Does AI Use Influence Students’ Academic Performance Directly and Indirectly Through Mental Health?

• In Event: Artificial Intelligence in Online Teaching and Learning

Abstract

The integration of generative AI (GenAI) in education presents both opportunities and challenges. This study investigates how two types of GenAI use—academic and personal communication/entertainment—affect postsecondary students’ academic performance directly and indirectly through mental health. Using data from the Healthy Minds Study and structural equation modeling, a serial multiple mediator framework was applied. Results show that academic use is positively associated with higher grades, whereas non-academic use is linked to greater mental health challenges and indirectly to lower performance via its negative impact on mental health. These findings highlight the need for guidance that promotes academic uses of GenAI while addressing mental health risks, informing evidence-based strategies for balanced AI integration in education.

Author

• Sunha Kim, University at Buffalo - SUNY