A Theory-Guided LLM Pedagogical Agent for STEM+C Scaffolding Without Over-Reliance

2026-05-28

AI Summary

This work proposes Copa, a multi-agent, multimodal collaborative peer agent grounded in an Evidence-Decision-Feedback (EDF) framework, which systematically integrates social cognitive theory and social constructivism to address the common oversight of learning theories in existing large language model–based tutoring agents. By providing adaptive dialogic scaffolding that fosters students’ meaning-making rather than delivering direct answers, Copa mitigates cognitive offloading and overreliance. The system delivers interpretable, personalized feedback and was evaluated in computational modeling experiments with 33 high school student groups. Results demonstrate significant improvements in students’ conceptual understanding and self-efficacy, with no observed signs of dependency, thereby validating Copa’s theoretically informed design and its effectiveness in promoting autonomous learning.

Abstract

LLM pedagogical agents are proliferating, yet recent findings have raised questions about their adherence to established theories of learning and, by extension, their educational value. Concerns regarding cognitive offloading, over-reliance, and "gaming" behaviors persist and remain largely unaddressed. In response, we developed Copa, an agentic, multi-agent, multimodal Collaborative Peer Agent for STEM+C learning. Copa is built on top of the Evidence-Decision-Feedback (EDF) framework, grounding its interactions in Social Cognitive Theory and Social Constructivism and promoting sense-making through adaptive, dialogic support rather than answer-seeking. In an authentic high school computational-modeling study (n=33 dyads), we demonstrate that Copa (1) supports students' confidence building and ability to verbalize conceptual understanding without causing dependence; and (2) provides adaptive feedback personalized to learners that is interpretable with respect to students' multimodal input data. These findings position theory-guided, multimodal LLM agents as a promising path toward classroom AI integration that amplifies students' reasoning rather than replacing it.

Problem

• LLM pedagogical agents
• over-reliance
• cognitive offloading
• educational value
• learning theory

Innovation

• theory-guided LLM
• multimodal pedagogical agent
• adaptive scaffolding
• social constructivism
• cognitive offloading prevention