I've been reading The Old New Logic, a collection of essays. Frank C. Keil, a psycholologist, has an interesting article on the ways that philosophy provides testable hypotheses to research psychologists:
"...although it might seem that work in the philosophy of mind should have the most direct link to psychological research, in fact work in the philosophy of logic, of science, and metaphysics has often had a more fruitful albeit unintended influence."
As examples, he gives Putnam & Kripke's distinction between artifacts and natural kinds, and Fred Sommers' category-tree theory. Both of these provided grist for research that established that children as young as 5 recognized the distinctions and rules they were putting forth.
The point seems to be that philosophers sometimes give very explicit descriptions of deep rules about how the world works. This affords psychologists the chance to test grown-ups and children - to see whether they implicitly observe these rules.
Learning logic is like that too:
Implicitly old, explicitly new.