Development of a cognitive diagnosis system for intelligent tutoring: An application to transportation problem-solving

The purpose of this study is to design and develop a cognitive diagnosis system for intelligent tutoring. The cognitive diagnosis system proposed in this dissertation is called PSCD (Production System for Cognitive Diagnosis). PSCD is intended for intelligent tutoring with possible relevance to decision aiding. This dissertation is based on two assumptions. First, it is assumed that perspectives and methodologies of cognitive science provide a starting point for integrating decision-making studies and cognitive studies (Simon, 1979). Second it is assumed that integrating theories and building an integrated theory or theories into a cognitive architecture provide opportunities to realize practical applications (Newell, 1990). Basing on these two assumptions, the study attempts to integrate three theories of human information processing and to build the integrated theory into PSCD. Three theories considered in this study are heuristic search, production systems architecture, and constructivism. These three theories act as constraints in setting the design specification of PSCD in the huge space of alternative designs. PSCD that are developed henceforth is applied to a counting task for initial validation, and to a transportation problem solving task which is the main task of this dissertation