Optimizing VR-UX: analysis and adaptive recommendations for enhancing immersion and reducing motion sickness

Abstract

This study presents an adaptive recommendation framework to enhance comfort and immersion in virtual reality (VR) by actively reducing motion sickness. Unlike prior research that views VR user experience (UX) as static, this approach integrates statistical analysis with dynamic system design. Using a Kaggle dataset of 1,000 entries, we applied descriptive statistics, Spearman correlation, Kruskal-Wallis tests, and regression to explore relationships among session duration, motion sickness, immersion, headset type, and user demographics. Findings show that session duration alone does not significantly predict motion sickness or immersion (R²=0.00, p>0.05), but certain user profiles, such as individuals over 30 using PlayStation VR, are more prone to discomfort. These insights inform a four-module framework: user profiling, real-time duration monitoring, rule-based adaptation logic (such as slowing navigation speed or adding a virtual nose for visual stability), and personalized in-VR recommendations. The system is compatible with Unity and Unreal Engine and integrates with commercial headset software development kits (SDKs). Future validation will use A/B testing, standardized questionnaires, simulator sickness questionnaire /immersion presence questionnaire (SSQ/IPQ), and physiological metrics. This work shifts VR design toward personalized, responsive systems that prioritize user well-being and immersive engagement.