VanLehn, K., Freedman, R., Jordan, P., Murray, C., Rosé, C. P., Schulze, K., Shelby, R., Treacy, D., Weinstein, A., & Wintersgill, M. (2000). Fading and deepening: The next steps for Andes and other model-tracing tutors. In Gauthier, Frasson, VanLehn (Eds.), Intelligent Tutoring Systems: 5th International Conference: Vol. 1839. Lecture Notes in Computer Science (pp. 474-483). Springer-Verlag Berlin & Heidelberg GmbH & Co. K.

Model tracing tutors have been quite successful in teaching cognitive skills; however, they still are not as competent as expert human tutors. We propose two ways to improve model tracing tutors and in particular the Andes physics tutor. First, tutors should fade their caffolding. Although most model tracing tutors have scaffolding that needs to be gradually removed (faded), Andes' scaffolding is already "faded", and that causes student modeling difficulties that adversely impact its tutoring. A proposed solution to this problem is presented. Second, tutors should
integrate the knowledge they currently teach with other important knowledge in the task domain in order to promote deeper learning. Several types of deep learning are discussed, and it is argued that natural language processing is necessary for encouraging such learning. A new project,
Atlas, is developing natural language based enhancements to model tracing tutors that are intended to encourage deeper learning.

For a PDF full article version, click here (993KB).