Actor-Observer Asymmetries

Actor-Observer Asymmetries Definition

Social psychologists speak of an observer perspective when someone perceives, thinks about, or makes a judgment about another person, and they speak of the actor perspective when someone thinks or makes a judgment about himself or herself. So if Jared storms out the door and Evelyn wonders why he does that, Evelyn is in the observer perspective and Jared is in the actor perspective. When the actor and the observer arrive at different judgments, we are faced with an actor-observer asymmetry.

Actor-Observer Asymmetries Importance

Actor-Observer AsymmetriesWhy are actor-observer asymmetries interesting? Actor and observer are the two fundamental perspectives in social cognition: People make judgments either about self or about others; there is no third. So to understand the nature of social cognition, scientists must understand the nature of these two perspectives, especially the conditions under which they differ. That is because some of the biggest challenges of social life involve the discrepancy between actor and observer perspectives. For example, people typically know why they act the way they do, but often they are confused about why others act the way they do. Similarly, to get along with others it isn’t enough to understand our own goals and attitudes; we need to understand other people’s goals and attitudes as well—especially when they might be different from our own. Actor-observer asymmetries cause gaps in people’s understanding of the social world, and scientific research on actor-observer asymmetries tries to identify these gaps and perhaps sharpen people’s tools to bridge the gaps—tools such as explanation, perspective taking, and negotiation.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% OFF with 24START discount code

The Classic Hypothesis of Actor-Observer Asymmetries

The primary actor-observer asymmetry social psychologists have studied is an asymmetry in causal attribution—in how actors and observer explain social behavior or social outcomes. Suppose a student received a D on the statistics exam; why did he or she receive this grade? As observers, we might think the student didn’t study or just isn’t good at statistics. But if the student is asked to explain the D, the student might say that the exam was very hard or that the teacher must have graded it harshly. This difference in explanations is typically described as one between observers citing person causes—causes that reside in the actor (the student didn’t study or lacks ability)—and the actor citing situation causes—causes that lie outside the actor (the exam was hard or the teacher graded harshly). This is in fact what social psychologists Edward E. Jones and Richard Nisbett formulated in 1972 as the now classic actor-observer hypothesis: Actors tend to explain their own behavior with situation causes, whereas observers tend to explain the actor’s behavior with person causes. Virtually all textbooks in social psychology and general psychology mention this hypothesis and describe it as a well-established truth. But a hypothesis is only as good as the research evidence that supports it, so what does the research say?

Empirical Tests of Actor-Observer Asymmetries

Recently, Bertram Malle reviewed more than 100 research articles that had tested the classic actor-observer hypothesis. When the results of all these articles were averaged, there was very little evidence that the hypothesis is true. How little evidence? Researchers can measure the strength of a hypothesis by determining how much better it allows them to predict an event than a blind guess would. If researchers try to predict whether an actor will cite a person cause or a situation cause, they could either guess (e.g., flip a coin), and will by chance be correct in 50% of cases, or they could use the actor-observer hypothesis. If they rely on this hypothesis, they will be correct in 53% of the cases. Thus, the classic actor-observer hypothesis is barely better than a blind guess.

There are situations, however, when the classic actor-observer hypothesis does better. If researchers want to predict how actors and observers explain negative events and if they follow the hypothesis that the actor will provide a situation cause, they will be right in about 57% of the cases. Unfortunately, the opposite happens when they want to predict how actors and observers explain positive events. If they bet again on the actor giving more situation causes, they will be wrong in 56% of the cases. This means that the opposite hypothesis is actually true: For positive events, actors give more person causes and observers give more situation causes. If the classic actor-observer hypothesis holds reasonably true for negative events but the opposite hypothesis holds true for positive events, it means that on average (across events), there may just be no actor-observer asymmetry.

But this finding contradicts intuitions. Actors do know more about their own goals and feelings and about their own history (e.g., past exam grades, past actions). Shouldn’t that lead to an asymmetry between actors and observers in how they explain behaviors and outcomes (even positive ones)? The answer is yes— but the relevant differences cannot be seen if the explanations are interpreted as simple decisions between “person causes” and “situation causes.”

