A Model for Generative CAI and Concept Selection.

This paper describes a model for generative computer-assisted instruction. This model has served as the basis for the design of a CAI system used to teach problem-solving in an introductory course in digital systems design. The system individualizes the instruction which each student receives in accordance with its record of his past performance. In addition, a heuristic technique is used to determine the best path for each student through the tree of course concepts. The refinement of this method of concept selection is described. An evaluation of the GCAI system and results of classroom usage are also presented.