Stability of network models linking personality to conspiracy mentality before and during the COVID-19 pandemic [Author Accepted Manuscript]

Discussions about potential intervention targets, antecedents, and consequences of conspiracy beliefs often rely on comparing bivariate correlations, which can mask intricate patterns. Therefore, the present study adopts a multivariate network approach to gain nuanced insights into the relationships between personality variables and conspiracy mentality. Established and less-studied correlates of conspiracy mentality (i.e., bedtime procrastination, life satisfaction, locus of control, neuroticism, political cynicism, self-efficacy, and self-esteem) were examined together in network models at the aggregate score level and item level. Notably, network stability was examined across different samples before (N = 403) and during (N = 193) the COVID-19 pandemic. The main findings are: (a) The strength and sign of the relationships were often stable across bivariate and network analyses (e.g., positive relationships between political cynicism and conspiracy mentality); however, there were exceptions, such as an inconsistent link between life satisfaction and conspiracy mentality. (b) While many network relationships and centrality indices were similar before and during the COVID-19 pandemic, some noteworthy exceptions indicate that interventions targeting implausible conspiracy beliefs may benefit from tailoring to external circumstances. (c) Certain influential network elements were identified that could inform future interventions (e.g., increasing politicians' transparency and accountability).