Children's Attributions 15
American experimenters. Another problem, now related to the God results, is the anomalous
performance of 6-year-old Maya children. This is less easily explained, but may be due to small
sample size. To fully account for the general delay in performance in relation to humans, and the
6 year olds’ performance in relation to God, further studies are needed.
Now turning to the general discussion of the theoretical positions and their predictions,
we can say that, while our results do not address the question of whether children consider the
mother as a special kind of agent, they do add to the US findings in speaking against the idea that
young children need to use humans as a basis to reason about God, which is the rationale behind
all similarity positions. In this sense, the Maya results go in the direction of the non-similarity
position, thus providing cross-cultural evidence for the perspective advanced by Barrett and
collaborators.
It is important to emphasize that this implies simply that young children do not treat God
and humans in the same way in terms of attribution of beliefs. That children truly understand
God as a different sort of agent, and not just a human with a few strange properties (e.g.
infallible beliefs, ability to make mountains etc.) is difficult to disambiguate. Also, by no means
do the data here support the claim that children’s concepts of God are completely independent of
their understanding of people in general and their parents in particular. For example, Christian
theology teaches about a God who practiced self-anthropomorphization by becoming human in
the form of Jesus of Nazareth.
The present results, however, clearly demonstrate that Yukatek young children, as well as
American young children, do not treat God as merely human. For this reason, this work joins the
growing literature that provides evidence against Piaget’s notion that young children cannot treat
other agents as importantly different from humans. For example, contrary to Piagetian