Comparing drivers’ visual attention at Junctions in Real and Simulated Environments

Robbins, Chloe J.; Allen, Harriet A.; Chapman, Peter · 2019 · OpenAlex-citations

DOI: 10.1016/j.apergo.2019.05.005

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 the behavioral validity of high-fidelity driving simulators by comparing drivers’ visual attention at junctions in simulated versus real-world environments. While simulators are widely used for driving research, previous validation studies often focused on simple tasks like speed regulation. This research addresses a gap in validating complex, higher-level cognitive tasks, specifically visual search strategies at intersections, which are critical for safety but involve intricate head and eye movements. The authors hypothesized that simulator validity would depend on task demand, predicting greater discrepancies between simulated and real driving in low-demand scenarios compared to medium-demand ones. The study employed a 2 × 2 repeated-measures design with 15 participants who drove in both a high-fidelity simulator (NITES facility with a 360-degree screen) and an instrumented real vehicle. Participants navigated six junctions in each environment, categorized as either low demand (controlled by traffic lights and road geometry) or medium demand (requiring drivers to judge gaps in traffic to pull out). To ensure comparability, the simulator replicated the exact road geometrics of the Nottingham junctions used on-road, and traffic levels in the simulation were "yoked" to match the traffic density observed during the real-world drives. Eye and head movements were recorded using head-mounted trackers in both settings to capture fine-grained visual search data. The results indicated that broad search strategies, measured by the frequency and size of head movements, were not significantly different between the simulator and real-world driving, suggesting that simulators accurately capture general scanning behaviors. However, significant differences emerged in fine-grained eye movement metrics. Mean fixation durations were longer in the simulator than on-road, particularly in low-demand situations. This suggests lower visual engagement with the simulated environment when task demands are minimal. Conversely, as task demand increased to medium levels, drivers exhibited longer junction crossing times, more head movements, shorter fixation durations, and larger saccadic amplitudes in both environments, with the differences between simulator and real-world performance diminishing. The findings imply that high-fidelity driving simulators are valid tools for investigating drivers’ visual attention at junctions, provided the driving task involves at least moderate demand. The study highlights that while simulators may underrepresent visual engagement in low-demand scenarios due to reduced environmental richness, they effectively replicate the complex visual search strategies required for more demanding driving tasks. This supports the use of simulators for safety research involving intersection maneuvers, while cautioning against generalizing results from low-demand simulated tasks to real-world behavior.

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-prismaquant on 2026-06-25; verification: verified.

Topics

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

• simulator validity fidelity
• simulator training transfer
• visual occlusion
• attention allocation
• useful field of view
• situational awareness

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).

• Methodological Resource: tool software, validation psychometrics, measurement protocol