Author manuscript, published in "International Journal of Artificial Intelligence in Education (IJAIED) 12 (2001) 85-99"International Journal of Artificial Intelligence in Education (2001), 12, 85-99Bridging the Gap Between Empirical Data on Open-EndedTutorial Interactions and Computational ModelsJohn Cook, Learning Technology Research Institute, University of North London, 166-220Holloway Road, London N7 8DB, UK.j.cook@unl.ac.ukAbstract. In this paper we present an approach to using empirical data on human teacher-learner interactions to guide the development of a pedagogical agent for supporting musicalcomposition learning. Our approach to bridging the gap between tutorial interaction analysisand computational models, intended for use in learning support systems, is a new one. Wesupport our claim by pointing out that most of the previous work in the area of using humantutors as models has been conducted in domains that are more procedural than the open-endedsubject area investigated here. However, the approach described in this paper seems applicableto most, if not all, domains, whether open-ended or not. In the paper we describe how anempirical study of teacher-learner interactions was linked, specifically modulated, to theconstruction of a pedagogical agent called MetaMuse. Empirically derived state transitionnetworks were used to provide a semi-open, goal-oriented interaction plan. One distinctivefeature of MetaMuse is its use of student input, ...