Non-linear parameterization of spatial decision making in immersive virtual environment

• Published in: 2024 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW) pp. 389 - 395
• Main Authors: Sunny, M. J. M., Basu, Aryabrata
• Format: Conference Proceeding
• Language: English
• Published: IEEE 16.03.2024
• Subjects: Decision making; Motors; Navigation; Neural networks; Three-dimensional displays; User interfaces; Virtual environments; Virtual reality-Human-centered computing-Spatial navigability-Spatial decision-making-SHAP Analysis-Neural networks
• DOI: 10.1109/VRW62533.2024.00076

We explore spatial navigability in virtual environments, focusing on the role of reflexive motor skills and decision-making processes. Maze-solving scenarios serve as a highly effective tool for stimulating spatial reasoning. Both expert and non-expert navigators' responses to these spatial experiences may vary, depending on their individual capacity for spatial decision-making and reflexive motor abilities. Virtual mazes provide an unique opportunity to measure human decision-making capacity and motor skills within a reasonable threshold of second degree of mediation with gaming controllers. Factors such as task completion times, relative response times, spatial positioning, movement velocities, and others can be used to assess a person's ability to navigate a virtual environment (VE). The study contrasts the performance of expert and non-expert navigators, revealing that individual differences in spatial decision-making and reflexive motor proficiency lead to divergent experiences within these synthetic spaces. Additionally, this paper identifies the parameters of user trajectory that significantly influence users' expertise in navigating VEs, as explained through the trained neural network by SHAP (SHapley Additive exPlanations) analysis [10]. The observed discrepancies in performance determinants between immersive and non-immersive settings indicate that non-expert users face a steeper learning curve within immersive VEs. This suggests a need for tailored navigational mechanics in future virtual reality applications and studies, designed with a nuanced understanding of users' navigational expertise. By doing so, we can enhance the accessibility and efficacy of VEs as tools for both recreational and practical applications, ensuring a more inclusive and user-friendly experience for all participants.