The Limits of Repetition in the Illusion of Consensus

Humans rely on social agreement to assess the credibility of information, but an illusion of consensus can arise through repeated exposure to the same source (dependent consensus), which can make people overestimate credibility. The current work investigated the limits of this illusion across multiple exposures to claims based on either an independent or dependent consensus. In two studies (N = 1015), participants rated their belief in a claim after exposure to an increasing number of consensual claims within a mock social media environment. Initially, when exposed to only a small number of claims, people were equally persuaded by dependent and independent consensus, but independent consensus had a greater impact on beliefs as exposures accumulated. Bayesian nonlinear modeling showed that independent consensus led to a higher asymptotic level of belief and more rapid belief revision than dependent consensus. These effects were robust over variations in claim valence (Experiment 1) and topic (Experiment 2) and were further supported by the free-text justifications participants provided. Our findings show that the illusion of consensus may be less of a concern than previous research suggested, at least in environments where a high amount of repeated exposure to the same claim is common. They also provide some evidence for a novel mechanism contributing to belief in repeated claims, namely, that repetition of such claims by a secondary source is perceived as a proxy for claim credibility.