fundamental Attribution Error than the average person.

This illusion of rationality makes it difficult to correct for biased thinking, as individuals assume their own conclusions are based solely on objective facts.

---

Cultural Differences

Psychological inquiry indicates that the “naive scientist” does not operate within a cultural vacuum.

While foundational models suggest a universal cognitive process, the application of attributional logic varies significantly between individualistic and collectivistic societies.

Individualistic cultures, predominant in Western nations, emphasize personal agency and autonomy.

Collectivistic cultures, common in East Asia and Latin America, prioritize social harmony and interdependence.

These cultural lenses dictate whether an observer defaults to dispositional or situational explanations for human conduct.

---

Analytic versus Holistic Cognition

The divergence in attributional styles stems from fundamental differences in how individuals perceive their environment.

Analytic Thinking

Individualistic cultures socialize members to adopt an analytic thinking style.

This approach focuses on the discrete properties of an object or person in isolation from its surroundings.

Consequently, Westerners are predisposed to dispositional attributions; they assume actions reflect an individual’s internal traits or values.

Holistic Thinking

Collectivistic cultures foster a holistic thinking style.

This orientation emphasizes the “whole picture,” accounting for the context, social relationships, and situational pressures surrounding an event.

Individuals from these backgrounds are naturally more attuned to situational attributions, viewing behavior as a reaction to external forces.

---

Neuropsychological and Developmental Evidence

Neuroscience suggests that cultural conditioning leaves a biological footprint on the brain.

fMRI studies point the same way. Asking East Asians to ignore context, or Americans to attend to it, increases activation in higher-order cortical regions.

This indicates that individuals must expend greater cognitive effort to perceive the world in a manner that contradicts their cultural training.

Miller Study (1984)

• Aim: To determine if cultural attribution patterns are innate or learned through socialization.

• Procedure: Americans and Hindu Indians at four age levels described one prosocial and one deviant behavior from their own experience. The levels were roughly 8, 11 and 15 years, plus adults.

• Findings: The youngest children of the two cultures did not differ. The gap emerged and widened with age, with dispositional explanations rising among Americans and contextual ones rising among Indians.

• Conclusions: Humans are born with flexible cognitive styles; specific attributional biases are cultural products acquired over time through social reinforcement.

---

Cultural Mediators of Attributional Biases

Culture significantly alters the prevalence and direction of common cognitive biases, such as the Fundamental Attribution Error (FAE) and the self-serving bias.

Fundamental Attribution Error

In collectivistic societies, the FAE is significantly less prevalent.

While individuals in these cultures recognize personality traits, they rarely ignore the situational context to the extent that Westerners do.

Priming studies have shown that bicultural individuals can shift their attributional style based on cultural cues.

For instance, Hong Kong students primed with American imagery made more dispositional attributions than those primed with Chinese imagery (Hong et al., 2000).

Reversals of the Self-Serving Bias

In the United States and Canada, the self-serving bias is the norm. People take credit for success and blame the situation for failure.

However, in many Asian cultures, this pattern is often reversed due to the value placed on modesty.

• Success: Westerners attribute success to internal talent. Japanese media more often credits an athlete’s coaches and family.

• Failure: Westerners blame the situation for failure. In Japan and Korea, people often blame their own lack of effort, a self-criticism that works as social glue.

---

Defensive and Group-Level Attributions

Cultural context also influences how societies justify inequality and manage intergroup conflict.

Just-World Hypothesis

The defensive belief that “people get what they deserve” is most prevalent in societies with extreme wealth disparities.

Research indicates that participants in India and South Africa score higher on just-world scales than those in more egalitarian Western nations.

In these contexts, attributing poverty to a victim’s flaws serves to justify systemic injustice and alleviate the observer’s anxiety about their own safety.

Ultimate Attribution Error (UAE)

At the group level, the UAE demonstrates how cultural and religious identities skew judgment to favor the “in-group.”

