AD

L2: Attribution, Motivated Reasoning, and Self in Social Psychology

Attribution, Covariation, and External vs Internal Explanations

  • Core idea across attribution theories: we make inferences about why people behave the way they do, often favoring internal explanations (about the person) over external ones (about the situation), because internal explanations feel more predictive and coherent for future behavior.

  • Discounting principle (external causes can reduce the perceived internal cause): when external factors could explain behavior, we discount the role of internal traits or motives.

    • Classic illustration (1960s study): participants interacted with a graduate student who either praised or criticized them.
    • Two conditions:
    • Condition A: Flattery or criticism only.
    • Condition B: Flattery or criticism plus an invitation to participate in a future study.
    • Findings:
    • In Condition A (no further request), flatterers were liked more than flatterers who received a follow-up request, and critics were liked least.
    • In Condition B (follow-up request), flattery followed by a request was discounted; people assumed the flatterer’s politeness was self-serving.
    • Conclusion: external constraints (the study invitation) reduced the impact of flattery on liking; people discount external motives when attributing behavior.

  • Attribution theory and its extreme forms can be cognitively demanding and occasionally unrealistic in everyday life; critics argue it overfits to multi-variable scenarios and is not always used in real-time judgments.

  • Covariation model (Kelley): we attribute behavior to one of three sources by looking for covariation across three dimensions: the person, the entity (the target of the behavior), and the circumstances.

    • Three sources of attribution:
    • The person (dispositional attribution): quirks or personality traits of the actor.
    • The entity (the target of the behavior): properties of what is being acted upon (the object of the behavior).
    • The circumstances (the context): situational factors in which the behavior occurred.
    • Three diagnostic criteria used to infer the source:
    • Distinctiveness: does the behavior occur with this entity but not with others?
      • High distinctiveness means the behavior is unique to this entity; low distinctiveness means it occurs with many entities.
    • Consensus: do others react similarly in the same situation?
      • High consensus means many people react the same way; low consensus means few do.
    • Consistency: does the actor behave this way across time and circumstances?
      • High consistency means the behavior is repeatable; low consistency means it’s variable.
    • MacArthur et al. (1972) study on laughter at a comedian:
    • Variants combined the three criteria (distinctiveness, consensus, consistency) with a baseline participant named John.
    • Conditions manipulated:
      • Consensus: high (most people laughed) vs. low (only John laughed).
      • Consistency: high (John always laughs) vs. low (John seldom laughs).
      • Distinctiveness: high (John laughs at this comedian but not others) vs. low (John laughs at all comedians).
    • Findings: attributions tended to differ depending on the pattern of criteria; sometimes behavior was attributed to the person (John), sometimes to the entity (the comedian), and sometimes to the situation (the circumstances), depending on the combination of criteria.
    • Key takeaway: attribution is a function of multiple interacting criteria, not a single cue.
    • Practical implications and criticisms:
    • The model is complex and may not map neatly onto everyday judgments.
    • In real life, people often rely on a subset of criteria or shortcuts, and even when the criteria are available, their influence can be limited by cognitive load, time pressure, or the case’s salience.
    • Experimental limitations:
    • Other studies show consensus information sometimes has limited impact on helping behavior (e.g., a confederate manipulated perceived consensus about how many people help someone having a seizure but found no effect on actual helping).

  • Covariation model critiques highlight: attribution under real-world conditions often ignores some criteria; people sometimes attribute behavior based on single, salient cues or rely on socially desirable explanations.

  • Distinctiveness, consensus, and consistency illustrated with simple mapping:

    • Distinctiveness (D): is the behavior specific to this target or common across targets? (High D means the behavior is rare for this target across different targets.)
    • Consensus (C): do others behave this way in the same situation? (High C means many others behave similarly.)
    • Consistency (Co): does the person behave this way across time? (High Co means the behavior is stable over time.)
    • Overall attribution emerges from the pattern across D, C, and Co, with different patterns favoring person vs entity vs context explanations.

  • Summary of implications:

    • People often prefer internal explanations for others’ behavior, unless there are strong, clear external constraints.
    • The covariation model provides a structured way to analyze attribution but is criticized for being overly complex for everyday judgments.
    • Real-world evidence shows mixed support for the universality of the model, and context matters (e.g., social desirability, impression management).

  • Self-interested or socially motivated attributions have consequences for prediction, compliance, and social interaction. Interpretation of others’ behavior influences expectations about future actions and social judgments.

  • Transition to related theories:

    • Bem’s self-perception theory, cognitive dissonance, and the broader motivational and social-cognitive processes that shape how we infer both others’ and our own motives.

  • Key formulas and numbers referenced:

    • Percentage in consensus manipulations: 50\% of the time, shocks terminated in the red-light condition; varied by condition in the experiments described. (Descriptive figures used to illustrate causal attributions and impression management concerns.)
    • Covariation criteria and their interaction are described qualitatively rather than as a single numerical formula, but the framework can be summarized as a function of the three cues (D, C, Co) and their high/low configurations across situations.

