3/52.3.2 A Critical Analysis
Although the theoretical model is quite good, ITS has no real breakthroughs. ITSs suffer from the usual AI problems; it is very hard to store knowledge in an appropriate form, and most systems are combinatorially explosive when trying to predict how the student will react. [SB82].Rosenberg [Ros90] analysed ITSs in a very critical way: ``My readings convinced me that ITS work is characterised by two major methodological flaws. First, ITSs are not well grounded in a model of learning; they seem more motivated by available technology than by educational needs. ... Second, positive claims for ITSs are based on testing that typically is poorly controlled, incompletely reported, inconclusive, and in some cases totally lacking''
Rosenberg also complains that too little protocol analysis is made, saying that the real experts, teachers and students, are not consulted enough.
Just when artificial intelligence in general, and intelligent tutoring systems in particular, will have their final breakthrough is still an open question. The idea of using neural nets instead of rule based systems is an interesting approach which could give ITSs some future success.