Considering Student’s Emotions in Computational Educational Systems

This work aims at handling the affectivity in a learning interaction between student and tutor in MACES educational system: a collaborative environment with a multi-agent architecture based on Vygotsky’s Socio-cultural Pedagogical Approach. The student’s emotions are handled by the Mediating Agent of the system which has the goal of motivating and engaging the student to learn as well as promoting a positive mood in the student which is more appropriate to learning. We believe that the student can experience a more positive feeling through the exposure to emotional behaviour and encouragement messages sent by the Mediating Agent. In order to accomplish its function, it should recognise student’s emotions to respond appropriately. Thus, it catches the student's emotions by his observable behaviour, stores this information in an affective model; and applies affective tactics according to these emotions. As these affective tactics can be emotive behaviour, we chose to represent the Mediating Agent as a lifelike character who has a personality and which interacts with the student through messages and emotive animations.