The role of political devotion in sharing partisan misinformation and resistance to fact-checking

Abstract

Online misinformation poses a significant threat to global challenges such as pandemics and climate change. To understand what drives individuals to share misinformation, we conducted two pre-registered experiments with conservatives and far-right supporters in the US and Spain (N = 1,609) and a neuroimaging study with far-right supporters in Spain (N = 36). Individuals who felt their personal identity was fused with their political group were more likely to share misinformation, especially around sacred moral issues (e.g., immigration and nationalism). Far-right supporters showed increased activity in brain regions associated with theory of mind in response to posts with sacred values, highlighting the social nature of misinformation sharing. Analytical thinking was unrelated to misinformation sharing around sacred values (vs. non-sacred values) and fact-checks had little or no effect, especially among hyper-partisans. We provide evidence that political devotion plays a key role in misinformation sharing and discuss practical implications of our findings.

Understanding and Combating Misinformation Across 16 Countries on Six Continents

“…A third theoretical perspective rooted in political psychology implicates ideology in susceptibility to falsehoods (31)(32)(33)(34)(35). We find consistent associations with participants' responses to two items from the World Values Survey regarding government policies: Valuing democracy was associated with higher truth discernment in all countries (meta-analytic b=0.045, z=8.80, p<0.001); and fiscal conservatismas measured by the endorsement of individual responsibility over government supportwas associated with worse truth discernment in most countries (meta-analytic b=-0.015, z=4.63, p<0.001).…”

The Misleading count: An identity-based intervention to counter partisan misinformation sharing

“…However, recent studies and a meta-analysis suggest that the effect of accuracy nudges may be relatively weak (9), especially among conservatives, Republicans, and far-right supporters (10,11). This is in line with the Identity-Based Model of Political Belief (12), which posits that individuals are more likely to believe and share misinformation when their partisan motives outweigh accuracy concerns (13).…”

“…Extreme partisanship. Interventions to counter misinformation are often less effective when partisan incentives outweigh accuracy concerns, for instance, when misinformation is framed in terms of group-relevant sacred values, and for individuals who highly identify with the group (11). Thus, we tested if the Misleading count was also less effective in these conditions.…”

“…Indeed, laypeople are relatively good at distinguishing low-quality news content (25), raising the possibility of crowdsourcing accuracy judgements. This strategy could be particularly useful for misinformation on sacred moral values, i.e., rigid stances on moral issues that are resistant to trade-offs (e.g., attitudes towards immigration, and universal healthcare, see (26,27)), which people are more likely to share as compared to misinformation on non-sacred issues (e.g., infrastructure, see (11)). Crowdsourcing only from the in-group may also contribute to correcting inaccurate perceptions of in-group consensus over particular issues, which could help reduce misperceived polarization (see (28)).…”

Knowing Well, Being Well: well-being born of understanding: The Urgent Need for Coordinated and Comprehensive Efforts to Combat Misinformation

“…39 Second, accuracy primes can be moderated by partisanship such that they appear to be ineffective for people who strongly identify as conservative, at least in the United States. [38][39][40] Third, there is some indication that the effect conferred by accuracy primes occurs predominantly in the first few seconds after exposure (the treatment effect dissipated after participants rated about 7 headlines), although this initial finding requires further investigation. 39 Finally, a recent study into the performance of "nudge" interventions (which accuracy primes/nudges fall under) in controlled laboratory experiments compared to "in the wild", found that the take-up effect of such interventions in the lab is about 6.2 times larger than in the field (8.7% vs 1.4% 41 ).…”