Kids are fans of Bayes Theorem

Last time I gave a small (and humble) introduction to Bayes' Theorem. I hope it got you all interested in this small little formula, because I think it is an important tool for any skeptic endeavour. And to be honest, it was just meant as an introductory paragraph or two for the scientific study I will discuss today. Guess I got a little carried away ...

It's important to note that we already use Bayes' Theorem implicitly. That's why a couple of researchers from Berkeley and Edinburgh devised an interesting study, as reported on Futurity (research itself is behind a paywall). They devised a couple of interesting tests, where unusual combinations of blocks put in a toy machine would "make it go" (music and light), and then put preschoolers and undergrads in front of them to test it. The results were comparable to results from the Marshmelllow Challenge: kids did it better than adults.

https://www.youtube.com/watch?v=bHQ0DemKcEA

Why? Well, because of Bayes. Not explicitly of course, no toddler started writing out his priors and consequents. But young children picked up on new evidence that showed that the toy machine could "go" because of combining different types of blocks. Adults meanwhilestayed longer with prior assumptions, like that only one specific type would make the toy machine "work". Not only it shows that children are capable of abstract reasoning and like to explore various unconventional hypothesis. The "results are consistent with the predictions of a hierarchical Bayesian model", was the conclusion from the article.

Indeed it was. As we saw last time, evidence and evidence evaluation (how typical are the results if a hypothesis would be true or false) form an important part of the formula. The kids in the study showed that they could act on evidence, and adapt their reasoning. As I noted last time, that is exactly why the Bayes' Theorem is such an interesting tool for skeptics. And I also speculate that if Christmas would be repeated every month, kids would pick up within a year if Santa really exists or not ...