On Distributed Cognition While Designing an AI System for Adapted Learning

Abstract

When analyzing learning, focus has traditionally been on the teacher, but has in the recent decades slightly moved toward the learner. This is also reflected when supporting systems, both computer-based and more practical equipment, has been introduced. Seeing learning as an integration of both an internal psychological process of acquisition and elaboration, and an external interaction process between the learner and the rest of the learning environment though, we see the necessity of expanding the vision and taking on a more holistic view to include the whole learning environment. Specially, when introducing an AI (artificial intelligence) system for adapting the learning process to an individual learner through machine learning, this AI system should take into account both the learner and the other agents and artifacts being part of this extended learning system. This paper outlines some lessons learned in a process of developing an electronic textbook adapting to a single learner through machine learning, to the process of extracting input from and providing feedback both to the learner, the teacher, the learning institution, and the learning resources provider based on a XAI (explainable artificial intelligence) system while also taking into account characteristics with respect to the learner's peers.