Principal Investigator: Vincent Aleven
CoPrincipal Investigator(s): Nikol Rummel, Kenneth Holstein
Organization: Carnegie-Mellon University
Abstract:
This project will create and demonstrate new technology that supports dynamically-differentiated instruction for the classroom of the future. This new vision centers on carefully-designed partnerships between teachers, students, and artificial intelligence (AI). AI-powered learning software will support students during problem-solving practice, providing either individual guidance (using standard intelligent tutoring technology) or guidance so students can effectively collaborate and tutor each other. These learning activities are constantly adjusted to fit each student’s needs, including switching between individual or collaborative learning. The teacher “orchestrates” (instigates, oversees, and regulates) this dynamic process. New tools will enhance the teacher’s awareness of students’ classroom progress. The goal is to have highly effective and efficient learning processes for all students, and effective “orchestration support” for teachers. We will implement and test this vision in the context of AI-enhanced mathematics learning in middle school. The proposed work will greatly enhance current understanding of how to design effective AI-based “co-orchestration” tools that draw on complementary strengths of teachers, students/peers, and AI agents to make the vision of the dynamically-differentiated classroom feasible. It will provide insight into the new classroom dynamics that arise. The work may ultimately contribute to more individualized K-12 education. The work will create a testbed that could be used to explore and rigorously test a wide range of interesting hypotheses regarding co-orchestration and dynamic differentiation of individual and collaborative learning.
Effective orchestration of dynamically-differentiated instruction poses significant challenges in terms of design and technical implementation. Although existing AI-based learning technologies support forms of dynamic differentiation of instruction, they tend not to support dynamic combinations of individual and collaborative learning; in fact, most only support one of these two learning modes. In addition, existing teacher support tools tend to focus only on enhancing teacher awareness, not on supporting teachers’ in-the-moment decision-making and action, and not on supporting dynamic interleaving of individual and collaborative learning. In the proposed work, we tackle this challenge by integrating and extending four strands of work: intelligent tutoring systems technology; a learning environment to support combinations of individual and collaborative learning; adaptive technology support for mutual peer tutoring; and a mixed reality tool (“smart glasses”) to support teacher/AI co-orchestration. Building on this foundation, we create and demonstrate technology support for dynamically-differentiated instruction by three strands of work. First, we create AI-based tutoring software capable of supporting both individual learners and students doing mutual peer tutoring. Second, we develop a tool to support the intelligent, real-time co-orchestration, between students, teachers, and AI agents, of dynamically differentiated combinations of collaborative learning and individual learning. We do so through design-based research, prototyping, and classroom piloting. Third, we evaluate the newly-created technology for dynamically-differentiated collaborative classroom in schools, to gain an initial understanding of its benefits and challenges, and the changes in classroom practices and learning outcomes that it brings about.
This award reflects NSF’s statutory mission and has been deemed worthy of support through evaluation using the Foundation’s intellectual merit and broader impacts review criteria.
