Magalie Ochs and Claude Frasson
It is well known that emotions have a strong influence on a person’s cognitive abilities; a fact that can be used by a tutor to improv e a student’s learning capacities. Conditions are that a tutor knows the current emotional state of the learner as well as the emotional impact of certain actions taken and, finally, how the student’s learning capacities depend on his or her emotions. In this paper, we propose the concept of an emotionally intelligent tutoring system (EITS) and make a step toward its practical realization. We are developing a tool that will allow an intelligent tutoring system to improve a learner’s capacities by actively influencing his or her emotions. We combine CBR methods with graph knowledge representation for modeling and predicting efficiently a learner’s emtoinal reaction to specific situations that can occur during a learning lesson, with the final objective of promoting any particular users learning capacities.