Integrating RAG-Enhanced AI to Support Cognitive, Affective, and Psychomotor Balance in K–12 Lesson Plans

• In Event: AERA Roundtable Session Wednesday 1:45 pm - Gold 3
  In Roundtable Session: Innovative Tools for Supporting Social Emotional Learning

Abstract

Although Social-Emotional Learning (SEL) is widely recognized as essential to whole-child development, affective objectives remain underrepresented in K–12 lesson design. This study investigates the use of an RAG–enhanced AI system to strengthen SEL integration and rebalance instructional goals, aligned with Taiwan’s 108 Curriculum Guidelines. Using Bloom’s Taxonomy, the CASEL framework, and psychomotor learning theory, we analyzed 58 lesson plans by pre-service teachers. A transformer-based classifier revealed that only 17% of objectives targeted the affective domain. After AI-driven augmentation, SEL coverage rose to 32%, with significant gains in self-management and decision-making. This research presents a scalable, explainable AI framework for embedding SEL into curriculum planning—advancing emotional depth, equity, and teacher agency in pre-service teacher education.

Author

• Shun-Ching Hsieh, National Chengchi University國立政治大學