AI to Support Guided Experiential Learning

Abstract

Kolb defined experiential learning as “the process whereby knowledge is created through the transformation of experience” [1], often expressed as learning-by-doing. In the context of this workshop, Guided Experiential Learning (GEL) is a pedagogical framework for learning-by-doing that emphasizes longitudinal skill development and proficiency gained through focused, repetitive practice under real world-like conditions [2]. GEL often requires scaffolding psychomotor, affective, and cognitive skill acquisition across multi-modal learning experiences, including games and simulations delivered using virtual, augmented, and mixed reality based applications. In addition, the skills targeted using a GEL framework are generally developed over time via episodic events with controlled conditions dictated by learner states and learning theory [1, 3].

The complexity of designing and assessing experiential learning in the technology-enabled, data-rich environments in which GEL takes place make GEL an ideal candidate for using AI. AI can assist in the design, delivery, and evaluation of experiential events that contribute to longer-term skill and proficiency objectives and to optimize learning. This workshop addresses the research challenges involved in applying AI to GEL. These include multi-modal data strategies, in which data comes from physical, virtual, and mixed-reality training systems; AI models for estimating and predicting skill acquisition and competency; designing learning experiences and assessments to produce optimal data for AI-based models and that provide data under variations in condition and complexity; and applications of AI to instructional support, feedback and coaching [4].