Do Children Attribute False Beliefs to God?

Humans are fallible. This is a fact that most adults understand. God, on the other hand, is not, by definition. This is also a fact that most adults, in most cultures, recognize. An interesting empirical question is whether children's views of the difference between the ability of humans and God to hold false beliefs is like adults, i.e., they recognize that humans can hold false beliefs, while God does not. There has been a great deal of research on children's ability to attribute false beliefs to humans. This research has been motivated by the view that a complete theory of mind requires the understanding that individuals can hold false beliefs¹. Most of it uses the false-belief paradigm that is also used to test for the presence and extent of theory of mind in both children and non-human primates. In developmental research, the prototypical false-belief task involves a puppet show, in which children view a doll² named Sally place an interesting object (e.g., a toy) into a a basket, and then leave the stage. After Sally is gone, a second doll, named Ann, removes the object from the basket, and places it into a box. Children are then asked where Sally will look for the object³.

If they understand that people can have false beliefs, children should answer that Sally will look in the basket, where she placed the objectbefore she left. However, up until about age four, children almost always answer that Sally will look in the box where Ann placed the object after Sally left. This indicates that until age four, children are not able to understand that people act based on both desire and belief. Prior to this, in the first year of life, children recognize the difference between animate and inanimate objects, and understand that people are capable of initiating their own actions, but do not seem to understand that these actions are guided by the desire's of the actor. By about age three, children begin to understand that other individuals have desires, and that their behavior is guided by them, but do not understand that beliefs also determine behaviors.

The interesting question about children's God concepts, then, is how their attributions of true and false beliefs to God resemble or differ from their attributions to humans at different ages. This is the question that Knight, Sousa, Barrett, and Atran attempt to answer in a recent paper. They recognizes two general possibile developmental lines. In the first, children's attributions of belief to Gods and humans resemble each other, even after children begin to attribute false beliefs to humans, and only diverge at some later age. This possibility is derived mainly from the work of Piaget, and the resulting popular belief among researchers that children's God concepts are based, analogically, on their human concepts. The second possible line of development involves separate God and human concepts, so that as soon as children begin to attribute false beliefs to humans, they do not so for God.

To test these different possibilities, Knight et al. used a variant of the false-belief task, which they call the "surprising contents" task. This version involves presenting a child with a box containing a picture of its contents on the outside (e.g., crackers), and then asking them what she thinks is inside. After the child has given the desired answer (the object in the picture), the experimenter opens the box, and shows the child that, in fact, there is something other than the pictured object (e.g., a rock) inside the box. Then, the experimenter shows the child a doll who had not seen what was actually inside the box, and asks the child what the doll thinks is inside. The logic of this task is the same as the task with the Sally and Ann dolls. Children who do not understand that people can hold false beliefs will answer that the doll thinks there is a rock inside the box, while children who do understand this will answer crackers. In their version, Knight et al. referred to the doll as either a human or God, and observed whether children's answers differed in these two conditions.

In a previous study using this version of the false-belief task with U.S. children⁴, God and human concepts diverged at about the same age that they began attributing false beliefs to humans. In other words, at age 4, when children generally begin to attribute false beliefs, children continued to attribute only true beliefs to God. Knight et al replicated this study with Mayan children from the Yucatán peninsula (these children believe in the Catholic God, and thus should have similar God-concepts to those of U.S. children). They found that the Mayan children's God concepts did not differ significantly from their human concepts until much later in development, at about age 7. This seems to be due to the fact that the Mayan children did not begin to attribute false beliefs to humans until about that age.

Thus, while there do appear to be cross-cultural differences in the development of theory of mind, it appears that human and God concepts diverge from each other as soon as children begin to attribute false beliefs to humans, whether that's at age 4 (in U.S. children) or age 7 (in Mayan children). Thus, the popular Piagetian view that children's God concepts are based on their human concepts appears to be wrong. While the data doesn't rule out the possibility that children initially develop their God concepts based on human concepts (e.g., at age 2 or 3), it's clear that God concepts are separate from human concepts from a very early age.

¹ Dennett, D. (1978). Beliefs about beliefs. Behavioral and Brain Sciences, 1, 568-570.
² Results are the same regardless of whether dolls or people are used, and since its easier to get children to pay attention to dolls, developmental researchers usually use dolls. See Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 62, 655-684.
³ This version was first presented in Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function on wrong beliefs in young children’s understanding of deception. Cognition, 13, 103-128.
⁴ Barrett, J. L., Richert, R. A., & Driesenga, A. (2001). God's beliefs versus mother’s: The development of nonhuman agent concepts. Child Development, 72, 50-65.