Crash Risk: Eye Movement as Indices for Dual Task Driving Workload

Kang, Julie J; Bian, Zheng; Andersen, George J · 2009 · Crossref

DOI: 10.17077/drivingassessment.1343

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 eye movements serve as indices of workload during dual-task driving, specifically examining whether increased task difficulty leads to "looking but failed to see" errors or cognitive tunneling. Motivated by the high prevalence of driver distraction in crashes, the researchers sought to determine if drivers reduce their useful field of view (UFOV) or restrict visual scanning under high cognitive load. The study tested two competing hypotheses: the "look but failed to see" hypothesis, which predicts increased saccades to compensate for a shrinking UFOV, and the cognitive tunneling hypothesis, which predicts decreased saccades as drivers focus attention narrowly. The experiment utilized a driving simulator with eight licensed college students. Participants performed a car-following task, maintaining a 21-meter distance behind a lead vehicle (LV) whose speed varied according to sinusoidal profiles. Task difficulty was manipulated by varying the LV’s average speed (60 vs. 100 km/h) and the amplitude of its speed changes (100% vs. 120%). Simultaneously, drivers performed a light detection task, identifying color changes in peripheral light arrays. Eye movements were recorded at 250 Hz using an Eyelink II tracker, measuring saccades, fixations, pupil size, and blinks. Performance metrics included car-following accuracy (headway distance, RMS error) and light detection accuracy and reaction times. The results indicated that as the amplitude of the LV’s speed changes increased (higher difficulty), the number of saccades significantly decreased. Contrary to the "look but failed to see" hypothesis, there was no evidence of a shrinking UFOV; instead, drivers were more accurate at detecting light changes in the higher amplitude condition. This improved accuracy correlated with fewer saccades, suggesting that reducing eye movements allowed for more efficient processing of visual information. There were no significant differences in fixation duration, number of fixations, pupil size, or blinks. Car-following performance remained stable across conditions, indicating that drivers had sufficient attentional resources to maintain driving standards while adapting their visual strategy. The findings support the interpretation that drivers adapt their scan strategy by reducing saccadic activity under higher workload conditions, rather than experiencing cognitive tunneling or a reduced UFOV. This reduction in saccades appears to optimize performance by minimizing the visual information loss that occurs during eye movements. The study concludes that fewer saccades reflect increased concentration and efficient resource allocation, challenging the assumption that increased workload necessarily degrades peripheral detection performance. These results imply that eye movement patterns, particularly saccade frequency, can serve as valid indicators of driving workload and adaptive visual strategies.

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.

• eye movements scanning
• useful field of view
• peripheral attention
• temporal
• inattentional change blindness
• road complexity

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: behavioral performance data
• Methodological Resource: measurement protocol
• Theoretical Contribution: theory or model