Multi-Modal Evaluation of Adaptive Interface Design and Railway Grade Crossing Infrastructure, in Simulated Driving Environments

Abstract

This study examines the usability and cognitive workload associated with two interface concepts ‒ Context-Driven Adaptive Dashboard System and a Bring-Your-Own- Device (BYOD) graphical interface ‒ tested within a high-fidelity driving simulator across urban and rural routes. Nineteen participants completed realistic driving scenarios, during which physiological, behavioral, and subjective data were collected. Usability was assessed using the System Usability Scale, while mental workload was measured with the NASA Task Load Index. The results show that the context-aware interface achieved a 20.9% higher usability score compared to a Bring-Your-Own-Device interface (71.7 vs. 59.3, p = 0.0105). However, workload levels did not differ significantly across the interfaces. The experiment also analyzed driver behavior at both secured and unsecured railway grade crossings using eye-tracking technology. Eye-tracking analysis revealed unsecured crossings elicited 30.2% more fixations, a 13.8% increase in fixation frequency, and a 26.4% decrease in average fixation duration (p < 0.01), reflecting elevated visual search activity and uncertainty. While statistical comparisons of driver risk behavior at crossings yielded limited significance, observed trends consistently pointed to safer actions at secured crossings. These findings underscore the importance of adaptive interface design and intelligent infrastructure in reducing driver distraction and enhancing safety in both everyday and critical driving situations.