Reifying Implicit Planning in Geometry: Guidelines for Model-Based Intelligent Tutoring System Design

Kenneth R. Koedinger, John R. Anderson

Abstract

This chapter addresses the problem of how basic cognitive science research can be translated into effective ³cognitive tools² for learning. We present a set of guidelines that have been used to design a second generation Intelligent Tutoring System (ITS) for geometry and are intended more generally to aid the design of other computer or non-computer based learning environments. Similar efforts (Anderson, Boyle, Corbett & Lewis, 1990; Anderson, Boyle, Farrell & Reiser, 1987; Collins, Brown, & Newman, 1989) have focused on how cognitive theories or principles can be generally applied to instructional design in any domain. In contrast, the aim of this chapter is to illustrate how cognitive science methodologies can be used in the instructional design process and we focus on a particular class of domains, ones where instructional innovations may pay off most. While instructional innovations can certainly benefit from the application of general instructional principles, a large share of the instructional benefit often derives from insights (of researchers or teachers) into the particular instructional domain itself. This chapter describes a set of methodologies intended to help instructional designers gain such insights.