SRLAgent: Enhancing Self-Regulated Learning Skills through Gamification and LLM Assistance
Project Overview
The document explores the use of generative AI in education through the lens of SRLAgent, a system designed to bolster self-regulated learning (SRL) skills in college students by integrating gamification and adaptive support. It highlights the challenges students face in developing SRL skills and presents a framework aligned with Zimmerman's three-phase SRL model. The findings indicate that SRLAgent effectively enhances SRL skills, motivation, and engagement by providing personalized feedback and gamified learning experiences. Despite the positive outcomes, the study acknowledges challenges related to students' over-reliance on AI support and emphasizes the necessity for ongoing adjustments to the system to optimize its effectiveness. Overall, the document illustrates the potential of generative AI to facilitate personalized learning experiences while also addressing the complexities involved in fostering student independence in their learning processes.
Key Applications
SRLAgent - A gamified LLM-assisted learning system
Context: Higher education targeting college students, particularly freshmen who face academic challenges in self-regulated learning.
Implementation: Developed based on a formative study that included interviews and surveys, the system was built using Minecraft to provide an immersive learning environment, utilizing LLMs for real-time feedback.
Outcomes: Significant improvements in SRL skills (p< .001) and higher engagement compared to baseline conditions. The system promotes effective goal-setting and self-reflection.
Challenges: Challenges include potential over-reliance on AI support, which may inhibit independent learning, and the need for personalized adaptation to address varying student needs.
Implementation Barriers
User Engagement
Students may focus on achieving rewards rather than engaging deeply with content, leading to superficial learning.
Proposed Solutions: Implement adaptive fading strategies to gradually reduce AI support as students become more proficient.
Personalization
Many existing educational tools do not provide sufficient personalization and adaptability to individual learner needs.
Proposed Solutions: Integrate more dynamic feedback mechanisms and personalized learning pathways to enhance user experience.
Project Team
Wentao Ge
Researcher
Yuqing Sun
Researcher
Ziyan Wang
Researcher
Haoyue Zheng
Researcher
Weiyang He
Researcher
Piaohong Wang
Researcher
Qianyu Zhu
Researcher
Benyou Wang
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Wentao Ge, Yuqing Sun, Ziyan Wang, Haoyue Zheng, Weiyang He, Piaohong Wang, Qianyu Zhu, Benyou Wang
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai