MCAT content library / Psychology & sociology / Attribution: weigh the person and the situation
An observed action can fit several explanations. Ask which alternatives the person had and which comparison would separate the causes.
About 16 minutes, including the worked examples and checks.
On this page
An action is evidence; an attribution is an explanation
A volunteer sends a brief reply to a detailed request. An observer might infer impatience, an overloaded queue, a restrictive messaging interface, or some combination. The reply is observed; its cause is not directly visible. Attribution is the process of explaining behavior through features of the person, the situation, or their interaction. AAMC category 8B asks how environmental information, cultural context, and self-perceptions shape those explanations.1
A dispositional explanation invokes a person’s characteristics, such as a preference or a tendency that extends beyond the immediate event. A situational explanation invokes circumstances such as instructions, incentives, workload, or available alternatives. The fundamental attribution error concerns giving the person too much explanatory weight while giving relevant situational constraints too little. It does not mean that every dispositional explanation is wrong.2
Ask what behavior was actually possible
If a person could freely choose among several responses, the chosen response may be informative about preferences. If an interface allowed only one response, observing it gives much less information about which response the person would prefer. The action can look identical in the two situations even though its evidential value differs. Classic attribution research examined how knowledge of constrained choice changes judgments of an actor’s attitude.3
This is a reasoning problem about alternatives. Before concluding that someone is habitually unhelpful, ask whether helping was available, understood, and compatible with the task’s constraints. Acknowledging a constraint does not require deciding that the person is exceptionally helpful. Sometimes the warranted conclusion is that the available observation does not distinguish the competing explanations.
Worked experiment: change the observer’s information
In an original teaching experiment, a volunteer answers a visitor using a three-word message. The volunteer’s interface required selection from prewritten messages; no longer answer could be sent. Eighty observers watch the same recording. Researchers randomly assign 40 observers to receive the accurate interface explanation and 40 to receive only neutral information about the recording’s length. Neither group meets the volunteer.
| Information supplied | Mean unhelpfulness rating, 0–10 | Mean predicted length of a future unrestricted reply |
|---|---|---|
| Interface restriction explained | 4.0 | 18 words |
| Recording length explained | 7.0 | 8 words |
The restriction-information group gives a rating three points lower. Its prediction for an unrestricted reply is ten words longer. The volunteer did not change across these conditions: the manipulated variable was what observers knew. Thus, the experiment concerns attribution and prediction by observers, not improvement in the volunteer’s behavior.
The random assignment supports attributing the observed group contrast to the information condition within this experiment. It does not make 4.0 the objectively correct personality rating, establish the volunteer’s enduring disposition, or show that each individual observer revised a previous judgment. Only post-information group means were measured. A lower mean is also not evidence that all informed observers made the same inference.4
The added explanation makes a situational account available and weakens the diagnostic value of brevity. The neutral-information condition controls for receiving an extra statement, but the content necessarily differs. If the explanation also cues what researchers expect, demand characteristics remain an alternative explanation for part of the rating difference. A private prediction of later behavior or an additional less explicit manipulation could help evaluate that concern.
Test the explanation with a new comparison
To learn about the volunteer, obtain behavior where the restriction is absent. Use comparable requests and opportunities, vary interface condition in a counterbalanced order, and observe more than one interaction. If replies become detailed whenever the restriction is removed, an explanation based on fixed unwillingness alone becomes less adequate. It still does not follow that the volunteer has no stable traits.
Three comparisons are useful. First, does the same person behave similarly when the same situation recurs? Second, does the person’s behavior change when the relevant circumstance changes? Third, do other people behave similarly under that circumstance? Repeated brevity in the same restrictive interface supports repeatability, but it does not isolate a stable personal preference: the constraint repeats too.
Suppose another volunteer gives equally short replies under the restriction and detailed replies without it. That comparison strengthens the case that the interface matters. It does not prove identical motives in both people. Conversely, if one volunteer stays brief in every setting while others become detailed, the evidence permits a stronger person-specific hypothesis, provided the relevant opportunities and task understanding are comparable.
Your own response can become an untested reference point
Observers sometimes treat their own choices as unusually representative. The false-consensus effect describes a tendency to estimate greater agreement with one’s own response and to interpret a different response as more revealing of the other person’s disposition. It is a claim about judgments under particular conditions, not a rule that every estimate of agreement is inflated.5
Imagine an observer who likes short messages and assumes most visitors do too. They may interpret the volunteer’s brevity as ordinary efficiency. Another observer who values detailed explanations may read the same message as unusually dismissive. Neither observer has measured what this visitor wanted. Asking about the recipient’s request and the available interface is more informative than projecting the observer’s preference onto everyone involved.
