Commentary on Kurt VanLehn's "The Behavior of Tutoring Systems"

The paper by Kurt VanLehn has been designed for two audiences: one who does not know very much about ITSs and the other who does. The paper has much to offer both audiences. For the former it gives an indication of the educational performance and complexity that ITSs are now able to offer in a way that focuses on the pedagogy and not on the underlying mechanisms. For the latter audience it offers a way of thinking about the nature of ITSs that helps to unify the field and demonstrates common themes across a range of research programmes. The paper is interesting and useful to read and makes many wise points about ITSs and their development. This commentary on the paper is divided into two parts. The first deals with the basic ontology set out in VanLehn's paper and offers some comments on it. The second goes through the paper sequentially and makes some comments on individual sections and subsections (but not all of them).