https://orcid.org/0000-0002-2343-2776
![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
The motivational account of religion—that belief fulfills a psychological need—has been both historically popular and empirically supported. It is also potentially informative about religious interpretations of negative events (e.g., that they are part of God’s benevolent plan). Yet, it is not clear what cognitive mechanism(s) might link negative events to religious belief, and what motivates belief in gods that cause these events. We proposed that a repressive coping style is an important factor because it involves an interpretive bias that both downplays threat and also emphasizes benefits afforded by ambivalent stimuli (including god concepts), potentially facilitating the construction of supernatural concepts that are positive and relevant enough to attract and sustain belief. In the current research, across three studies, we found that repressors were more likely to be religious, and more likely to interpret familiar, unfamiliar, and experimenter-created gods positively. Repression partially explained the positive views of gods held by religious individuals, and this relationship strengthened when distressing events were associated with gods, suggesting that repression helps to sustain belief. Furthermore, the relationship between repression and belief in familiar and unfamiliar gods was partially explained by god positivity, supporting a motivated reasoning account of belief formation.
Acknowledgments
The authors thank the editor, two anonymous reviewers, and researchers in the University of Otago’s Social Cognition Laboratory for helpful feedback.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The data that support the findings of this research are publicly available in the Open Science Framework at http://doi.org/10.17605/OSF.IO/94Z7X.
Notes
1 Motivated reasoning is also facilitated by traits sometimes ascribed to gods, such as ambiguity, abstraction, and metaphor (Dunning et al., Citation1989; Kunda, Citation1990; Maass et al., Citation1989; Suls et al., Citation2002; Thibodeau & Boroditsky, Citation2011)
2 For each of our studies, we recruited more than the required number of participants in anticipation of exclusions (see below), but no participants were added after analysis.
3 See Dance (Citation2015) for a discussion about the validity and reliability of internet studies and of Mechanical Turk. Although participants recruited online are more diverse in their age and geography than undergraduate or local participant pools, there may be greater potential for inattention, distraction, cheating, and familiarity with psychological research. To partially address these concerns in our studies, we employed quality controls (participants had a >95% past approval rate for completed tasks, were excluded if they took part in similar studies with our lab and were asked to not discuss the survey with anyone), enforced a minimum reading time for some tasks, and included qualitative and quantitative attention checks (e.g., time-based exclusion criteria; see below).
4 Continuous measures of repression have been employed in other research to test its correlational associations (e.g., Szentagotai & Onea, Citation2007). Reviews comparing continuous measures of repression with the more common method of treating repressors as a group based on arbitrary cut-offs for anxiety and social desirability find that the two methods produce comparable findings (Derakshan et al., Citation2007; Myers, Citation2010).
5 Neutral (control) priming was not included because our hypotheses principally concerned the effect of individual differences in repression on god positivity when gods are associated with negative events. We believe this is a chronic association, and the priming procedure is intended only to boost its signal.
6 Items were randomly ordered within each of the scales used in our studies.
7 Inattention shown by a small minority of participants can be a problem for internet studies (Dance, Citation2015), as evidenced by implausibly fast response times (e.g., one participant completed the STAI scale in 6.1 s; another completed the VOG scale in 5.3 s). Although a large majority of participants gave the materials sufficient attention, it would reduce data quality to include participants who did not read them. These time constraints were employed on the basis that it would not be possible to read and respond to the materials in less time. For consistency, the same time constraints were applied in our later studies. Correlation matrices that apply all exclusions at once (i.e., a stable sample) can be found in the supplemental materials, although the results are qualitatively unchanged.
8 The mean time to complete these tasks plus two standard deviations was 615 s.
9 We define “religious” or “religiosity” as identifying with a religion in demographic questions, and “nonreligious” as not. All participants coded as “other” (e.g., spiritual) were not included in any religiosity analyses in any of our studies. We considered this binary measure of religious belief and practice to be concise and routine enough to not arouse suspicion about the purpose of the study or to explicitly prime god concepts.
10 All confidence intervals are at the 95% level.
11 Direct and indirect effects are on a log-odds metric. Bootstrap 95% confidence intervals were based on 20,000 samples.
12 This is conceptually non-causal because we measured prior religiosity, but it determines whether our causal model could, in principle, be supported.
13 To limit the number of conditions, folk psychology and ambiguity were not manipulated, although each stimulus received one folk psychology violation.
14 For consistency, we applied the same a priori power analysis to each of our studies.
15 The three-way interaction was larger, r = -.23, when two-way interactions were excluded.
16 English language skill was tested with the training task, with exclusions made prior to analysis.
17 Gender showed no association with repression. Age was consistently associated with repression (as expected; Erskine et al., Citation2016) but did not interact with repression to predict god positivity in any of our studies.
18 Physiological anxiety (that contradicts self-reports) would help to identify individuals with repressive tendencies. Nevertheless, the present operationalism is well-supported in the literature (Myers, Citation2010) and, across the present set of studies, neither trait anxiety nor social desirability (alone) were consistently associated with god positivity (see Supplemental Materials), supporting the construct validity of the combined measure.