Promoting Self-Regulation Progress and Knowledge Construction in Blended Learning via ChatGPT-Based Learning Aid
Ting-Ting Wu, Hsin-Yu Lee, Pin-Hui Li, Chia-Nan Huang, Yueh-Min Huang- Computer Science Applications
- Education
This study combines ChatGPT, Apple’s Shortcuts, and LINE to create the ChatGPT-based Intelligent Learning Aid (CILA), aiming to enhance self-regulation progress and knowledge construction in blended learning. CILA offers real-time, convergent information to learners’ inquiries, as opposed to traditional Google search engine that provide divergent information. By addressing questions promptly, CILA minimizes interruptions during the performance phase of self-regulation progress. The tool records learners’ questions and answers, aiding self-reflection in self-regulation progress. We evaluated self-regulation progress using motivation, engagement, and self-efficacy as indicators. Findings show that CILA’s intervention effectively improves self-regulation progress and knowledge construction, offering benefits over divergent information in blended learning contexts with respect to amotivation, intrinsic motivation, and behavioral engagement. This research highlights the potential of incorporating large language models like ChatGPT in educational settings to support teachers and students.