Someone is pulling the strings: hypersensitive agency detection and belief in conspiracy theories

Abstract

We hypothesised that belief in conspiracy theories would be predicted by the general tendency to attribute agency and intentionality where it is unlikely to exist. We further hypothesised that this tendency would explain the relationship between education level and belief in conspiracy theories, where lower levels of education have been found to be associated with higher conspiracy belief. In Study 1 (N = 202) participants were more likely to agree with a range of conspiracy theories if they also tended to attribute intentionality and agency to inanimate objects. As predicted, this relationship accounted for the link between education level and belief in conspiracy theories. We replicated this finding in Study 2 (N = 330), whilst taking into account beliefs in paranormal phenomena. These results suggest that education may undermine the reasoning processes and assumptions that are reflected in conspiracy belief.

Keywords:

• Conspiracy theories
• hypersensitive agency detection
• intentionality bias
• paranormal beliefs
• education

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The current study was part of a larger-scale investigation including measures that form part of a separate project. These were the extent to which participants felt that they themselves would take part in real-world conspiracies if placed in the situation of the alleged conspirators (Douglas & Sutton, 2011), personal need for structure (Thompson, Naccarato & Parker, 1989), desirability of control (Burger & Cooper, 1979), just world beliefs (Rubin & Peplau, 1975), trust (Goertzel, 1994; Yamagishi & Yamagishi, 1994), the dark triad of personality (Jonason & Webster, 2010), and immanent justice reasoning based on ambiguous scenarios (Callan, Sutton, Harvey & Dawtry, 2014). We also included paranormal beliefs in this study (Eckblad & Chapman, 1983) but due to an error in the randomisation of scales, we were not able to analyse the results.

² A meta-analysis showed that the weighted mean correlation between conspiracy belief and education level across Studies 1 and 2 (fixed effects) was significant, r = −.21, 95% CI = [−.29, −.13], Z = 4.93, p < .001 (Lipsey & Wilson, 2001).

³ We measured these variables again in Study 2, and in contrast to Study 1 they were significantly correlated. Note however that a Fisher r-to-z transformation revealed that the coefficients for Study 1 and Study 2 were not significantly different (z = −.161, p = .107). Therefore, the different finding across the two studies could simply be due to variation around conventional p-values.

⁴ As was the case for Study 1, Study 2 was part of a larger-scale investigation including measures that form part of a separate project. These were uniqueness-seeking (Snyder & Fromkin, 1977), preference for intentional causes as attributions (McClure et al., 2007), social capital (Cozzolino, 2011; Welzel, Ingelhard & Deutsch, 2005), just world beliefs (Rubin & Peplau, 1975), and immanent justice reasoning based on ambiguous scenarios (Callan et al., 2014).

⁵ Participants also completed a scale of generic conspiracist ideation (Brotherton, Pickering & French, 2013) where they were asked to rate the extent to which they think that 10 statements are true (e.g., “The government is involved in the murder of innocent citizens and/or well-known public figures, and keeps this a secret”, 1 = definitely not true, 5 = definitely true; α = .95). Brotherton et al. (2013) have argued that measuring conspiracy beliefs via agreement with popular conspiracy theories can be problematic because many conspiracy theories are culture-specific. However, the patterns of results were the same as for the well-known conspiracy theories scale we report so we do not report the results for the Brotherton et al. (2013) scale here.