A Multidimensional Model of Metaverse Adoption: Insights from a Hybrid SEM-ANN Analysis

The Metaverse is transforming digital interaction and creating new opportunities for immersive learning in higher education. However, limited research has examined the behavioral, technological, and risk-related factors shaping students’ adoption of Metaverse technologies. To address this gap, this study develops and validates a multidimensional adoption model by integrating the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), Task-Technology Fit (TTF), and Protection Motivation Theory (PMT). Using a quantitative design, data were collected from 760 university students through an online survey and analyzed through a hybrid Partial Least Squares-Structural Equation Modeling (PLS-SEM) and Artificial Neural Networks (ANN) approach. The PLS-SEM results showed that performance expectancy, effort expectancy, social influence, facilitating conditions, hedonic motivation, price value, task-technology fit, response efficacy, self-efficacy, and perceived vulnerability significantly predicted Metaverse adoption, while response costs and perceived severity were not significant. ANN analysis further improved predictive accuracy and identified performance expectancy and self-efficacy as the most influential determinants. The study contributes a comprehensive framework for understanding Metaverse adoption and offers practical insights for designing effective, user-centered, and secure immersive learning environments in higher education.