A comparative simulator study of reaction times to yellow traffic light under manual and automated driving
DOI: 10.1016/j.trpro.2021.01.032
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 driver reaction times (RT) to yellow traffic lights under manual versus automated driving conditions, addressing the safety implications of automation-induced mental underload. The research was motivated by existing literature suggesting that automated vehicles increase RTs by 1–1.5 seconds compared to manual driving, potentially leading to hazardous situations such as crossing intersections on red. The authors aimed to quantify these differences and analyze how driver experience and system failure frequency impact response times. The research utilized two experiments conducted on a dynamic driving simulator at the University of Padua. Experiment 1 involved 24 participants manually driving through 40 trials approaching signalized intersections, with 32 trials featuring a yellow light onset. Experiment 2 involved 33 participants (28 valid for analysis) in an automated vehicle mode. After familiarization, participants encountered three programmed system failures where the automated vehicle failed to stop for a yellow light, requiring manual intervention. The study measured the time between the stimulus (yellow light or system failure) and the driver’s braking action. Statistical analyses employed non-parametric tests, including Kruskal-Wallis, Friedman’s, and Mann-Whitney U tests, to account for sample size and assumption violations. Results from Experiment 1 indicated that manual driving RTs were significantly influenced by driver habits. Daily drivers exhibited lower median RTs (1.64s) compared to non-daily drivers (2.02s), and those accustomed to suburban/highway driving reacted faster (1.59s) than those used to urban/local roads (1.96s). Surrounding traffic conditions did not significantly affect RT. In Experiment 2, RTs to automated system failures were significantly higher than manual driving RTs. Specifically, the median RT for the first system failure was 3.41s, decreasing to 2.67s by the third failure, indicating a learning effect where drivers reacted faster to subsequent failures. However, even the fastest automated failure responses remained significantly slower than manual driving responses. Driver characteristics such as daily usage and road type did not significantly impact RTs during automated failures, likely because these factors relate to manual driving experience rather than automated system interaction. The study concludes that automation leads to significantly longer reaction times, primarily due to mental underload and reduced situational awareness. The findings confirm that drivers take longer to react to critical events in automated vehicles than in manual ones, with the initial failure eliciting the slowest response. These results highlight the importance of considering automation-induced delays in traffic engineering and safety assessments, suggesting that current RT standards may be insufficient for automated vehicle scenarios. The significant difference between manual and automated RTs supports the need for further research into mitigating the effects of mental underload in automated driving systems.
Key finding
Reaction times for drivers intervening in automated vehicle failures were significantly higher than those for manual drivers, with the longest delays occurring during the first system failure.
Methodology
simulator
Sample size: 57
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. Discovered via author_sweep_intake on 2026-05-28.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-28 |
| archive | success | openalex | — | — | 9 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-04 |
| chunk | success | chunk | — | — | 1 | 2026-06-04 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-04 |
| enrich | success | — | — | — | 1 | 2026-05-28 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
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
- Theoretical Contribution: computational model