I'm a big fan of Bayesian reasoning, and I like the idea of applying it to human intelligence. Usually it is in terms of failures of reasoning, compared with the gold-standard of Bayes. That's why I like this post from Andrew Gelman's blog about this post from Slate where they present evidence that even very young children think in a Bayesian way. The discussion centers around contrasting children and adults in terms of the priors they have when approaching a new problem. Gelman makes the point that the failures in adults can also occur because out internal models tend to be discrete and the real-world is continious. It is very interesting to see how these sorts of mappings from theory to reality are made. In general, how do we infer what internal representations of models we are actually employing?