New Hypotheses of Actor-Observer Asymmetries

People’s explanations of behaviors and outcomes are more complex than the person-situation dichotomy suggests. First, people make a sharp distinction between unintentional and intentional events. Unintentional events (e.g., tripping, being sad) are explained by causes, and—if needed—these causes can be classified as located in the person or in the situation. But when it comes to intentional actions, people have a more sophisticated approach. They recognize that one can explain a person’s action by mentioning the reasons the person had for acting—in light of the goal and beliefs held by the person pursuing the action (e.g., “I studied all week because… I knew the test counted for 60% of my grade, and I really want to do well in this class”). Such reason explanations are the most common explanations people give for intentional actions. In addition, people sometimes explain intentional actions by referring to background factors, such as the person’s personality, culture, childhood experiences, unconscious forces—all things that can influence intentional actions but are not the reasons for which the agent chose them (e.g., “She studied all week, never went out because… she is from a hardworking family, she’s very dedicated”). These explanations are called causal history of reason explanations. When Ann says, “I voted for him because I wanted to see a more open-minded social policy,” she is giving a reason explanation; when Blake says, “Ann voted for him because she grew up in a liberal family,” Blake is giving a causal history of reason explanation. Blake’s explanation implies that Ann had some reason, but he may not know the specific reason and therefore offers a background factor that he does know about.

Research shows that actors give far more reason explanations (relative to causal history of reason explanations) than observers do. Knowing about this asymmetry allows us to be right in 67% of cases (and wrong in 33% of cases) when predicting actors’ and observers’ explanations. So this is a powerful asymmetry, and it holds whether the explained action is negative or positive.

There are other features of explanation that show actor-observer asymmetries. Among the reasons people give to explain actions are some that refer to the agent’s thoughts or beliefs that went into the action (called belief reasons) and some that refer to the agent’s goals or desires (called desire reasons). Desire reasons focus on what the agent wants (and doesn’t have), whereas belief reasons highlight the agent’s thinking and rational consideration of the world. Research has found a strong actor-observer asymmetry here: Actors offer more belief reasons (relative to desire reasons) than observers do, and knowing about this asymmetry allows researchers to be right in 62% of the cases when predicting actors’ and observers’ explanations.

There are a few other interesting asymmetries, described in more detail in the literature, but this much is clear: The intuition that actors and observers give different explanations is true after all. But to capture these differences, it isn’t enough to talk about person and situation causes; researchers must consider how people actually explain behavior: with causal histories of reasons, reasons, belief reasons, and so on.

Researchers also have begun to explore why these asymmetries exist and have identified two main processes. One is cognitive: how much the explainer knows about the behavior or outcome. Giving reason explanations, especially belief reasons, requires specific knowledge that observers sometimes lack, and that is in part why actors offer more (belief) reasons. The second process is motivational: whether the explainer is specifically trying to portray the agent (self or other) in a positive light. Here, reasons and especially belief reasons make the agent look more rational and “in control,” so actors prefer to offer those kinds of explanations.

Research on the original actor-observer asymmetry in attributions had a strong impact on the study of other asymmetries, and social psychologists discovered a number of them. For example, in social interactions, actors focus their attention more on their own experiences, whereas observers focus more on the other person’s actions. Also, most people consider their own personality to be more complex and less fixed than other people’s personalities.

What social psychologists have learned from this research is that people face a fundamental challenge in social life: Perceiving, understanding, and reasoning about people are different when they are about oneself than when they are about another person. This challenge must be met, and the gaps between actors and observers overcome, if social interactions are to be successful.


  1. Alicke, M. D., Dunning, D. & Krueger, J. I. (Eds.). (2005). The self in social perception. New York: Psychology Press.
  2. Jones, E. E., & Nisbett, R. E. (1972). The actor and the observer: Divergent perceptions of the causes of behavior. In E. E. Jones, D. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.), Attribution: Perceiving the causes of behavior (pp. 79-94). Morristown, NJ: General Learning Press.
  3. Malle, B. F. (2006). The actor-observer asymmetry in causal attribution: A (surprising) meta-analysis. Psychological Bulletin, 132, 895-919. Retrieved from http://