Underload on the Road: Measuring Vigilance Decrements During Partially Automated Driving
DOI: 10.3389/fpsyg.2021.631364
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This narrative review addresses the safety risks associated with partially automated driving (PAD), specifically focusing on vigilance decrements caused by cognitive underload. As vehicles with SAE levels 1–3 automation become common, drivers are required to monitor the system rather than actively control the vehicle. This creates a monotonous, low-demand environment where drivers experience cognitive underload, leading to passive fatigue and mind-wandering. These states impair the driver’s ability to detect and respond to critical events, posing significant safety risks when the automation fails. The paper aims to characterize the methodologies used to measure these vigilance decrements, highlighting their advantages and limitations to guide future research. The authors conducted a comprehensive review of existing literature, analyzing studies that measured driver vigilance during PAD. They categorized measurement tools into two primary types: offline measures and online measures. Offline measures, such as the NASA Task Load Index (NASA-TLX) and the Dundee Stress State Questionnaire (DSSQ), rely on self-reported data collected before or after driving sessions to assess workload and engagement. Online measures, collected in real-time during driving, include behavioral metrics like response times to safety-critical events or detection tasks, as well as psychophysiological indicators such as eye-tracking, heart rate, and EEG. The review excluded studies involving secondary tasks or sleep-related fatigue to isolate the effects of underload. The findings indicate that offline measures consistently show lower workload and engagement scores during PAD compared to manual driving, reflecting the disengagement associated with underload. However, these self-report tools are subject to recall bias and cannot capture real-time fluctuations. Online measures provide higher temporal sensitivity, revealing that response times to critical events are significantly slower during PAD than during manual driving. Furthermore, response times degrade over the duration of a PAD session, confirming that vigilance declines over time. The review also notes that higher levels of automation (e.g., SAE Level 2 vs. Level 1) correlate with larger vigilance decrements due to reduced driver engagement. While online measures like critical event responses effectively demonstrate the consequences of vigilance loss, they require careful design to avoid inducing artificial high vigilance through frequent stimuli. The significance of this work lies in its systematic characterization of how vigilance is measured in automated driving contexts. By distinguishing between the retrospective nature of offline measures and the real-time precision of online measures, the authors provide a framework for evaluating driver states. The review underscores that cognitive underload is a critical safety factor in PAD, necessitating robust monitoring methods. The authors conclude that future research must consider the specific limitations of each measurement type and account for factors like automation level and task duration to develop effective countermeasures for vigilance decrements, ultimately enhancing transportation safety until full automation is achieved.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-17 |
| archive | success | unpaywall | — | — | 2 | 2026-06-25 |
| extract | success | cached | — | — | 2 | 2026-06-25 |
| clean | success | clean | — | — | 1 | 2026-06-18 |
| chunk | success | chunk | — | — | 1 | 2026-06-18 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-18 |
| promote | success | — | — | — | 1 | 2026-06-17 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-25 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-18 |
| verify | success | — | — | — | 1 | 2026-06-26 |
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.
- vigilance
- automation complacency bias
- situational awareness
- automation surprise
- drowsiness
- mode 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).
- Empirical Findings: physiological data
- Theoretical Contribution: theory or model, conceptual framework