Affective polarization, (mis)perceived polarization and negative meta-perceptions

Recent studies show that those citizens who perceive greater ideological polarization and/or have exaggerated perceptions of the extent to which their political opponents view them negatively, are more affectively polarized. In other words, there is evidence that both (mis)perceived polarization and negative meta-perceptions fuel interpartisan hostility (Moore-Berg et al., 2020; Druckman et al., 2022). Scholars have, however, suggested that the relationship between these phenomena may be bi-directional or even reciprocal in nature, and the present study aims to explore whether the reverse may be true. Our contribution is two-fold, first we investigate whether affective polarization (AP) bolsters biased perceptions of (ideological) polarization and negative meta-perceptions. Second, we move away from common valanced measures of affective bias (e.g. feeling thermometers), and instead consider the extent to which anger, fear, disgust, and contempt might contribute to (mis)perceived (ideological) polarization on the one hand, and biased meta-perceptions of AP on the other. Our assumption of a bi-directional (and potentially reciprocal) relationship is informed by psychological literature concerning ‘projection’, or more specifically, ‘counter-projection’. Counter-projection – whereby group members assign oppositional and/or negative traits, opinions, values etc. to the outgroup – is particularly likely to occur where groups dislike the other and are politicized in nature (Denning & Hodges, 2021), and thus highly likely to emerge in the context of AP. Moreover, we know that anger as a discrete emotion is particularly relevant, as incidental anger bolsters (mis)perceived polarization (Huber et al., 2015), and likely 'amplifies’ other cognitive processes behind negative meta-perceptions (Fernbach & van Boven, 2022). Given that different negative emotions impact personal judgement, information processing and behaviors differently, it is also reasonable to assume that different emotions may lead to (different) exaggerated perceptions and meta-perceptions. We rely on a unique survey-experiment fielded in nine countries, with stimuli designed to either heighten or dampen existing AP, to test these assumptions. Our results will provide further insight into the role of discrete negative emotions underpinning AP and provide evidence as to whether we should (re)consider the presumed causal link between partisan hostility, (mis)perceived polarization and negative meta-perceptions.