Perceived supernatural support heightens battle confidence: a knife combat field study

ABSTRACT

Religiosity has been historically linked with propensities for both antisocial aggression and prosocial bravery, and prior research employing indirect measures affirms that envisioning the support of supernatural agents promotes confidence in engaging in violent conflict. Here, we provide the first test of this hypothesis within a realistic combat paradigm (i.e., simulated knife fighting). We primed the presence of supernatural aid and collected measures of trait religiosity as well as political orientation, which typically covaries with religiosity and has been similarly linked with battle confidence in prior research. Consistent with predictions, participants evinced greater confidence in their own performance in an imminent knife battle following a guided visualization exercise analogous to prayer summoning supernatural aid. Moreover, individual differences in trait religiosity comparably predicted battle confidence, an effect that was not accounted for by covarying differences in political orientation, which also predicted battle confidence if analyzed independently of religiosity. Against expectations, we observed limited effects of the visualization manipulation or religiosity, and no effect of political orientation, on coalitional confidence in the groups participants fought alongside. These results, derived from unusually valid methods, are discussed as they extend prior research on the confidence-enhancing effects of perceived supernatural support.

KEYWORDS:

• Religion
• risk-taking
• violence
• threat
• aggression
• political orientation

Acknowledgments

We thank Paige McNorvell for organizing data collection, and thank our research assistants Shahe Dishakjian, Evyn Mirasol, Taylor Lowe, Angad Thakral, and Sabrina Belen. We also thank Adam Sparks and Dan Fessler for helpful feedback.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 It should be noted that, in a series of cross-cultural studies, Ginges, Hansen, and Norenzayan (2009) find that participation in religious community activities correlates with support for suicide attacks, whereas religiosity as indexed by prayer frequency does not, suggesting that coalitional affiliation associated with religiosity may drive support for at least some forms of coalitional aggression to a greater extent than does perception of access to supernatural support related to religious beliefs.

2 Follow-up tests confirmed that the overall pattern of findings relating the supernatural visualization, religiosity, and political orientation to personal battle confidence remains statistically significant in models including the entire raw, unfiltered sample.

3 Data collection occurred over two weekends. On the second weekend (N = 44), a new item exploring rematch confidence was added immediately after each battle: “If there were a rematch, how confident are you that you would perform better than members of the opposing squad?” (1 = Not at all well; 9 = Extremely well). Analyses and discussion of this exploratory measure are provided in the Supplemental Online Material.

4 The model revealed a significant interaction between the visualization condition and the order of presentation (see Table 1). The effect of the Supernatural visualization was more pronounced in the first battle than in the second, in a pattern which appears to reflect a tendency for participants to feel greater confidence before the second battle than they did before the first (see SOM Table S1). The increase in battle confidence before the second battle plausibly owes to a practice/habituation effect which, while potentially interesting with regard to the psychology of combat, appears orthogonal to our hypotheses regarding the influence of perceived supernatural support. Importantly, the supernatural visualization remains a significant predictor of personal battle confidence when removing the order*visualization interaction term (or any of the other covariates) from the model.

Additional information

Funding

This work was supported by the US Air Force Office of Scientific Research under Award #FA9550-115-1-0469.