The Bias-and-Expertise Model: A Bayesian Network Model of Political Source Characteristics.

Area of Science:

• Cognitive Science
• Political Communication
• Computational Social Science

Background:

• Source credibility influences political polarization and misinformation.
• Existing Bayesian models of information integration lack a bias component.
• Political sources often exhibit bias, affecting information credibility.

Purpose of the Study:

• To develop a novel Bayesian Network model incorporating source bias and expertise.
• To predict how individuals update beliefs and perceptions of sources.
• To validate the model's predictive power in understanding information processing.

Main Methods:

• Developed a novel Bayesian Network model integrating source bias and expertise.
• Conducted two studies to test the model's predictions.
• Utilized testimony to assess belief and perception updates.

Main Results:

• The model successfully predicted how individuals update beliefs and perceptions of bias and expertise.
• Demonstrated the model's validity in a preregistered replication and extension study.
• Confirmed the integration of bias and expertise influences information processing.

Conclusions:

• The novel Bayesian Network model effectively captures how people process information from biased and expert sources.
• This model offers a more comprehensive understanding of source credibility in political contexts.
• Findings have implications for combating misinformation and political polarization.