The Heart Trumps the Head- Desirability Bias in Political Belief Revision

Understanding Political Belief Revision

• Importance: Revision of political beliefs is vital for societal perceptions and decisions.

• Study Overview: A preregistered study with 900 participants tested two prominent theories of belief revision: desirability bias and confirmation bias, particularly relevant in the context of the 2016 U.S. presidential election.

  • Participants expressed their desired candidate to win versus who they believed would win.

  • They were then confronted with evidence either supporting or contradicting their desires/beliefs.

Theories of Belief Revision

• Desirability Bias: Individuals tend to give greater weight to information that aligns with their desires.

  • Asymmetry exists where desirable information is more likely to be incorporated than undesirable.

• Confirmation Bias: Individuals prefer information that confirms their existing beliefs, leading to potential belief polarization.

  • Even when balanced information is presented, confirmation bias manifests in belief updating.

Research Context and Methodology

• Context: The heated political climate during the 2016 elections provided a fertile ground for observing biases:

  • Example: Trump supporters possibly believed Clinton would ultimately win due to polls, creating a conflicting scenario for bias effects.

• Participants: Data collected from 900 U.S. residents (59% female, Mage = 37.89) through Amazon Mechanical Turk.

• Experimental Design: Participants categorized based on who they desired versus believed to win:

  • Filler tasks used to obscure the research purpose.

  • Evidence was manipulated to present polling results supporting either candidate based on their initial belief.

Findings of the Study

• Desirability Bias Observations: Participants updated beliefs more significantly when evidence supported their desired outcomes compared to evidence that didn’t.

• Confirmation Bias Limited Evidence: Little independent confirmation bias was found; participants showed a inclination to disbelieve contradictory information.

• Statistical Results: Significant interactions were observed, particularly:

  • Participants updated beliefs more when receiving desirable vs. undesirable info, illustrating a robust desirability bias.

  • Disconfirmation bias surfaced when participants received conflicting information.

Implications for Theories and Practice

• Theoretical Contributions: The results emphasize that desirability influences belief updating significantly, separate from confirmation effects.

• Practical Relevance: Understanding these biases helps in addressing how information is processed in political contexts.

• Previous Research Correlation: Aligns with past studies noting a propensity for individuals to incorporate favorable information over less favorable, affecting beliefs regarding factual matters.

Robustness and Additional Analysis

• Power Analysis: Adjustments to sample size were made to ensure substantial power for detecting meaningful effect sizes.

• Interactions of Variables: Influences of previous exposure and initial confidence levels were analyzed:

  • High initial confidence constrained belief updating, fostering disconfirmation bias.

• Ideological Differences: Exploration of differences between Trump and Clinton supporters regarding desirability bias.

  • Notable finding of a greater frequency of desirability bias among Trump supporters, likely influenced by identity and motivation.

Conclusion

• Understanding political belief revision can address societal polarization over factual beliefs and implications for democratic functionality.

• Emphasizing desirability bias as a determinant factor in how new information is incorporated allows for reevaluation of beliefs in a politically charged atmosphere.