TOWARD USER MODELS FOR INTELLIGENT COMPUTER-ASSISTED-LEARNING - AN EMPIRICAL-STUDY

A methodology was established that made it possible to observe students interacting with a highly unconstrained computer-based learning system. The system was unconstrained in that it was not subject to the limitations of current computer-based technology. Thirty volunteer postgraduate librarianship students learned about PRECIS indexing, using a specially designed learning system that could respond as flexibly and intelligently as a human expert to natural language queries about the subject matter. Interactions were logged and analyzed. The evidence obtained supports the notion that individuals employ distinct strategies when interacting with information sources in a learning context. The strategies observed in the present study display close similarities with styles and strategies previously found by other researchers using more constrained methodologies.