Building an Affective Learning Companion

摘要

About a half century ago, the computer became a model, metaphor and modelling tool privileging the cognitive over the affective, and engendering theories in which thinking and learning are viewed as information processing and affect is ignored or marginalised. In the last decade there has been an acceleration in efforts to redress this imbalance, developing technologies that can begin to measure and manage the role of affect, enabling new theories and interventions in which affect and cognition are appropriately integrated with one another. This invited keynote presents a vision for developing an automated learning companion that jointly supports a learner's affective and cognitive needs. In particular, I will describe our efforts at MIT to invent several of the affective technologies to enable such a learning companion. The talk will show examples of the state of the art with respect to affect sensing and recognition and with respect to developing strategies for responding intelligently to learner affect.