Motivated reasoning or biased prior impressions: Updating trust towards partisan sources based on evidence

A prominent explanation for the growing trend of belief polarization in political issues is that people engage in motivational reasoning to affirm their ideology and to protect their partisan identities. An alternative explanation is that people seek the truth but use partisanship as an a-priori criterion of what constitutes a credible source of political information. In two experiments, we test these competing explanations by employing a dynamic setting where accuracy-motivated individuals can gradually learn which partisan source is credible based on evidence. In each round, Democrats and Republicans are asked to update their prior estimations about a neutral or a politically relevant topic based on information they receive from another partisan (i.e., ingroup or outgroup source). Both partisan groups initially trusted ingroup sources more than outgroup sources, a pattern that was stronger among partisans with high affective polarization. However, across rounds, this identity bias declined, or even changed direction, as supporters of both groups gradually learned to trust reliable sources more than unreliable sources irrespective of source’s partisanship. Importantly, the partisanship of the sources and the political relevance of the shared information did not affect the learning rate. In contrast, the strength of evidence regarding sources’ actual reliability influenced the learning rate. These findings demonstrate that partisans do not exhibit a persistent motivation to protect their identities. Although they exhibit an initial bias in trusting partisan sources, they are motivated by accuracy and, in the presence of strong evidence, they gradually learn to trust credible sources irrespective of partisanship.