#392: Designing an AI-Powered Reflective Partner: Translating Self-Regulated Learning Theory into Generative Chatbot Scaffolds

This interactive tool paper presents the design and development of an AI-powered chatbot that scaffolds college students’ self-regulated learning (SRL). Grounded in Pintrich’s SRL framework and informed by meta-analytic evidence on effective SRL interventions, the chatbot is designed as a reflective learning partner that supports metacognitive monitoring, strategic deployment, and affective regulation. Using a design-based research (DBR) approach, this study explores how SRL theory can guide the development of generative AI–mediated interactions. We describe the theoretical rationale underlying the tool, articulate key design principles, present the prototype’s architecture, and discuss implications for improving evidence-based, equitable AI scaffolds to enhance undergraduate learning outcomes.