Truth sensitivity and partisan bias in responses to misinformation.

Abstract

Misinformation represents one of the greatest challenges for the functioning of societies in the information age. Drawing on a signal-detection framework, the current research investigated two distinct aspects of misinformation susceptibility: truth sensitivity, conceptualized as accurate discrimination between true and false information, and partisan bias, conceptualized as lower acceptance threshold for ideology-congruent information compared to ideology-incongruent information. Four preregistered experiments (n = 2,423) examined (a) truth sensitivity and partisan bias in veracity judgments and decisions to share information and (b) determinants and correlates of truth sensitivity and partisan bias in responses to misinformation. Although participants were able to distinguish between true and false information to a considerable extent, sharing decisions were largely unaffected by actual information veracity. A strong partisan bias emerged for both veracity judgments and sharing decisions, with partisan bias being unrelated to the overall degree of truth sensitivity. While truth sensitivity increased as a function of cognitive reflection during encoding, partisan bias increased as a function of subjective confidence. Truth sensitivity and partisan bias were both associated with misinformation susceptibility, but partisan bias was a stronger and more reliable predictor of misinformation susceptibility than truth sensitivity. Implications and open questions for future research are discussed.

Understanding belief in political statements using a model-driven experimental approach: a registered report

“…Cognitive reflection did not demonstrate an improvement in truth discernment, as we hypothesised (H1). This outcome contrasts with the consistent evidence reported in several studies that specifically examined this question in the context of belief in fake news 16 , 29 , 86 , 87 . The underlying reasons for this disparity remain uncertain and warrant further investigation.…”

Incrimination Through Innuendo

“…Moreover, target group membership did not moderate question-innuendo effects, although overall impressions did significantly differ across the three target groups in a manner that is consistent with the intended manipulation (i.e., impressions of political outgroup members were significantly less favorable than impressions of political ingroup members and nonpolitical targets). Thus, counter to the strong effect of partisanship in the acceptance of misinformation (e.g., Batailler et al, 2022; Gawronski, 2021; Gawronski et al, 2023), partisanship seems to play a less significant role for question-innuendo effects. In other words, although acceptance of misinformation is much more common for ideology-congruent than ideology-incongruent information, incriminating innuendo in questions seems to shape attitudes and opinions irrespective of whether the innuendo is consistent or inconsistent with the audience’s ideological beliefs.…”

“…While these concerns call for further research on the impact of partisanship on question-innuendo effects, several aspects of the current studies and prior research on related questions may help to address potential concerns about counteractive method-related factors. First, using the same categorical approach to recruit Prolific workers who identify as either Democrat or Republican, prior work has found extremely large effects of partisanship in truth judgments of political misinformation (e.g., Gawronski et al, 2023). Because this work also relied on evaluatively unambiguous statements and explicit instructions to judge the truth of the presented statements, lack of ambiguity and presence of explicit instructions fail to explain the nonexistent effect of partisanship on the size of question-innuendo effects in the current studies.…”

“…Theories of partisan identity suggest that people readily accept information that is congruent with their ideological beliefs and dismiss information that is incongruent with their ideological beliefs (Van Bavel & Pereira, 2018). Consistent with this hypothesis, partisanship has been identified as a major factor in responses to political (mis)information, in that people accept false information as true when it is congruent with their ideological beliefs and reject true information as false when it is incongruent with their ideological beliefs (e.g., Batailler et al, 2022;Gawronski, 2021;Gawronski et al, 2023). Applied to question-innuendo effects, these findings suggest that incriminating innuendo in questions may have a stronger impact when the target is a political outgroup member (because people readily accept information that is congruent with their ideological beliefs).…”

Meta-perception and misinformation

“…Partisanship strongly impacts how individuals evaluate misinformation itself [9,10], and the sources of (mis)information [11]. But work isolating the features of misinformation, and how they impact evaluation and downstream attitudes, is sparse.…”