Al-Powered Socratic Learning for Psychological Statistics: Enhancing Flipped Classroom Practiceswith Generative Intelligent Tutoring Systems

Abstract

This study integrates the Socratic Playground Intelligent Tutoring System (ITS) into a flipped classroom model for a psychological statistics course. The system, based on an AI-supported Learning Management System (LMS) with a Socratic-style large language model, is combined with flipped classroom teaching to explore the impact of adaptive pre-class tasks, personalized grouping, continuous post-class reflection, and various teaching methods on instruction effectiveness. It investigates the optimal timing, methods, and strategies to improve teaching and learning outcomes, with the goal of enhancing classroom interaction, learning atmosphere, student interest, and learning effectiveness, while fostering problem-solving, critical thinking, and autonomous learning skills. The study also provides data-driven insights to refine the system. The process includes pre-class task assignments, in-class personalized grouping, post-class reflection, and evaluation of classroom configuration to assess effectiveness and inform improvements. The results aim to optimize teaching and system design, supporting future implementation and refinement.