Cognitive workload evaluation of landmarks and routes using virtual reality

Abdurrahman, Usman Alhaji; Zheng, Lirong; Yeh, Shih-Ching · 2022 · Crossref

DOI: 10.1371/journal.pone.0268399

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates how the availability of landmarks and route complexity influence navigational efficiency, cognitive workload, and learning transfer during driving. Motivated by the need to understand how spatial information affects driver behavior and safety, the researchers hypothesized that insufficient landmarks and difficult routes would increase psychophysiological activation and error rates, thereby hindering the application of acquired knowledge. The experiment utilized a virtual reality (VR) driving system involving 79 undergraduate participants with no prior driving experience. The study employed a 2x2 between-group design manipulating two variables: landmark availability (sufficient vs. insufficient) and route difficulty (easy vs. difficult). Participants were divided into groups navigating routes with either seven landmarks (sufficient) or three to five landmarks (insufficient). Easy routes featured three turns and traffic lights, while difficult routes included five of each. Data collection involved tracking psychophysiological metrics—specifically heart rate and pupil size via eye-tracking and heartbeat sensors—and driving performance metrics, including task completion time, wrong turns, and collision counts. To evaluate cognitive load, the researchers applied statistical t-tests and employed machine learning classifiers (SVM, ANN, Naïve Bayes, KNN, and Decision Tree) combined with three data fusion strategies: feature-level, decision-level, and hybrid-level fusion. The results demonstrated that insufficient landmarks and difficult routes significantly increased cognitive workload, evidenced by higher heart rates and larger pupil sizes compared to conditions with sufficient landmarks and easy routes. Specifically, participants in groups with insufficient landmarks exhibited significantly higher physiological arousal and made more navigation errors. Navigational efficiency was markedly higher for routes with sufficient landmarks, where participants completed tasks faster and with fewer mistakes. The study also found that high cognitive loads negatively impacted learning transfer, making it difficult for participants to apply knowledge acquired during initial training. Among the analytical methods, the hybrid-level data fusion approach achieved the highest accuracy in classifying cognitive workload states, outperforming other classification methods. These findings underscore the critical role of environmental landmarks in reducing cognitive load and enhancing navigational efficiency. The study suggests that designing driving routes with adequate visual cues can mitigate driver stress and error rates. Furthermore, the successful application of VR and multi-modal data fusion provides a robust framework for future research in traffic safety, offering a method to analyze driver cognition and performance in controlled, immersive environments.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• mental demand
• workload measurement
• nav systems
• road complexity
• cognitive capacity variation

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data
• Methodological Resource: metric or index
• Theoretical Contribution: theory or model