Using Neural Imaging and Cognitive Modeling to Infer Mental States while Using an Intelligent Tutoring System
Functional magnetic resonance imaging (fMRI) data were collected while students worked with a tutoring system that taught an algebra isomorph. A cognitive model predicted the distribution of solution times from measures of problem complexity. Separately, a linear discriminant analysis used fMRI data to predict whether or not students were engaged in problem solving. A hidden Markov algorithm merged these two sources of information to predict the mental states of students during problem-solving episodes. The algorithm was trained on data from one day of interaction and tested with data from a later day. In terms of predicting what state a student was in during any 2 second period, the algorithm achieved 87% accuracy on the training data and 83% accuracy on the test data. Further, the prediction accuracy using combined cognitive model and fMRI signal showed superadditivity of accuracies when using either cognitive model or fMRI signal alone