• Taylor and Jaggi (1974):

  • Aim: To observe how intergroup prejudice affects causal attributions.

  • Procedure: Hindu participants in India were asked to explain the positive and negative behaviors of both Hindus (in-group) and Muslims (out-group).

  • Findings: Desirable in-group acts and undesirable out-group acts were both credited to internal dispositions. Undesirable in-group acts and desirable out-group acts were explained away as situational.

  • Conclusions: Attributional logic is often hijacked by group identity to maintain collective self-esteem and reinforce existing social stereotypes.

---

Critical Evaluation

Attribution theory has survived sixty years of testing, but not intact. Its models describe how an ideal perceiver should reason, and real perceivers rarely oblige.

Normative Models Versus Real Perceivers

Kelley’s covariation model and correspondent inference theory are normative accounts. They set out how someone would reason given full information, unlimited time and no motive to distort.

The error-and-bias literature supplies the descriptive picture, and the two rarely match. McArthur’s (1972) finding captures the gap precisely: people do apply covariation logic when the data are handed to them, yet they largely ignore consensus information.

The two bodies of work are best read as a pair. The normative models are the benchmark, and the biases are what actually happens.

The Naive Scientist Overstates Human Rationality

Heider, Jones and Davis, and Kelley all model the perceiver on a scientist calmly weighing evidence. Critics argue this flatters ordinary reasoning.

People are cognitive misers who prefer cheap, single-cause explanations (Fiske & Taylor, 1991). They naively assume that simple explanations are better than complex ones (Jones & Nisbett, 1971). They also fail to gather the consensus data the model assumes, and are swayed by whatever happens to be salient.

Weiner’s work answers part of this charge by treating attribution as emotional and goal-driven rather than coldly analytic.

Culture Limits How Far the Theory Travels

The fundamental attribution error, the self-serving bias and the ultimate attribution error were established largely on Western, individualistic samples.

Because they weaken, vanish or reverse in collectivistic settings, they cannot be assumed pan-cultural (Miller, 1984; Morris & Peng, 1994). Much classic work risks presenting a Western pattern as human nature.

That makes “fundamental” a misnomer. Miller’s developmental data show the dispositional default grows with age inside individualistic cultures rather than arriving with the human mind.

Contemporary Research

Research since 2015 has stopped asking whether the classic biases exist and started measuring how large, how universal and how modifiable they are.

How Large Is the Self-Serving Bias?

• Aim: To quantify whether athletes claim success internally and deflect failure externally, and to identify what moderates the effect.
• Procedure: Allen, Robson, Martin and Laborde (2020) pooled 69 studies of 10,515 athletes. Standardised mean differences compared internal and external attributions for success and failure, with formal tests for moderators and publication bias.
• Findings: A moderate bias for personal outcomes (SMD 0.62) and a comparable one for team attributions (SMD 0.63). The effect was moderated by sample sex, performance level and world region.
• Conclusions: The self-serving bias is reliable and moderate well beyond the laboratory. Its size is conditional, which fits a flexible esteem-management strategy rather than a hard-wired universal.

As a synthesis of 69 studies, this outweighs any single experiment. Its limits are that the pooled studies are mostly cross-sectional and self-report, and publication bias inflates the estimate somewhat.

The significant world-region moderator matters for the cultural argument above. The meta-analysis confirms the bias and re-states its cultural boundedness in the same breath.

Can the Bias Be Turned Down?

Genschow, Rigoni and Brass (2017) ran six studies on belief in free will and the correspondence bias. The stronger a person’s belief in free will, the larger their bias.

Two of the six were experiments. Weakening belief in free will experimentally reduced the bias, which in turn changed how harshly participants prescribed punishment and reward.

That carries a causal claim: an ideological belief, not just cognitive load, moderates over-attribution. Read alongside the meta-analysis, the modern evidence retains the classic biases but re-describes them as moderate, culturally graded and modifiable.

