Latent Profiles of AI Tool Usage, Cognitive Load, and Fatigue Among Filipino College Students
Year: 2026
Type: conference-paper
Study design: Observational Study
Abstract: The integration of AI tools into higher education has introduced new patterns of student engagement that vary considerably in their associated cognitive and wellbeing outcomes. This study applied latent profile analysis (LPA) to identify distinct subgroups of Filipino college students based on their levels of AI tool integration, cognitive load, and fatigue and sleep disruption, and examined whether these profiles differed significantly on AI literacy, self-efficacy, and positive AI outlook. A cross-sectional survey was conducted with 391 university students from three Philippine higher education institutions. LPA identified four profiles: Minimal AI Users $(11 \%)$, Burdened AI Users (40.7%), Efficient AI Users ($\mathbf{3 6. 6 \%}$), and Overwhelmed AI Users ($\mathbf{1 1. 8 \%}$). One-way ANOVA with Tukey-corrected post-hoc comparisons revealed significant between-profile differences on all three outcome variables (all $\boldsymbol{p}{\lt}.001$), with self-efficacy showing the largest effect $\left(\eta^{2}=.375\right)$. Overwhelmed AI Users reported the highest self-efficacy and most favorable AI attitudes despite carrying the heaviest cognitive and physical burden, while Efficient AI Users demonstrated that high AI integration is achievable without substantial wellbeing costs. These findings show the heterogeneous nature of student AI experiences and highlight the need for profile-sensitive institutional interventions that simultaneously promote AI competence and safeguard student wellbeing in the Philippine higher education context.
Institution: Pampanga State University
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7165201019