Self-Perception, Dissonance, and Motivated Reasoning

  • Bem’s self-perception theory: we infer our own motives from our behavior, especially when internal cues are ambiguous.

    • Core idea: similar to attribution to others, we infer our own motives by observing our behavior and the surrounding context.
    • Conditions favoring self-attribution from behavior:
    • Ambiguous internal cues about motives.
    • When behavior is not clearly explained by external constraints.
    • Caveats and critiques:
    • Impression management: people may misreport to present themselves in a favorable light (e.g., claiming behavior was driven by genuine motivation when it was not).
    • In real life, the motive inference may be confounded by the desire to look good or to avoid negative impressions.

  • External vs internal attribution in action: complex interplay between our own behavior and how we interpret it.

  • Cognitive dissonance (Festinger, 1950s): how we maintain consistency between beliefs and actions when they conflict.

    • Classic study: participants perform a boring task for an hour.
    • Afterward, they are asked to tell the next participant that the task is interesting.
    • Conditions:
    • Control: no manipulation of payment or justification.
    • $1 condition: participants are paid $1 to tell the next participant that the task is interesting.
    • $20 condition: participants are paid $20 to tell the next participant that the task is interesting.
    • Findings:
    • The $1 condition led participants to report higher interest in the task than the $20 condition or control.
    • The idea: with only $1, participants experienced insufficient justification for lying and thus changed their attitude to reduce cognitive dissonance, leading to an internal shift toward believing the task was interesting.
    • The $20 condition provided external justification for lying, so less attitude change occurred because there was a clear external reason for lying.
    • Implications and caveats:
    • Impression management and demand characteristics can confound interpretation.
    • The study illustrates internal attitude change as a way to resolve dissonance when external justification is insufficient.

  • Motivation and education interventions:

    • Attribution condition (academic performance): attributing success in mathematics to students’ hard work and ability (internal attributions) was more effective at improving performance than persuasion (telling students to do better) or reinforcement (praise for effort).
    • Key takeaway: people’s performance can be enhanced by fostering adaptive attributions that emphasize controllable internal factors (e.g., effort, strategy).
    • Persuasion as an intervention is often ineffective and can even be counterproductive.
    • Positive reinforcement is helpful but typically less effective than appropriately framed attributional messages.

Motivated Reasoning and Dual-Process Theories

  • Core idea: rationality is sometimes used not to discover truth but to defend preexisting beliefs; information is evaluated through motivated processing.

    • Kahneman’s system 1 vs system 2 framework:
    • System 1: fast, intuitive, emotional, and heuristic-driven processing.
    • System 2: slow, deliberate, analytical, and less emotionally driven processing.
    • Implications: when confronted with high-threat or high-risk information, people may default to system 1 processing, leading to emotionally driven judgments and biased evaluation of evidence.
    • Threat and risk framing (e.g., homicide rates, terrorism): emotional cues prime system 1 processing, reinforcing protective identity and us-vs-them thinking.
    • Motivated reasoning and identity protection can lead to selectively accepting or discounting information that aligns with prior beliefs or group memberships.

  • Two influential lines of research on motivated reasoning:

    • Kahan et al. (cognitive reflection task studies): Investigated how people evaluate expertise and evidence about contentious issues (e.g., climate change) based on political ideology.
    • Measures: cognitive reflection task (questions with an intuitive but incorrect answer requiring reflection) and ideology dimensions.
    • Two-dimensional ideological model: Egalitarian-Communitarian vs Hierarchical-Individualist.
      • EC: egalitarian concerns and community welfare; HI: hierarchy and individual rights.
      • People tend to distrust or distrustfully rate experts whose positions conflict with their own ideological positions.
    • Findings: credibility of experts is strongly moderated by the alignment between the expert’s expressed position and the evaluator’s ideology; consistency boosts perceived credibility for those aligned with the evaluator’s worldview.
    • Nyhan and colleagues on correcting misinformation:
    • Two corrective conditions across issues (e.g., Bush and WMDs; stem cell research): provide corrective information counter to mistaken beliefs.
    • Findings: corrections often fail to change beliefs; in some cases, corrections backfire or strengthen the original belief, especially among those with strong prior ideology.
    • The backfire effect: counterarguments can cause people to cling more strongly to their initial beliefs when faced with information challenging those beliefs.
    • Moderating factors: political allegiance and the perceived credibility of the correcting information influence whether beliefs are updated.
    • General takeaway: even when presented with factual corrections, motivated reasoning often leads to selective acceptance, with ideology shaping how information is interpreted and which evidence is considered credible.

  • Implications for public discourse and policy:

    • Facts alone are often insufficient to change beliefs; messaging and framing matter, and source credibility matters more when it aligns with the audience’s preexisting worldview.
    • The backfire effect complicates attempts to debunk myths in politically polarized environments.

  • Additional discussion points:

    • The debate about whether truth-seeking rationality has declined in importance relative to post-hoc rationalization of group beliefs.
    • The idea that in some contexts, belief justification may have historically served group cohesion or survival even at the expense of objective accuracy.

