The Role of Theory of Mind and Wishful Thinking in Children’s Moralizing Concepts of the Abrahamic God

ABSTRACT

Adults conceptualize God as particularly knowledgeable – more knowledgeable than humans – about moral transgressions. We investigated how younger (4- to 5-year-old) and older (6- to 7-year-old) children view God’s moral knowledge. Cultural narratives in the United States portray God as omniscient, which could lead children growing up in the United States to conclude that God knows their own and others’ behaviors. However, older children are better able to distinguish between different minds, and this ability (theory of mind, or TOM) may predict a tendency for older, versus younger, children to attribute greater knowledge to God. Consistent with the latter possibility, 6- to 7-year-olds viewed God as more knowledgeable of their own and others’ transgressions than did 4- to 5-year-olds. TOM partially mediated this difference. Further, children – particularly 4- to 5-year-olds – conceptualized God as more knowledgeable of others’ transgressions than of their own. Study 2 probed whether 4- to 5- year-olds’ responses were due to wishful thinking (e.g., they did not want God to know their transgressions and therefore reported that God would lack this knowledge). Supporting this prediction, 4- to 5-year-olds attributed greater knowledge to God of their own, versus others’, pro-social acts. The extent to which children attributed knowledge of these acts to God predicted their own propensity to behave pro-socially by sharing with others. This work expands current understanding of religious cognition, particularly its connections with moral judgment and theory of mind.

Acknowledgments

The authors thank Clara Apostolatos, Jenny Barshay, Grace Ding, Jared Fel, Devyani Goel, Lindsay Goolsby, Michael Herman, Brielle Internoscia, Rahil Kamath, Carolina Santiago-Robles, Simran Suresh, Aaliyah Triumph, and Haley Ward for assistance with data collection and coding; James Dunlea, Megan Goldring, and Ayse Payir for their feedback; and the Brooklyn Children’s Museum, Traphagen Elementary School, and Crescent Elementary School for providing testing space. This project was made possible through the support of grant #61080 from the John Templeton Foundation to LH. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the John Templeton Foundation. This work has also been supported (in part) by a Visiting Scholar award from the Russell Sage Foundation to LH. Any opinions expressed are those of the authors alone and should not be construed as representing the opinions of the Foundation.

Disclosure statement

The authors do not have any potential conflict of interest.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Data availability statement

Data for Study 1 are available here: https://doi.org/10.7916/d8-zhw7-8r98. Data for Study 2 are available here: https://doi.org/10.7916/d8-arb7-9z72.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at https://doi.org/10.7916/d8-070h-az45.

Notes

¹ Preliminary analyses compared members of different religious groups in two ways. First, we compared children whose parents reported raising them in any kind of religious background with children whose parents identified their affiliation as non-religious/atheist/agnostic. We made this comparison because children who are raised in a religious tradition may make different judgments about God’s mind than children who are not raised in a religious tradition. Second, we compared children from Protestant, Catholic, and “other Christian” backgrounds with all other children. We made this comparison because Christianity is the dominant religious group in the United States, and members of the dominant group may differ from members of other groups. Here and in Study 2, religious background did not reliably predict children’s attributions of knowledge to God. Further, our pre-registered analyses did not focus on religious background, as this was not the main variable of interest in these studies. Therefore, the results presented below collapse across this variable.

² Similar attributions of knowledge to God as those reported below emerged when including all respondents in analyses. The 95% confidence intervals for the indirect effects in the mediation models included zero when analyzing data from all respondents (attributions of knowledge regarding others’ transgressions: .13 [−.05, .31]; attributions of knowledge regarding one’s own transgressions: .19 [−.02, .42]). In Study 2, only one respondent was excluded, and the reason was that he/she was not in the age group targeted by this study. Therefore, we performed Study 2 analyses only on non-excluded participants.

³ Our pre-registration specified particular comparisons of theoretical interest rather than the omnibus ANOVA, and we report the results of these comparisons here. First, to test whether 6- to 7-year-olds attributed to God a “high” level of knowledge about transgressions, we used a one-sample t-test to compare mean responses across the five “self” transgressions to zero (the scale’s midpoint), t(69) = 7.42, p< .001, Cohen’s d= .89. We used a second one-sample t-test to perform the same analysis on mean responses across the five “other” transgressions, t(69) = 9.71, p< .001, Cohen’s d= 1.16. The fact that both means were significantly above zero indicates that older children did, in fact, attribute “high” levels of moral knowledge to God. Second, we tested whether 4- to 5-year-olds attributed to God less knowledge of others’ transgressions (a) than of their own transgressions and (b) than did 6- to 7-year-olds. To address the first of these possibilities, we conducted a paired-samples t-test comparing 4- to 5-year-olds’ responses to the “self” transgressions versus the “other” transgressions. Although 4- to 5-year-olds did distinguish between their own and others’ transgressions, the difference was in the opposite direction as what we expected: they were more likely to attribute knowledge to God when responding about others’ transgressions compared to their own (t(66) = −3.05, p= .003, Cohen’s d= −0.28). To address the second of these possibilities, we conducted an independent-samples t-test showing that 4- to 5-year-olds were less likely than 6- to 7-year-olds to attribute knowledge of others’ transgressions to God (t(133) = −5.19, p< .001, Cohen’s d= −.90). For completeness, we also investigated whether 4- to 5-year-olds were less likely to attribute knowledge regarding their own transgressions to God, and found that they were (t(135) = −5.83, p< .001, Cohen’s d= −1.00).

⁴ The Real Apparent Emotion Task did not include the same types of comprehension check questions as the other tasks. It did ask participants to recall components of the story, but the ability to demonstrate theory of mind on the main questions of interest did not depend on recalling those details (e.g., a child may understand that someone can display different emotions than the ones they truly feel without being able to readily recall why they want to do that). Therefore, our main analyses included responses to this task regardless of children’s answers to the memory questions. Because this task was the only one that did not measure false beliefs and therefore had a different type of set-up from the other tasks, we also re-ran the mediation analyses using only the three false belief tasks and excluding the Real Apparent Emotion Task. These models showed a significant indirect effect when the dependent variable was God’s knowledge of participants’ own transgressions (.22 [.05, .44]) and did not show a significant indirect effect when the dependent variable was God’s knowledge of others’ transgressions (.12 [−.03, .31]).

⁵ We pre-registered analysis for only the “other” mediation, but we present both here for completeness.

Additional information

Funding

This work was supported by the John Templeton Foundation [61080]; Russell Sage Foundation [Visiting Scholar Award].