Emotional Intelligence for Computer Tutors

We are developing intelligent tutors that elicit, sense, communicate, measure and respond to students ’ emotion. The vision is to use affect in conjunction with cognition to customize instruction and exploit the potential impact of responding to a student’s affective state. This paper describes software and hardware solutions to redress the cognitive vs. affective imbalance to teaching. We investigate problems of sensing and responding to affective experiences, such as frustration, motivation, and self-confidence. We describe theories and technologies to appropriately integrate affect into intelligent tutors and are investigating how to respond to emotions. 1. Connecting emotion to learning The obvious next frontier in computational instruction is to systematically examine the relationship(s) between affective components in feedback and learning outcome (performance) (Shute 2006). If computer tutors are to interact naturally with humans, they need to recognize emotion and express social competencies. Emotion is completely intertwined with cognition in guiding rational behavior, including memory and decision-making and the human brain is described as a system in which emotion and cognitive functions are inextricably integrated (Cytowic 1989). Emotion has been shown to be more influential tha