Like-Minded Narratives: Unraveling Truth-Learning in Political Elections

• In Event: Information Transmission

Abstract

The notion of like-mindedness in politics is usually associated with political polarization, where individuals tend to align with others who share similar beliefs and values. Following Eliaz & Spiegler (2020), this paper extends this concept by emphasizing the similarity between narratives to depict individuals who share analogous causal models as represented by Directed Acyclic Graphs (DAGs). In political elections, voters often encounter divergent opinions on policy implementation, stemming from conflicting narratives about political reality. We formalize the voter scrutiny process, wherein individuals critically assess the narratives presented by candidates in comparison to their own existing narratives.

The voter scrutiny process outlined in the model unfolds as follows: In the electoral scenario, two candidates vie for support, alongside a dedicated public representative acting on behalf of the people. The public has a prior narrative, known to the public but not to the candidates. Simultaneously, the candidates know a true narrative undisclosed to the public. There exists a certain probability for candidates to discover the public's prior narrative, and reciprocally, the public can learn the true narrative with a certain probability. Upon acquiring knowledge of the true narrative, the public adjusts their prior narrative accordingly. The public's selection of candidates bases both on the expected utility derived from the candidates' announced narratives, and the like-mindedness gauged by the distance between the candidates’ and the public’s own narratives represented by DAGs. Essentially, the public’s decision-making process is built upon the behavioral assumption that combines “like-mindedness” and “wishful thinking,” integrating the alignment of narratives with the anticipated utility induced by those narratives. As the public evaluates their subjective anticipatory utility, the candidate’s problem is to choose narrative-policy pair to decide whether to tailor their narratives for pandering or to adhere to a truthful narrative.

The findings highlight two significant insights: Firstly, when the public actively engages in the effort to learn the truth, candidates are more inclined to align with the true narrative. This suggests a positive connection between the public’s commitment to truth-learning and the candidates’ tendency to provide accurate information. Conversely, when the public is slow or lax in their pursuit of the truth, candidates are more likely to cater to public sentiments or preferences. This indicates that candidates’ narratives may be influenced by the perceived preferences or biases of the public, especially when there is a delay or reluctance in the public’s truth-learning process. In the face of diverse narratives prevalent among different social groups or political parties, the dedicated pursuit of truth becomes a unifying force. Public engagement in earnest efforts to understand facts diminishes divisive impacts. This commitment serves as a bridge, fostering shared understanding and alleviating polarization, thereby promoting cohesive and well-informed public discourse. These findings underscore the dynamic relationship between the public’s truth-learning behavior and the responsiveness of political candidates, shedding light on how the intensity of truth-learning efforts can influence the narratives presented by candidates in political elections, and how to create a more inclusive and less polarized political environment.

Author

• Ke Liang, Princeton University