80 MPH and out-of-the-loop: Effects of real-world semi-automated driving on driver workload and arousal
archive: archived pipeline: cataloged
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates whether real-world semi-automated driving reduces driver arousal and workload, potentially leading to decreased monitoring and safety risks. Motivated by the Yerkes-Dodson law, which posits that low arousal impairs performance, and prior observational data suggesting that driver assistance systems may cause "out-of-the-loop" disengagement, the authors examined the physiological and behavioral effects of using Tesla Autopilot. The research aimed to determine if relinquishing operational control to a level-2 automated system diminishes driver engagement compared to manual driving. The experiment involved 22 participants (14 males, 8 females, aged 21–35) who drove a Tesla Model S on a 63-mile highway route in both manual and semi-automated modes. The study utilized a within-subject design with counterbalanced condition order. During the drives, researchers recorded continuous electrocardiogram (ECG) data to calculate mean heart rate and heart rate variability (specifically TINN). Behavioral performance was measured using a peripheral detection task where participants responded to a vibrotactile stimulus on their left arm. Additionally, participants self-reported their mind-wandering state on a three-point scale. Data from ten participants were excluded from physiological analyses due to recording artifacts, and four were excluded from behavioral analyses. Results indicated that semi-automated driving significantly reduced physiological activation. Mean heart rate was lower in the semi-automated condition (M = 72.12 bpm) compared to manual driving (M = 75.03 bpm), t(11) = 3.28, p < .05. Conversely, heart rate variability (TINN) was higher in the semi-automated mode (M = 468 ms) than in manual mode (M = 413.33 ms), suggesting increased parasympathetic activity associated with relaxation. Behaviorally, participants responded significantly slower to the peripheral detection task in semi-automated mode (M = 1068 ms) than in manual mode (M = 879 ms), t(17) = 2.69, p < .05. No significant differences were found in respiration rates or self-reported mind-wandering between the two conditions. These findings suggest that semi-automated driving induces a state of under-arousal, characterized by lower heart rate and increased heart rate variability, which correlates with slower reaction times to peripheral stimuli. The discrepancy between objective physiological/behavioral signs of disengagement and the lack of self-reported mind-wandering implies that drivers may lack accurate self-awareness of their declining vigilance. The authors conclude that semi-automated systems may not mitigate safety risks from human error but could instead exacerbate them by causing driver disengagement. This highlights a critical safety concern where reduced workload leads to decreased monitoring, potentially resulting in delayed responses to traffic hazards, a pattern consistent with recent accidents involving automated vehicle control systems.
Key finding
Relative to manual driving, semi-automated driving (Tesla Autopilot with Autosteer) on a 63-mile interstate route lowered physiological arousal (heart rate 72.12 vs 75.03 bpm) and slowed peripheral detection response time, with more self-reported mind-wandering — an out-of-the-loop pattern at 70-80 mph.
Methodology
on_road
Sample size: 22
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 | — | — | — | 1 | 2026-05-06 |
| archive | success | owner_recovery | — | — | 24 | 2026-08-07 |
| extract | success | cached | — | — | 6 | 2026-08-23 |
| clean | success | clean | — | — | 3 | 2026-08-10 |
| chunk | success | chunk | — | — | 3 | 2026-08-10 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 3 | 2026-08-10 |
| enrich | success | — | — | — | 1 | 2026-05-06 |
| promote | success | — | — | — | 1 | 2026-05-06 |
| summarize | success | llm | qwen3.8-27b-gittensor | summ-v5 | 3 | 2026-08-23 |
| tag | success | vector_similarity | — | — | 32 | 2026-08-11 |
| verify | success | — | — | — | 2 | 2026-08-09 |
Summary generated by qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- situational awareness
- hands on hands off engagement
- automation surprise
- mode awareness
- automation complacency bias
- automation
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