ITR/PE: Social Intelligence in Interfaces for Educational Software

University Of Southern California, Los Angeles CA

Investigators

Abstract

This project will develop a social intelligence capability for computers, for use in educational software applications. Social intelligence plays a critical role in human tutoring and teaching. By modeling social intelligence in software it should be possible to create educational software that is aware of the learners' attitudes, that interacts harmoniously and responds to learner needs, that is able to develop rapport with learners, and that uses responsiveness and rapport to influence learner motivation. This could be particularly valuable for learners who have motivational problems such as low self-confidence. The social intelligence model will be realized in an animated pedagogical agent, that uses a combination of speech and nonverbal gestures, that can express emotions and attitudes, and that reacts to and adapts to learners over time. It will respond to a variety of learner inputs, including learner performance on assigned problems, conversational dialog, and eye gaze and facial expressions recognized using computer vision techniques. The model will be applied in the context of the Virtual Factory Teaching System, a simulation-based learning system for teaching industrial engineering concepts. The effectiveness of the model will be tested by comparing learner performance with and without the assistance of a socially intelligent agent.