How Grounding in Runnable Schemas Contributes to Producing Flexible Scientific Models in Experts and Students

This chapter takes a more global point of view and attempts to explain why grounding models in runnable schemas appeared to be so desirable for both experts and students in the previous chapters. Sections I and II of this book show that for experts, analogical reasoning can be difficult, risky, and time consuming. This raises the question of why the subjects who used analogies went to the trouble to do so. A simple answer might be that it was their strategy of last resort – that they could not think of any other way to make a prediction. However, in the case of S2’s solution to the spring problem in Chapter 6, the subject already has a prediction with a high confidence level at the point where he complains that his “understanding is still quite low” and continues to pursue analogies. He goes on to eventually ground his solution in runnable schemas of twisting and bending. But he did not do this primarily to increase confidence in his solution, which was already quite high. The hypothesis to be developed here is that such subjects were attempting to construct new understandings that were as deep as possible by grounding these attempts in prior understandings. In particular, the subjects generated analogies in order to find relevant anchoring schemas constituting their prior understandings. This paid off when a schema was found that was runnable with conviction and that could be used as a grounding element in the construction of a model for the target situation.