Self, Identity, and Social Belonging

  • Personal vs social identity

    • Self-concept comprises personal identity (traits, quirks) and social identity (group memberships such as gender, student status, ethnicity).
    • Social identity contributes to how we define ourselves and how we are perceived within groups.
    • The self is dynamic and context-dependent: different situations can prime different aspects of self.

  • Language and self-concept: a dynamic prime of identity

    • TseMei/Tru research with bilingual Hong Kong students:
    • Task: describe “Who am I?” in two languages: Chinese vs English.
    • Findings:
      • In English, descriptions tended to emphasize personal traits and individual differences (more individualistic descriptors).
      • In Chinese, descriptions emphasized group membership and interdependence (more collectivistic descriptors).
    • Interpretation: language can prime different aspects of social identity; identity is fluid and context-sensitive.

  • Self-esteem and its measurement

    • Self-esteem reflects how positively or negatively we evaluate ourselves; can be global or domain-specific (e.g., academic self-esteem).
    • Relations with outcomes:
    • Some longitudinal data suggest self-esteem relates to academic performance cross-sectionally, but not consistently over time; the relationship can be bidirectional or confounded by other factors.
    • Measurement approaches:
    • Self-report scales (e.g., direct items about self-worth). Pros: simple; Cons: susceptible to social desirability and response biases; confronting items may suppress honesty.
    • Implicit measures (I/Me priming): tests reaction times to self-relevant positive traits (e.g., “I am smart, warm”). Probes implicit associations and can circumvent some social desirability biases.
    • Example items and concerns:
    • Direct self-esteem item: "All in all, I am inclined to think that I am a failure" (confronting, may induce defensiveness).
    • Positive items: "I take a positive attitude towards myself."
    • The World No Youth Study example (contextual note): cross-sectional link between self-esteem and academic performance observed; longitudinal follow-up showed limited predictive validity, underscoring bidirectionality and context sensitivity.

  • Self-concept and time perception

    • People tend to view their past selves as more stable and trait-like, while their present selves are viewed as more situational and malleable (actor-observer bias extended across time).
    • Pronin & Ross (time-based self-perception): comparing self at present vs five years ago;
    • Present self judged as more influenced by situations, past self judged as more fixed or trait-like.
    • Age differences: patterns hold across younger and older groups; both show this temporal self-distinction.

  • Self and social belongingness: hierarchical concept

    • Social Identity Theory: the self is comprised of two levels of identity:
    • Personal identity (unique traits and dispositions).
    • Social identity (group memberships and the importance of belonging to groups).
    • The sense of self can be hierarchical: one’s social identities can take on greater or lesser prominence depending on context.
    • Examples of situational priming of identity:
    • Being a student is more salient in a university lecture setting than at home.

  • Important methodological note: time and identity research emphasizes dynamic, context-dependent self-definition rather than a static, unchanging self.

  • Practical implications:

    • Interventions aimed at improving performance or well-being may need to consider both personal and social identity components.
    • Language, culture, and context can shape self-perception and motivation, affecting behavior in educational and work settings.

  • Summary of key methods for studying self and identity:

    • Self-descriptions in different languages to reveal cultural/collective emphasis.
    • Actor-observer bias paradigms applied across time.
    • Implicit measures of self-esteem to capture automatic associations rather than self-reported beliefs.
    • Longitudinal designs to distinguish cause and effect between self-esteem and performance.

Concluding Notes: Implications for Research and Practice

  • Attribution theories (including Kelley’s covariation model) provide a structured lens for interpreting others’ behavior, but real-world judgments are often simplified, inconsistent, or influenced by social desirability and cognitive load.

  • Bem’s self-perception theory highlights a potential route to understanding self-motivations, especially under ambiguity; but self-presentation concerns can complicate interpretations.

  • Cognitive dissonance and subsequent attitude change illustrate how people rationalize actions inconsistent with their beliefs; but results can be sensitive to the amount of external justification and to demand characteristics.

  • Motivated reasoning emphasizes that rational processing is frequently oriented toward sustaining beliefs and identity, not simply seeking truth; dual-process theories offer a framework to understand when people rely on fast, emotional judgments versus slow, analytical assessment.

  • Corrective information can fail to shift beliefs and may even backfire for certain groups, underscoring the need for careful messaging and understanding of audience ideology and identity.

  • The self is fundamentally social and context-dependent; language, culture, and group belonging shape self-concept and motivation, with implications for education, leadership, and intergroup relations.

  • Ethical and practical implications:

    • Be mindful of how contextual cues (language, framing, authority cues) might prime different identities and influence judgments.
    • When designing interventions (education, health, climate communication), consider attribution styles, audience ideology, and potential backfire effects.
    • In research, account for impression management and social desirability biases, especially in self-report measures.

  • Final takeaway: human reasoning is a product of complex, often competing forces—dispositional attributions, situational cues, self-perception, cognitive biases, and motivated reasoning—shaping how we understand others, ourselves, and the world around us.