Inferring Learning Strategies from Students’ Errors

Student models in an Intelligent Tutoring System (ITS) aim at representing both correct and incorrect knowledge of the student in a particular domain of learning. This data serves as the basis upon which the Tutor can dispense remedial feedback. This paper presents a framework in which learners’ errors are diagnosed, classified and linked to cognitive processes associated with learning. The cognitive processes we identify are those of generalization, misattribution and ignorance. Our system applies to the domain of Second Language Acquisition and our approach uses Chomsky’s linguistic theory of Principles and Parameters. We particularly concentrate on the acquisition of the Phrase Structure component.