In their target article, Norenzayan et al. develop a cultural evolutionary theory of the origins of prosocial religions. The authors argue that through this cultural evolution, beliefs and behaviors that are prosocial toward one's own social group are retained across generations. Specifically, learners attend to cultural information via content-based mechanisms, context-based mechanisms, and credibility enhancing displays (CREDs).

We agree with the authors’ argument that context-based mechanisms, along with content- based mechanisms, should be considered in the transmission of religious beliefs. However, although the authors argue for early transmission of such beliefs and behaviors, they cite surprisingly little developmental research to examine how beliefs develop in individuals. We argue that including a developmental approach alongside a cultural evolutionary approach is critical to understanding this phenomenon. A more thorough inspection of the developmental literature supports some aspects, but not all, of the authors’ argument.

Specifically, research on children's learning from others (cf. “testimony”) indicates that even at a young age, children are able to attend to individuals (“cultural models,” sect. 2.3, para. 2) to understand the world in which they live. Children are sensitive to a wide variety of cues and can shift their judgments depending on the kind of information available to them. For example, pre-primary school children (ages 4–6 years) are receptive to whether the informants are in consensus with one another when deciding whom to trust in learning about unfamiliar objects (Chen et al. Reference Chen, Corriveau and Harris2013; Corriveau et al. Reference Corriveau, Fusaro and Harris2009). Young children are especially attentive to testimony provided by informants belonging to the same social group, such as those who share the same ethnicity (Chen et al. Reference Chen, Corriveau and Harris2013) or accent (Corriveau et al. Reference Corriveau, Kinzler and Harris2013; Kinzler et al. Reference Kinzler, Corriveau and Harris2011).

Finally, experience with a particular social group, such as a religion (e.g., attending church services, Christian schools), can also impact children's judgments (Corriveau et al. Reference Corriveau, Chen and Harris2015; Vaden & Woolley Reference Vaden and Woolley2011; Woolley & Cox Reference Woolley and Cox2007). Thus, although the cultural evolutionary approach the authors propose can explain how testimony from prosocial religions allowed for the expansion of large-scale cooperative societies, the developmental approach is necessary to explain the spread of prosocial religion across generations. Even before children can explicitly identify the degree to which they belong to a social group, they are more inclined to learn from and to socialize with members of that group (Gaither et al. Reference Gaither, Chen, Corriveau, Harris, Ambady and Sommers2014).

What about children's use of CREDs when acquiring prosocial religious beliefs? The authors cite Lanman (Reference Lanman2012) as evidence that children are sensitive to the CRED of their religious parents and are less likely to commit to the same religion if their parents do not appear to be particularly religious (sect. 5.1, para. 3). However, our recent research suggests that CREDs might not play as important a role – at least not for young children (Corriveau et al. Reference Corriveau, Chen and Harris2015). In our research, we found that 5- and 6-year-old children who have any religious experience – either at school, at home, or both – have the ability to suspend disbelief in unexpected phenomena, in contrast to their secular peers.

Children with religious experience were not only more likely to consider characters in religious stories as real, but they were also more likely to consider characters in fantastical stories as real; these results hold even when the fantastical stories were less familiar to them and when these stories explicitly referenced magic. The religious children's decisions regarding the storybook characters did not vary depending on the type (i.e., school, home) or the amount (i.e., whether children experienced religion at home and at school, or just at home or school) of religious experience they received, which suggests that this susceptibility is less dependent on the actual content (or CREDibility) of the religious testimony.

Moreover, our findings are not consistent with the authors’ argument regarding the impact of cultural evolution on our social instincts related to “kinship, reciprocity, status, and reputation” (sect. 5.3, para. 2). Specifically, the authors suggest that cultural evolution co-opted these social instincts to help people of the same religion feel a strong affinity for other members in the religious group (although the authors themselves acknowledge that these social instincts are not unique to religion). In the United States, approximately half of the population consider themselves Protestant Christians (Pew Form on Religion 2008), and cues to Christianity and other religions (e.g., holidays and celebrations in school) can often be found in children's everyday lives. Despite this potential presence of religious cues, the secular group of children in our study (Corriveau et al. Reference Corriveau, Chen and Harris2015) – those who did not attend religious school or come from religious families – did not appear readily inclined toward religiosity, even if they interacted with religious peers on a daily basis in the classroom.

The study of religion and beliefs has become increasingly prevalent in the field of developmental psychology. Recent research has shown that both children and adults readily differentiate religious ideologies from factual and preference-based beliefs (Heiphetz et al. Reference Heiphetz, Spelke and Banaji2013a; Reference Heiphetz, Spelke, Harris and Banaji2013b), although understanding of certain characteristics associated with religiosity – such as the concept of omniscience – appears to deepen as children age (Barrett et al. Reference Barrett, Richert and Driesenga2001; Lane et al. Reference Lane, Wellman and Evans2014). A consideration of developmental changes in children's understanding of beliefs is necessary in understanding how prosocial religions have evolved over time.