Recognizing and responding to student affect

Abstract

This paper describes the use of wireless sensors to recognize student emotion and the use of pedagogical agents to respond to students with these emotions. Minimally invasive sensor technology has reached such a maturity level that students engaged in classroom work can us sensors while using a computer-based tutor. The sensors, located on each of 25 student's chair, mouse, monitor, and wrist, provide data about posture, movement, grip tension, facially expressed mental states and arousal. This data has demonstrated that intelligent tutoring systems can provide adaptive feedback based on an individual student's affective state. We also describe the evaluation of emotional embodied animated pedagogical agents and their impact on student motivation and achievement. Empirical studies show that students using the agents increased their math value, self-concept and mastery orientation. © 2009 Springer Berlin Heidelberg.