See, this is exactly the sort of thing kids should be doing in their very first Science class. Not blindly being filled with derived facts with no explanation as to how they were derived, but rather giving them (simple) problems without any sort of solution method, and having them find possible methods on their own. (And the problems don't have to be from the natural world—they can simply be patterns, like this one is.)

After a few hundred, the test would be to look at all the ways they solved all the problems, and come up with a generalized method for problem-solving. Once they derive the scientific method, they pass—and get to learn all the things previous scientists found out using the method. I imagine it would forever change their attitude about those facts—they'd either be eager to build on them with their own experiments, or to know how they were derived and replicate/refute them :)