Cybersickness remains a major obstacle to the widespread adoption of immersive virtual reality (VR), particularly in consumer-grade environments. While prior methods rely on invasive signals such as electroencephalography (EEG) for high predictive accuracy, these approaches require specialized hardware and are impractical for real-world applications. In this work, we propose a scalable, deployable framework for personalized cybersickness prediction leveraging only non-invasive signals readily available from commercial VR headsets, including head motion, eye tracking, and physiological responses. Our model employs a modality-specific graph neural network enhanced with a Difference Attention Module to extract temporal-spatial embeddings capturing dynamic changes across modalities. A cross-modal alignment module jointly trains the video encoder to learn personalized traits by aligning video features with sensor-derived representations. Consequently, the model accurately predicts individual cybersickness using only video input during inference. Experimental results show our model achieves 88.4% accuracy, closely matching EEG-based approaches (89.16%), while reducing deployment complexity. With an average inference latency of 90ms, our framework supports real-time applications, ideal for integration into consumer-grade VR platforms without compromising personalization or performance. The code will be relesed at https://github.com/U235-Aurora/PTGNN.

Virtual reality offers the opportunity for immersive exploration, yet it is often undermined by cybersickness. However, how individuals strike a balance between exploration and discomfort remains unclear. Existing method (e.g., reinforcement learning (RL)) often fail to fully capture the complexities of navigation and decision-making patterns. This study investigates how curiosity influences users’ navigation behavior, particularly how users strike a balance between exploration and discomfort. We propose curiosity as a key factor driving irrational decision-making and apply the free energy principle to model the relationship between curiosity and user behavior quantitatively. Our findings indicate that users generally adopt conservative strategies when navigating. Also, curiosity levels tend to rise when the virtual environment changes. These results illustrate the dynamic interplay between exploration and discomfort. In addition, it offers a new perspective on how curiosity drives behavior in immersive environments, providing a foundation for designing adaptive VR environments. Future research will further refine this model by incorporating additional psychological and environmental factors to improve prediction accuracy.