How should rhythmic discontinuous murals in extra-long tunnels be evaluated? Evidence from driving simulation
Related Experiment Video
Updated: Nov 3, 2026
Evaluation of an Exclusive Spur Dike U-Turn Design with Radar-Collected Data and Simulation
Published on: February 1, 2020
Wenhui Dong1, Zhiqing Zhang1, Nale Zhao2
1Beijing Key Laboratory of Traffic Engineering, College of Metropolitan Transportation, Beijing University of Technology, Beijing, P.R. China.
Objectives:
Extra-long tunnels are visually monotonous environments that may weaken drivers' visual references and affect vehicle control. Tunnel sidewall murals can provide additional visual information, yet the effects of mural rhythm and discontinuous deployment on driver responses remain insufficiently quantified. This study evaluated discontinuous mural rhythm schemes in an extra-long tunnel using a driving simulation experiment.
Methods:
Six mural rhythm schemes were designed, and vehicle operation and eye-tracking data were collected from 27 licensed drivers. Indicators of visual search, visual workload, longitudinal control, and lateral control were analyzed. To compare the schemes, this study used a three-stage ANOVA-FCE-SPA framework: repeated-measures ANOVA examined indicator-level differences, FCE generated whole-tunnel and sectional rankings, and SPA described the identity, discrepancy, and opposition patterns observed alongside those rankings.
Results:
The results showed that discontinuous mural rhythm affected visual processing, speed regulation, and lateral control, with heterogeneous responses across indicators. Fixation duration, pupil-related indicators, lateral position, and steering wheel angle were more sensitive to rhythm variations, whereas speed- and acceleration-related indicators mainly reflected local responses. FCE showed spatial variation in integrated performance: the no-mural scheme A1 ranked highest for the whole tunnel, M1, and M2, whereas A4 ranked highest in M3, indicating that the relative performance of the tested schemes varied across tunnel sections. Set pair analysis further characterized relative indicator-state patterns across schemes and tunnel sections, with lateral position, steering wheel angle, and pupil-related indicators showing greater scheme- and section-specific variation.
Conclusions:
The ANOVA-FCE-SPA framework integrates indicator-level response analysis, comprehensive scheme evaluation across spatial scales, and structural interpretation of multi-indicator response patterns. It provides a quantitative basis for selecting mural rhythm parameters and optimizing tunnel sidewall visual environments in extra-long tunnels.
