Identifying Cybersickness Causes in Virtual Reality Games Using Symbolic Machine Learning Algorithms

Virtual reality developers face a persistent challenge in mitigating cybersickness, a condition that significantly hinders user retention and accessibility. By employing symbolic machine learning algorithms and random forest classifiers, researchers have successfully identified the primary mechanical triggers of discomfort without the need for external medical monitoring equipment. The analysis confirms that exposure duration, rapid rotation, and sudden acceleration are the most significant contributors to physiological distress during gameplay.

The research highlights a clear correlation between game genre and user susceptibility, noting that flight-based simulations induce symptoms more frequently than racing titles. Furthermore, individual experience levels play a critical role, as novice users demonstrate a higher sensitivity to the sensory conflicts inherent in virtual environments. These findings underscore the necessity of tailoring game mechanics to the target audience’s proficiency to ensure a comfortable experience.

To address these issues, designers should adopt proactive mitigation strategies integrated directly into the software architecture. Recommended interventions include the implementation of rotational blurring, tunneling effects, and real-time acceleration smoothing. Additionally, enforcing structured gameplay sessions—specifically limiting continuous exposure to five-minute intervals—serves as an effective safeguard against the onset of symptoms. By prioritizing lower-complexity control schemes and incorporating these technical design adjustments, developers can significantly improve the ergonomic quality of virtual reality experiences, ultimately fostering broader adoption across diverse user demographics.