Does the Style of Misinformation Condition its Effects? An Experiment in Brazil

Empirical evidence suggests that misinformation alters beliefs (Ecker et al., 2022) and potentially influences downstream attitudes such as outgroup perceptions and political preferences (Tucker et al., 2018). Research on misinformation’s impact on both beliefs and outgroup attitudes however remains limited. Specifically, little is known as to how different styles of outgroup-targeted misinformation may influence beliefs and whether they might in turn influence affective polarization. Distinguishing between two ideal-types of misinformation styles (journalistic and crude styles), we propose that the comparatively more professional-looking journalistic misinformation should be more likely to affect belief in false claims, while the comparatively more extravagant crude misinformation should be better able to induce polar-ization, despite having a lesser effect on beliefs. A survey experiment in Brazil partly confirms these hypotheses. Comparing the effect of exposure to a strong dosage of journalistic versus crude misinformation about president Lula da Silva to a control condition, we do overall find that journalistic misinformation affects beliefs to a larger extent, despite variations in effect sizes across measures and subgroups. Turning to the effect of the two styles on attitudinal and behavioral measures of polarization, we do not find clear, overwhelming support for our hypothesis that crude misinformation polarizes respondents more than journalistic-style misinformation. Results instead show that while misinformation can contribute to polarization, the overall impact is limited, and inconsistent across measures of polarization, regardless of its style. Besides, on the measures of polarization that exposure to misinformation does impact, journalistic misinformation is more effective on non-partisan individuals, while both styles show similar effects on petistas. This highlights that the evidence for polarization effects is weak and underscores the need for further research to fully understand the role of misinformation styles in contributing to polarization, and for more research on the measurement of polarization. These findings contributes to the literature on misinformation by highlighting that the style of misinformation shapes its effects, by contributing to the discussion about the potential downstream effects of misinformation, and by extending this literature to a Global South case (Brazil) where both the formats of misinformation prompts and levels of polarization may be more extreme than elsewhere.