References

Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), 49–74. https://doi.org/10.1037/0021-843X.87.1.49

Allen, M. S., Robson, D. A., Martin, L. J., & Laborde, S. (2020). Systematic review and meta-analysis of self-serving attribution biases in the competitive context of organized sport. Personality and Social Psychology Bulletin, 46(7), 1027–1043. https://doi.org/10.1177/0146167219893995

Eiser, J. R. (1983). From attributions to behaviour. In M. Hewstone (Ed.), Attribution theory: Social and functional extensions. Blackwell.

Fiske, S. T., & Taylor, S. E. (1991). Social cognition (2nd ed.). McGraw-Hill.

Genschow, O., Rigoni, D., & Brass, M. (2017). Belief in free will affects causal attributions when judging others’ behavior. Proceedings of the National Academy of Sciences, 114(38), 10071–10076. https://doi.org/10.1073/pnas.1701916114

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.

Heider, F. (1958). The psychology of interpersonal relations. John Wiley & Sons. https://doi.org/10.1037/10628-000

Heider, F., & Simmel, M. (1944). An experimental study of apparent behavior. The American Journal of Psychology, 57(2), 243–259. https://doi.org/10.2307/1416950

Hewstone, M. (1990). The “ultimate attribution error”? A review of the literature on intergroup causal attribution. European Journal of Social Psychology, 20(4), 311–335. https://doi.org/10.1002/ejsp.2420200404

Hong, Y. Y., Morris, M. W., Chiu, C. Y., & Benet-Martinez, V. (2000). Multicultural minds: A dynamic constructivist approach to culture and cognition. American Psychologist, 55(7), 709–720. https://doi.org/10.1037/0003-066X.55.7.709

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. https://doi.org/10.1016/S0065-2601(08)60107-0

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.

Kelley, H. H. (1973). The processes of causal attribution. American Psychologist, 28(2), 107–128. https://doi.org/10.1037/h0034225

Kelley, H. H. (1983). Perceived causal structures. In J. M. F. Jaspars, F. D. Fincham, & M. Hewstone (Eds.), Attribution theory and research: Conceptual, developmental and social dimensions (pp. 343–369). Academic Press.

Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. Plenum Press. https://doi.org/10.1007/978-1-4899-0448-5

McArthur, L. A. (1972). The how and what of why: Some determinants and consequences of causal attribution. Journal of Personality and Social Psychology, 22(2), 171–193. https://doi.org/10.1037/h0032602

Miller, D. T., & Ross, M. (1975). Self-serving biases in the attribution of causality: Fact or fiction? Psychological Bulletin, 82(2), 213–225. https://doi.org/10.1037/h0076486

Miller, J. G. (1984). Culture and the development of everyday social explanation. Journal of Personality and Social Psychology, 46(5), 961–978. https://doi.org/10.1037/0022-3514.46.5.961

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

Pettigrew, T. F. (1979). The ultimate attribution error: Extending Allport’s cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5(4), 461–476. https://doi.org/10.1177/014616727900500407

Pronin, E., Lin, D. Y., & Ross, L. (2002). The bias blind spot: Perceptions of bias in self versus others. Personality and Social Psychology Bulletin, 28(3), 369–381. https://doi.org/10.1177/0146167202286008

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

Taylor, D. M., & Jaggi, V. (1974). Ethnocentrism and causal attribution in a South Indian context. Journal of Cross-Cultural Psychology, 5(2), 162–171. https://doi.org/10.1177/002202217400500202

Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4), 548–573. https://doi.org/10.1037/0033-295X.92.4.548

Weiner, B. (1986). An attributional theory of motivation and emotion. Springer-Verlag. https://doi.org/10.1007/978-1-4612-4948-1

Weiner, B., & Kukla, A. (1970). An attributional analysis of achievement motivation. Journal of Personality and Social Psychology, 15(1), 1–20. https://doi.org/10.1037/h0029211