Learner emotions and performance in hypercasual VR games with adaptive AI difficulty

Hypercasual virtual reality games (HVRGs) are widely regarded as cost‐effective tools for rapid skill acquisition, yet the mechanisms that optimise their effectiveness and user acceptance remain insufficiently explored. This mixed‐methods empirical study investigates how playful emotions, characterised by engagement, enjoyment and anxiety, influence performance, measured through game scores, in HVRGs employing two distinct difficulty strategies: (1) progressive game difficulty (PGD) and (2) dynamic game difficulty (DGD). Using a mixed‐methods approach, data were collected from 50 adult participants during gameplay sessions of TargetShootingMania for this study. The findings aim to enhance understanding of the interplay between playful emotions and performance across difficulty strategies, contributing to the design of more effective and engaging hypercasual VR learning tools. The results indicate that the rule‐based AI‐enabled DGD significantly enhances performance by fostering higher levels of engagement and enjoyment while reducing anxiety compared to the PGD. The findings suggest that adaptive difficulty settings in HVRGs can better align with adult learners' capabilities, thereby optimising learning outcomes through enhanced emotional engagement. These insights contribute to the understanding of effective smart learning game design, emphasising the importance of tailoring game challenges to individual player needs to promote optimal learning experiences. The study underscores the potential of dynamic difficulty adjustments in smart learning to improve both the efficacy and enjoyment of learning through play. It also provides practical insights for game developers and educators, highlighting the importance of adaptive difficulty mechanisms in improving user engagement, learning outcomes, and enjoyment in HVRGs.