The Eye of Disaster: Development and Evaluation of a VR-Based Landslide Education System

This study develops an immersive Virtual Reality (VR) pedagogical framework to mitigate the spatial cognitive constraints inherent in conventional two-dimensional disaster-prevention education. It focuses on the 2024 Noto Peninsula landslide events in Japan. By integrating high-precision digital elevation models and satellite imagery, the produced system reconstructs post-disaster terrains to facilitate high-fidelity risk communication. The interaction logic, governed by the natural user interface principles, incorporates a multi-perspective switching mechanism that enables users to conduct comprehensive analyses of disaster sites across varying spatial scales.

 

The system architecture comprises two core modules: the "VR Geological Museum" for knowledge acquisition and the "Evacuation Simulation" for practical application, enabling deep transfer from conceptual understanding to survival skills. The former employs a task-driven strategy and a "macro-micro" dual-perspective observation model. It transforms abstract geological knowledge into intuitive interactive experiences through high-precision 3D reconstructions of landslide topography, effectively lowering the cognitive threshold for non-expert learners. Complementing this, the evacuation simulation module integrates official landslide-disaster warning area maps from the Geospatial Information Authority of Japan. Grounded in embodied cognition theory, this module implements a "trial-and-error" feedback mechanism. By navigating highly restored disaster evolution scenarios, users translate static warning information into dynamic survival capabilities, thereby completing the cognitive loop from theoretical understanding to behavioral practice.

 

The pedagogical efficacy of the system was empirically validated through a randomized controlled trial, utilizing multidimensional standardized metrics, including the Presence Questionnaire, the System Usability Scale, and the NASA Task Load Index for workload assessment. Experimental results demonstrate that the system significantly outperforms traditional text-based media in knowledge internalization, risk perception accuracy, and survival decision-making efficiency. The core contribution of this research lies in the deep integration of high-fidelity geospatial data with immersive interaction, establishing a verifiable technical paradigm for disaster education. This approach effectively dismantles barriers to professional knowledge. It enhances disaster preparedness and evacuation efficacy across diverse demographic backgrounds, providing a robust theoretical and technical foundation for the universalization of geohazard education.