Designing GenAI Tools for Personalized Learning Implementation: Theoretical Analysis and Prototype of a Multi-Agent System

2025/03/19 by Ling Zhang, Zijun Yao, Arya Hadizadeh Moghaddam
Computer Science · #Intelligent Tutoring Systems and Adaptive Learning #Online Learning and Analytics #Open Education and E-Learning

paper · doi:10.1177/00224871251325109

crossref issued 2025/03/19 · crossref published 2025/03/19 · crossref published-online 2025/03/19 · openalex publication_date 2025/03/19 · crossref created 2025/03/19 · crossref published-print 2025/05/01 · openalex created_date 2025/10/10 · crossref deposited 2026/05/01 · crossref indexed 2026/07/31 · openalex updated_date 2026/07/31

Abstract

Educator preparation, personalized learning (PL) implementation, and applications of Generative AI converge as three interrelated systems that, when carefully designed, can help achieve the long-sought goal of providing inclusive education for all learners. However, realizing this potential comes with challenges resulting from theoretical complexities and technological constraints. This article provides a theoretical analysis of the complex interconnectedness among these systems guided by the Cultural-Historical Activity Theory (CHAT). Building on the analysis, we introduce CoPL, a multi-agent system consisting of multiple agents with distinct functions that facilitate the complex PL design and engage pre-service teachers (PSTs) in dynamic conversations while prompting them to reflect on the inclusivity of agent-generated instructional suggestions. We describe the affordances and limitations of the system as a professional learning tool for PSTs to develop competencies for designing inclusive PL to meet diverse learning needs of all learners. Finally, we discuss future research on refining CoPL and its practical applications.

Citations

• “Vygotsky’s Neglected Legacy”: Cultural-Historical Activity Theory
• On the Dangers of Stochastic Parrots
• A Systematic Review of Research on Personalized Learning: Personalized by Whom, to What, How, and for What Purpose(s)?
• Aion Framework: Dimensional Emergence of AI Consciousness, Observer-Induced Collapse, and Cosmological Portal Dynamics
• Distributing Accountability, Not Capability: Phase Separation and the LLM Workflow Quadrant in Autonomous AI Agent Architectures
• A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT
• Retrieval-Augmented Generation for Large Language Models: A Survey
• Large Language Models are Not Stable Recommender Systems
• Agent AI: Surveying the Horizons of Multimodal Interaction
• A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
• Bias and Fairness in Large Language Models: A Survey
• A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
• Beware of metacognitive laziness: Effects of generative artificial intelligence on learning motivation, processes, and performance
• Expert thinking with generative chatbots.