Historically, efforts at user modelling in educational systems have tended to employ knowledge representations in which symbolic (or "linguistic") cognition is emphasized, and in which spatial/visual cognition is underrepresented. In this paper, we describe our progress in developing user models for an explicitly "spatial" educational application named HyperGami, in which students design (and construct, out of paper) an endless variety of three-dimensional polyhedra. This paper gives a brief description of the HyperGami system; discusses our observations (and experimental results) in understanding what makes certain polyhedral shapes difficult or easy to visualize; and describes the ideas through which we plan to augment HyperGami with user models that could eventually form the computational basis for "intelligent spatial critics." Important as these "linguistic" representations of knowledge are, there is an increasing body of evidence to sugges...
Michael Eisenberg, Ann Nishioka, M. E. Schreiner