Reimagining Teaching and Learning in the Age of Generative AI Agents

This collection supports and amplifies research related to SDG 4: Quality Education.

Generative AI is transforming the conventional dyadic teacher-student dynamic into a triadic framework centered around teacher-student-agent interactions, with the potential to further evolve into multi-agent collaborative educational networks. LLM-based agents are redefining educational roles by serving as instructional assistants (generating curricular content, designing pedagogical activities, facilitating assessment), learning companions (engaging in Socratic dialogue, providing socio-emotional and motivational scaffolding), and classroom analysts (structuring discourse, delivering formative insights). This reconfiguration is fundamentally altering instructional structures, temporal dynamics, and feedback mechanisms.

This Special Collection explores how LLM-based educational agents are restructuring the ecology of teaching and learning, emphasizing scalable, interpretable, and governable practices supported by empirical evidence across diverse disciplines, educational levels, and media formats.

Themes of Interest include, but are not limited to:

1. Theoretical Foundations and Models of Multi-Agent Learning Ecologies

This theme examines how multi-agent systems reconfigure teacher–student–agent interactions. Drawing on distributed cognition, collaborative learning, and ecological perspectives, it explores how learning processes emerge through coordination, role distribution, and knowledge-building across human and AI agents.

2. Design Principles for Multi-Agent Educational Systems

Focusing on the design of educational agents, this theme emphasizes innovations in agent architectures (e.g., RAG, chain-of-thought, multimodal interaction) alongside learning sciences principles such as scaffolding, inquiry, and formative assessment. It highlights how multi-agent design can foster deeper learning and support adaptive, transparent, and trustworthy practices.

3. Practical Implementations of Multi-Agent Companions and Assistants

This theme investigates real-world applications of educational agents functioning as companions and teaching assistants. Studies address how multi-agent support systems enable personalization, adaptive scaffolding, and diagnostic feedback across diverse disciplines, educational levels, and contexts, grounded in empirical evidence from learning sciences.

4. Multi-Agent Collaboration and Simulated Classroom Environments

This theme explores the use of multi-agent collaboration in both synthetic and authentic classrooms. It covers role distribution, conflict resolution, and collective reasoning, while leveraging simulations and agent-based modeling to test learning designs. Contributions emphasize how multi-agent systems extend collaborative learning theories and provide scalable tools for evaluating emergent classroom dynamics.

5. Cognitive and Learning Mechanisms in AI-Augmented Education

This theme investigates how interactions with educational agents shape cognitive and learning processes, drawing on theories of attention, memory, metacognition, and cognitive load. Studies may integrate findings from educational psychology, cognitive science, and neuroscience to examine how multi-agent systems influence learning pathways and adaptive scaffolding.

6. Socio-Emotional and Motivational Dynamics in Human–AI Learning

This theme focuses on the affective and motivational dimensions of multi-agent educational systems. It considers how agents function as companions to sustain engagement, provide emotional scaffolding, and foster belonging. Cross-cultural and psychological perspectives are encouraged to deepen understanding of trust, motivation, and socio-emotional support in AI-mediated learning.

7. Teacher–AI Collaboration and Pedagogical Transformation

This theme explores the evolving role of teachers in multi-agent learning ecologies. It emphasizes how educators design, coordinate, and reflect upon pedagogical practices with AI partners, including curriculum co-design, formative assessment, and data-informed decision-making. Studies may investigate how teacher identity, professional growth, and instructional effectiveness are reshaped in AI-rich contexts.

8. Equity, Ethics, and Policy in AI-Enhanced Learning

This theme addresses the societal and ethical dimensions of deploying multi-agent systems in education. Key issues include fairness, transparency, privacy, and inclusivity, with attention to how AI adoption intersects with equity and access. Policy-oriented contributions are welcome, providing frameworks for responsible governance of AI in diverse educational settings.

9. Future Directions in Learning Science and AI Integration

This theme considers the long-term implications of multi-agent systems for the science of learning. It highlights how AI can function both as a learning tool and as a research instrument, enabling new ways of studying cognition, collaboration, and development. Contributions may outline emerging paradigms and research agendas for the next generation of learning sciences.