Distracting or informative? Examining signage for cyclists using eye-tracking

Aasvik, Ole; Fyhri, Aslak · 2022 · Crossref

DOI: 10.55329/wxcy5694

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 whether a new signage system implemented in Oslo, Norway, effectively guides cyclists or acts as a distraction in complex traffic environments. Motivated by political goals to increase cycling shares and improve safety, the research addresses the lack of systematic knowledge regarding cyclists' visual attention and wayfinding strategies. Specifically, it examines if increased route information helps navigation or detracts from attention to other road users, potentially compromising safety. The researchers employed a mixed-methods approach involving eye-tracking and machine learning. Forty cyclists with limited local route knowledge navigated a designated route through Oslo’s city center. Data collection occurred in two phases: pre-intervention (June 2020, N=20) and post-intervention (September 2020, N=28), after the city introduced six new guidance measures, including road markings, route identity signs, and intersection maps. Participants wore Pupil Labs eye-tracking glasses and GoPro cameras with GPS. The study utilized a novel machine learning algorithm based on FasterRCNN NAS to automatically detect and categorize objects in the video footage, linking gaze data to specific traffic elements like pedestrians, vehicles, and signs. Route adherence was analyzed via GPS data at three critical intersections. Results indicated that the new signage increased the proportion of cyclists following the intended route at key intersections, with increases ranging from 12% to 22%. However, most cyclists did not strictly follow the entire signed route. Qualitative analysis of gaze behavior revealed that cyclists who deviated from the suggested route cycled faster and looked significantly less at route signs than those who adhered to it, though both groups noticed similar amounts of road markings. Crucially, the machine learning analysis found no statistically significant difference in the time spent looking at other road users or signs between the pre- and post-intervention periods. This suggests the new signage did not distract cyclists from monitoring their surroundings. The study concludes that the new signage strategy was informative rather than distracting, supporting safer navigation without compromising attention to traffic hazards. The authors highlight the utility of combining eye-tracking with machine learning for objective, large-scale analysis of cyclist gaze behavior, noting it reduces researcher bias compared to manual coding. However, they caution that gaze fixation does not equate to conscious awareness, as cyclists may perceive stimuli peripherally. The findings suggest that improved signage can facilitate route choice, particularly for slower cyclists, while future research should refine models to better account for individual variability and peripheral vision.

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 2026-06-26; verification: verified.

Topics

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

• external distraction
• vru facing ehmi
• vru conspicuity
• gaze based attention detection