Constructing Quantitative Problem Representations on the Basis of Qualitative Reasoning

Abstract

We describe the instructionally and psychologically motivated cognitive simulation model Sepia in which qualitative and quantitative problem representations in physics are conceptualized as complementary representations. The main emphasis is put on how qualitative problem representations can be taken advantage of to facilitate the construction of quantitative problem representations. The application domain is classical mechanics. Sepia has been implemented as a knowledge‐based system by means of an equation‐based representation language. Two different mechanisms are employed in the model to coordinate qualitative and quantitative problem representations. On the basis of these mechanisms not only can it be demonstrated how qualitative and quantitative problem representations can be coordinated to solve physics problems successfully and efficiently, but it can also be predicted how misconceptions come into play if problems are to be solved which ask for a precise quantitative solution. These predictions are tested empirically.