Many people, and American Christians in particular, view God as a benevolent protector. Those who believe in God may therefore expect that they can safely engage in potentially risky activities, secure in the knowledge that God will look out for their best interests and ensure good outcomes. Initial experiments supported this hypothesis, but recent attempts to replicate them failed. This unreliable pattern may reflect a false conceptual hypothesis, or an inadequate method: Both the initial reports and the replication attempts used outdated religious priming methods we now know to be ineffective. The present research aimed to clarify the relationship between thinking about God and risk-taking behaviors. A pilot study (N = 264) showed that American Christians do expect that God will protect them during risky activities; moreover, those who hold this expectation more strongly report greater willingness to take risks. This registered report then used a high-powered preregistered experiment to test whether having participants explicitly think about God's (presumably protective) influence over what happens in their lives increases their willingness to take risks. Results confirmed that American Christian participants were more willing to take non-moral risks when thinking about God, compared to a control condition, d = 0.28, 95% CI [0.12, 0.44], p < .001. This pattern was robust to different exclusion criteria and was consistent across domains of career, recreational, and social risks. By using the most recent and sensitive methods, this study provides a more definitive test of the conceptual hypothesis that thinking about God can influence risk-taking.

许多人，尤其是美国基督徒，将上帝视为仁慈的保护者。因此，那些信仰上帝的人可能期望他们能够安全地从事有潜在风险的活动，并确信上帝会照顾他们的最大利益并确保良好的结果。最初的实验支持了这一假设，但最近的复制尝试失败了。这种不可靠的模式可能反映了错误的概念假设或不充分的方法：最初的报告和复制尝试都使用了过时的宗教启动方法，我们现在知道这些方法是无效的。本研究旨在阐明对上帝的思考与冒险行为之间的关系。一项试点研究（N = 264）表明美国基督徒确实期望上帝会在危险活动中保护他们；此外，那些抱有这种期望的人更愿意承担风险。然后，这份注册报告使用了一项高性能的预先注册实验来测试，让参与者明确思考上帝对他们生活中发生的事情的影响（大概是保护性的）是否会增加他们冒险的意愿。结果证实，与对照条件相比，美国基督徒参与者在思考上帝时更愿意承担非道德风险，d  = 0.28，95% CI [0.12，0.44]，p < .001。这种模式对于不同的排除标准都是稳健的，并且在职业、娱乐和社会风险领域都是一致的。通过使用最新和最敏感的方法，这项研究为“思考上帝可以影响冒险行为”这一概念假设提供了更明确的检验。