Artificial intelligence use, academic engagement, and psychological well-being among secondary school students

Abstract

Artificial intelligence (AI) is rapidly becoming part of students’ educational and psychosocial environments. Yet evidence on its simultaneous influence on learning engagement and psychological well-being among secondary school students remains limited, particularly in low- and middle-income settings. This paper examined the emerging relationship between AI use, academic engagement, and psychological well-being among students. The study adopted an integrative literature-review design, synthesising four focal Nigerian sources with fourteen additional verified empirical, meta-analytic, and review sources, together with established theoretical and policy frameworks (Self-Determination Theory, Cognitive Load Theory, Social Cognitive Theory, the Transactional Model of Stress and Coping, and UNESCO's human-centred AI guidance), organised around the three constructs of AI use, academic engagement, and psychological well-being. The findings indicate that AI can support academic engagement through personalised assistance, rapid feedback, access to explanations, and learning support, with meta-analytic evidence reporting small-to-moderate positive effects on behavioural, cognitive, and emotional engagement and on learning outcomes. However, the same technologies may produce disengagement when they substitute for independent cognitive effort, facilitate academic dishonesty, or encourage excessive dependence and over-reliance. Psychological well-being presents a similarly dual picture: AI-enabled screening and conversational support may improve access, privacy, and early identification of psychological difficulties, while concerns about privacy, inaccurate outputs, emotional dependence, reduced human interaction, and inequitable access remain salient. The psychological effects of AI are best understood as contingent on the quality, intensity, purpose, and social context of use. The study concludes that AI should be positioned as a supervised, human-centred educational and psychosocial support tool rather than a replacement for teachers, counsellors, parents, or students' own learning processes. It recommends psychologically informed AI literacy, teacher supervision, clear school policies, human oversight of mental-health applications, and routine monitoring of students' engagement and well-being, and calls for longitudinal Nigerian research that examines AI use, engagement, and well-being within the same student population.