Attribution Notes

Attribution

Heider & Simmel Demo (1944)

  • Demonstrated that people automatically infer mental states of triangles and lamps.
  • This suggests it is natural to describe movements in terms of intentions, desires, and affective states.
  • Raises the question of whether we truly see intentions, desires, and states or if we are merely inferring invisible mental states.

Inferring Invisible States & Traits

  • We infer traits --> states --> intentions --> behavior.
  • It is useful to know the states and traits of others for empathy and predicting future behavior.

Attribution Process

  • Dispositional factors (traits): Internal characteristics (e.g., the ball is round).
  • Situational factors: External circumstances (e.g., someone hits the ball).
  • People tend to focus on one type of factor while ignoring the other.
Kelley’s (1967) Covariation Model
  • Distinctiveness: Does the person act this way in many situations? High distinctiveness means the behavior is specific to a situation.
  • Consensus: Do other people act this way in this situation? High consensus means the behavior is common.
  • Consistency: Does the person act this way consistently over time in this situation? High consistency suggests something about the person or the situation.
Example
  • Observing Debbie acting depressed in a conversation with a new person:
    • Distinctiveness: If Debbie always acts depressed, it's low distinctiveness, suggesting something about Debbie.
    • Consensus: If everyone looks depressed in this situation, it’s high consensus, suggesting something about the situation.
    • Consistency: If Debbie always looks depressed in this situation, it’s high consistency, suggesting something about Debbie.
Limitations of Covariation Analysis
  • It describes what we should do, not what we actually do.
  • It assumes we observe a lot of behavior over time, which is often not the case.

Correspondent Inference Theory (Ned Jones, 1965)

  • Attributions are based on a single behavior.
  • Formula: Behavior=Dispositions+SituationalFactorsBehavior = Dispositions + Situational Factors or B=D+SB = D + S
  • Therefore, D=B−SD = B - S, meaning DispositionalHostility=BehavioralHostility−SituationalPressureforHostilityDispositional Hostility = Behavioral Hostility - Situational Pressure for Hostility
Situational Constraints
  • We share knowledge of social norms.
  • We know how situations typically affect people.
    • Normative Behavior (B=SB = S): No dispositional inference should occur when a social norm creates the behavior. If a level 75 hostility behavior occurs with a level 75 situational provocation, we shouldn't assume the person is dispositionally hostile.
    • Counternormative Behavior (B > S): Suggests a correspondence between the behavior and internal dispositions. If behavior is stronger than the situational provocation, it must be due to disposition.
    • Example: Making lots of noise in the library.
Examples with Mark and Bill
  • Scenario 1: Mark punches Bill after Bill says he is sleeping with Mark's wife.
    • High situational provocation.
    • Conclusion: Mark is not necessarily a hostile person in general.
  • Scenario 2: Mark punches Bill after Bill bumps into Mark.
    • Low situational provocation.
    • Conclusion: Mark might be a pretty hostile person in general.

The Castro Study (Jones & Harris, 1967)

  • Subjects read pro-Castro or anti-Castro essays supposedly written by other students.
  • Subjects were told that the essay valence was either freely chosen or required.
  • Subjects had to rate how pro-Castro the essay writer really is, measuring the dispositional attitude.
  • Prediction: If the student was required to write pro-Castro, this should prevent dispositional attribution; the behavior could be attributed to requirement rather than true attitude.
Results
  • Even when students were required to write pro-Castro essays, subjects still rated them as having more pro-castro attitudes, although it was less pronounced than when students freely chose to write pro-Castro.
  • This is the correspondence bias.

The Correspondence Bias (Jones & Harris, 1967) / Fundamental Attribution Error (Ross, 1977)

  • The tendency to make dispositional attributions from observed behavior even when situational influences fully account for the behavior.
  • Disposition = Behavior - Situation
  • A major reason why “we don’t know why people do what they do.”

Why do we commit the FAE? (Gilbert & Malone, 1995)

  1. Overlooking Situational Constraints

    • Situations are often invisible.
      • Roles that we take on.
      • We ourselves can be the situation for others.
    • Situations are often in the mind of the actor.
      • Their subjective construal that you can’t see
      • e.g., previous threats.

  2. Inaccurate Theories of Situational Influence

    • Corollary 2a: We are unaware of the power of situations

  3. Salience

    • We don’t commit the FAE for ourselves, only for others. Why?
    • Partly for motivational reasons, but…
    • We are aware of how situations are influencing us more than we are aware of our own behavior.
      • I know why I was late, but for you I just see the behavior and judge you as irresponsible
    • Salience = how attention getting something is. Salient things are attention getting
      • Actor-Observer Effect
    • Consequence of Differential Attributions for self & other
      • In new meeting, don’t put yourself out there because of fear of rejection, but assume partner doesn’t put herself out there because she doesn’t like you

  4. Incomplete Corrections (Dan Gilbert, 1989)

    • Sequential Operations Model
      • Behavioral Characterization: Automatic
      • Initial Dispositional Attribution: Automatic
      • Situational Correction: Controlled
    • Corollary 3b: We often don’t know the bias in our judgments because they are made automatically
    • Example: “hostile punch” --> “hostile person” --> “It’s a play? Maybe not SO hostile” (only if motivated & not under load)
      • Gilbert, Pelham & Krull (1988)
        • Start with question about dispositional attitude – “How anxious is this woman in everyday life?”
        • See list of discussion topics
          • Anxiety-provoking topics
          • Calming topics
        • Subjects were made cognitively busy (or not)
        • Watch video of woman behaving anxiously
        • Assessed person’s true level of dispositional anxiety
      • Under cognitive load, anxious behavior is taken to indicate an anxious disposition
      • Without load, this initial attribution is corrected/adjusted based on situational information
      • Even when people correct, they are still making a strong dispositional attribution
      • Why? They anchor on the behavior as a starting point

  5. Cultural Bias?

    • East Asians explicitly value sensitivity to contextual factors more than European Americans (Norenzayan et al., 2002)
      • How much do you consider _ factors?
    • But when actually tested, they look the same (Lieberman et al, 2005)

Circumscribed Accuracy (Swann, 1984)

  • Why would we be built this way?
  • Why hasn’t the world gone to hell as a result?
  • We tend to see people in the same situation.
    • Obedient waiters
    • Sometimes you are the situation, so whenever you see them, the situation is present.
  • People choose their situations.
    • But not in experiments
  • Despite FAE, can still predict behavior, even though we might have the wrong explanation for it.