Self-serving attribution is a related but distinct pattern: explanations protect a favorable self-view, such as crediting one’s success to ability while explaining a comparable failure through external obstacles. The question is whether the explanatory standard changes with the self-relevant outcome. An external explanation can nevertheless be accurate; calling a statement self-serving requires more than noticing that it is convenient.2
Cultural context changes hypotheses, not the rules of evidence
Cultural learning can influence which causes receive attention and how a person-in-context is understood. Research comparing sampled cultural settings has examined differences in attention to situational constraints, while also finding dispositional reasoning across settings. That combination is incompatible with assigning one exclusive reasoning style to every member of a large cultural category.6
If two sampled programs differ in average trait ratings, examine the tasks, sampling, language, and information available before attributing the difference to culture. A program might teach its trainees about the specific constraint; translation might change the apparent freedom of choice. Group averages cannot establish what an unmeasured individual thinks, and a two-program comparison cannot demonstrate a universal cultural hierarchy.
For a new vignette, separate the observed action, the proposed cause, and the evidence that would distinguish it from alternatives. Then check whether the conclusion concerns an observer’s belief or the actor’s behavior. Explaining why a judgment changes and establishing what caused the underlying action are different research goals.
ACTIVE PRACTICE · LOCALIZE AND TRANSFER
Change What Observers Know Without Changing the Person
The original recording shows the same volunteer giving a three-word reply because their interface permitted only prewritten messages. Forty observers hear a correct restriction explanation and forty receive neutral recording-length information. Mean unhelpfulness ratings: 4.0 versus 7.0; predicted future unrestricted replies: 18 versus 8 words.
Show the Evidence Boundary
Experimental comparisons support bounded claims; other processes and unmeasured causal pathways remain possible. This is a learning exercise rather than calibrated score prediction.
Check your reasoning
Choose an answer before opening its feedback. Each option explains the reasoning behind that choice.
Check 1
A reviewer infers that a trainee dislikes collaboration because the trainee declines a joint assignment. The reviewer then learns that the trainee’s supervisor prohibited participation to avoid a scheduling conflict. What change in inference is most warranted?
A. Conclude that the trainee strongly prefers every collaborative task.
Review this choice. The constraint weakens evidence for dislike, but does not establish the opposite preference.
B. Treat the refusal as less diagnostic of a stable dislike of collaboration.
Correct. A directly relevant restriction provides another explanation for declining. The refusal supplies less information about what the trainee would freely choose.
C. Treat the scheduling restriction as proof that personal preferences never matter.
Review this choice. One constrained decision does not rule out personal influences in other settings or in other aspects of the decision.
D. Use the refusal as stronger evidence of dislike because a supervisor was involved.
Review this choice. The added information restricts choice; it does not strengthen the inference that refusal expresses a stable preference.
The principle to carry forward
Information about unavailable alternatives changes what an observed choice can reveal.
Check 2
A dispatcher writes terse instructions only in a system that limits message length. A colleague proposes that the dispatcher is generally unwilling to explain. Which observation would most directly challenge that explanation as sufficient?
A. A second observer also calls the restricted messages terse.
Review this choice. Agreement about the observed brevity does not distinguish its cause.
B. Many more messages from the same restricted system are terse.
Review this choice. The proposed constraint repeats in all those observations, so repetition alone does not isolate the person.
C. The dispatcher gives detailed instructions for comparable requests when the length restriction is removed.
Correct. Holding the person and request type comparable while changing the relevant restriction tests whether stable unwillingness alone accounts for the behavior.
D. The dispatcher reports preferring short messages in an unrelated social setting.
Review this choice. That preference could support a person-level hypothesis, but it does not directly test the effect of the specific system constraint.
The principle to carry forward
Repeat observations across the relevant change in context, not only within one unchanged setting.
Check 3
Students from two programs read the same translated vignette. Mean dispositional-attribution ratings differ, but the researchers have not checked translation equivalence or sampled beyond those programs. Which conclusion is supported?
A. A rating difference occurred in these samples under the administered versions of the task.
Correct. This describes the observed comparison while leaving translation, sampling and causal interpretation open.
B. All members of the first program’s broader cultural group ignore situational causes.
Review this choice. A sample mean cannot establish a rule about every member of a cultural group, and the measure does not test every possible cause.
C. Cultural membership caused the difference independently of task interpretation.
Review this choice. The study does not rule out interpretation differences and did not manipulate cultural membership.
D. The group with the lower mean made objectively correct attributions in every case.
Review this choice. The supplied mean does not provide a correctness criterion or individual-level results.
The principle to carry forward
Keep a sampled comparison separate from a universal or causal claim about culture.
Continue learning
Related lessons
- Personality explanations and the situations they must explain
- Experimental Controls: What a Comparison Can Show
- Self-concept, self-efficacy, and perceived control
Browse psychology & sociology · MCAT content library
MCAT preparation and roadmap · Aim High, Score Higher
Sources
1. AAMC: P/S Content Category 8B
2. OpenStax Psychology 2e: What Is Social Psychology?
3. Jones and Harris (1967): The Attribution of Attitudes
4. OpenStax Psychology 2e: Analyzing Findings
5. Ross, Greene, and House (1977): The False Consensus Effect
6. Choi, Nisbett, and Norenzayan: Causal Attribution Across Cultures