1632 - AI IN MENTAL HEALTH PROMOTION IN EDUCATIONAL SETTINGS: A SYSTEMATIC REVIEW AND META-SYNTHESIS OF USES, OUTCOMES, BENEFITS AND CHALLENGES

Session: D05S006 - Artificial Intelligence and learning
AUTHORS:
Simon Patricia (Hong Kong Baptist University ~ Hong Kong ~ Hong Kong) , Nalipay Ma Jenina (The Education University of Hong Kong ~ Hong Kong ~ Hong Kong) , Yema Dan Paolo (The Chinese University of Hong Kong ~ Hong Kong ~ Hong Kong) , Dela Cruz Isaiah (Massachusetts Institute of Technology ~ Cambridge ~ United States of America) , Liu Tongxi (University of Macau ~ Macao ~ Macao)
Abstract text:
Artificial intelligence (AI) integration into educational settings has opened new avenues for mental health promotion. Despite reviews and meta-analysis conducted on AI's effectiveness in addressing mental health issues, a comprehensive understanding of the uses, outcomes, benefits, and challenges of AI-driven mental health initiatives across all educational levels remains limited. This systematic review and meta-synthesis aim to critically examine the current landscape of AI applications in mental health promotion within educational environments, synthesizing evidence from qualitative studies on their uses, outcomes, benefits and barriers to implementation. We systematically searched six databases (WoS, Scopus, MEDLINE, Embase, PsycINFO, and Google Scholar) for peer-reviewed articles published between 2020-2025. Studies were included if they reported on AI-based interventions or tools aimed at promoting mental health among students across educational levels. Data were extracted and synthesized using thematic analysis and meta-synthesis techniques. Thirty-nine studies met inclusion criteria, encompassing a range of AI approaches such as chatbots, wearables, social robot, mobile agents, etc. We found that AI were used for assessment/diagnosis, adoption of therapist's role, enhancement of socio-emotional skills, facilitation of the therapeutic process, improvement of well-being and functioning, and management of mental health symptoms. Mental health outcomes addressed by AI were grouped into positive mental health, negative mental health, and physiological functioning. Reported benefits include positive user experience and engagement, adoption of therapist's positive characteristics, enhanced mental health and well-being of users, enhanced socio-emotional skills, enhanced therapeutic process, and privacy and confidentiality. Challenges identified were related to client factors, technology and design issues, negative attitude towards AI, lack of counsellor characteristics, and resource intensiveness. AI holds promise for mental health promotion, yet its implementation in educational settings is accompanied by notable practical and methodological challenges. Future research should prioritize transparent reporting, stakeholder involvement, and robust evaluation frameworks to maximize AI's potential while safeguarding student well-being.