Affect-Sensitive adaptive learning environments

点击翻译按钮获取中文摘要

Research on affect-sensitive adaptive learning environments investigates how to create user-adaptive educational tools that take both cognitive and affective elements into account to tailor the interaction to each student's learning needs. This research field has become increasingly prominent in recent years, due to both evidence of strong correlations between affect and learning, and advances in affect recognition, which make it now feasible to devise interactive tools that can be aware of and respond to a user affective states. This proposal aims to contribute to research on affect-sensitive learning environments by investigating how to close what we will call the affective loop, i.e. that ensemble of three phases that together allow for the principled addition of affective elements to an ALE: (i) affect detection, i.e. modeling/recognizing relevant user's states, (ii) selection of appropriate system responses; (iii) synthesis of the appropriate affective expressions. Of the three phases that form the affective loop, phase (i) and (iii) have seen the most advances in recent years. Still lacking are formal theories on how to use affect in computer-based pedagogical interactions, as well as a formal results on the benefits of affect-sensitive computer-based tutors. The long-term objective of this research is two-fold. We seek to close the affective loop in ALE by defining the theoretical underpinnings and technological components to devise computational models of effective affect-sensitive behaviors for computer-based tutors. We then want to provide empirical evidence of the effectiveness of such tutors.