Corrections of Political Misinformation: No Evidence for an Effect of Partisan Worldview in a U.S. Convenience Sample

Abstract

Misinformation often has a continuing effect on people’s reasoning despite clear correction. One factor assumed to affect post-correction reliance on misinformation is worldview-driven motivated reasoning. For example, a recent study with an Australian undergraduate sample found that when politically-situated misinformation was retracted, political partisanship influenced the effectiveness of the retraction. This worldview effect was asymmetrical, that is, particularly pronounced in politically-conservative participants. However, the evidence regarding such worldview effects (and their symmetry) has been inconsistent. Thus, the present study aimed to extend previous findings by examining a sample of 429 pre-screened U.S. participants supporting either the Democratic or Republican Party. Participants received misinformation suggesting that politicians of either party were more likely to commit embezzlement; this was or was not subsequently retracted, and participants’ inferential reasoning was measured. While political worldview (i.e., partisanship) influenced the extent to which participants relied on the misinformation overall, retractions were equally effective across all conditions. There was no impact of political worldview on retraction effectiveness, let alone evidence of a backfire effect, and thus we did not replicate the asymmetry observed in the Australian-based study. This pattern emerged despite some evidence that Republicans showed a stronger emotional response than Democrats to worldview-incongruent misinformation.

Can you believe it? An investigation into the impact of retraction source credibility on the continued influence effect

“…Results typically show that a direct retraction significantly reduces reliance on the critical information relative to the no-retraction control condition, but does not eliminate the influence down to the no-misinformation baseline (e.g., E c k e r , H o g a n , & L e w a n d o w s k y , 2 0 1 7 ; Ecker, Lewandowsky, & Apai, 2011). Continued influence has also been demonstrated with real-world news (Lewandowsky, Stritzke, Oberauer, & Morales, 2005), common myths (Ferrero, Hardwicke, Konstantinidis, & Vadillo, 2020;Sinclair, Stanley, & Seli, 2019;Swire, Ecker, & Lewandowsky, 2017), political misconceptions (Ecker & Ang, 2019; also see Ecker, Sze, & Andreotta, 2021;Nyhan & Reifler, 2010;Wood & Porter, 2019), with subtle and implicit misinformation (Ecker, Lewandowsky, Chang, & Pillai, 2014;Rich & Zaragoza, 2016), false allegations (Thorson, 2016; but see Ecker & Rodricks, 2020), and when the misinformation is presented initially as a negation that is later reinstated .…”

Sensitivity to misinformation retractions in the continued influence paradigm: Evidence for stability

“…Using both perspectives, researchers have been able to uncover several aspects related to the mechanisms underlying the CIE. For example, it has been shown that memory-related processes at both encoding (e.g., Ecker, Lewandowsky, Swire, et al, 2011) and retrieval (e.g., Swire, Ecker, & Lewandowsky, 2017) can contribute to the CIE. Furthermore, cognitive biases have been shown to influence the CIE through perceived source credibility (e.g., Ecker & Antonio, 2020; Guillory & Geraci, 2013), and political beliefs (e.g., Ecker & Ang, 2019; Lewandowsky et al, 2005; Nyhan & Reifler, 2010; but see Ecker et al, 2021; Swire-Thompson, DeGutis, & Lazer, 2020; Swire-Thompson, Ecker, et al, 2020; Wood & Porter, 2019). However, experiments in both lines of investigation have predominantly manipulated aspects of information presentation to observe their impact on the CIE. For example, studies have manipulated study-test delays and the number of misinformation/retraction repetitions (Ecker et al, 2011; Swire, Ecker, & Lewandowsky, 2017), the explicitness of the misinformation (Rich & Zaragoza, 2016), its emotional valence (Chang et al, 2019; Ecker, Lewandowsky, & Apai, 2011; Trevors & Kendeou, 2020), whether participants were warned about the CIE prior to misinformation presentation (Clayton et al, 2019; Ecker et al, 2010), or the trustworthiness of the source (e.g., Ecker & Antonio, 2020; Guillory & Geraci, 2013; Swire-Thompson, Ecker, et al, 2020).…”