Driver Expectations for System Control Errors, Driver Engagement, and Crash Avoidance in Level 2 Driving Automation Systems
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study investigates how driver expectations regarding SAE Level 2 driving automation systems influence driver engagement, response to system errors, and crash avoidance behaviors. The research was motivated by the increasing availability of vehicles with lateral (lane centering/keeping) and longitudinal (adaptive cruise control) automation, where capabilities vary widely among manufacturers. Drivers often form expectations through indirect sources like online videos or dealership training, which may not align with actual system limitations, potentially leading to overreliance or disengagement. The primary objective was to determine if manipulating driver expectations independently from actual system capabilities would alter how drivers interact with these technologies. The researchers employed a mixed experimental design involving 96 participants recruited from Blacksburg, Virginia. Participants drove a modified 2015 Infiniti Q50 equipped with adaptive cruise control and a customizable lateral control feature. The study utilized a four-condition between-subjects design that manipulated the congruence between participant training and vehicle capability. Training was either congruent or incongruent with the lateral feature’s actual performance (high or low capability), creating scenarios where expectations matched, exceeded, or fell short of system abilities. Participants first drove on public roads, then proceeded to the Virginia Smart Road test track. On the track, they performed three non-driving tasks (texting, watching video, and baseline driving) in varying orders. Midway through the session, participants experienced a surprise crash-imminent event, either a lane departure due to lateral feature failure or an imminent forward crash due to longitudinal limitations. Data were collected using a data acquisition system to measure hands-on-wheel behavior, eyes-off-road time, and response times, alongside subjective measures of trust and acceptance. The results indicated that driver training significantly influenced engagement behaviors on public roads, independent of the vehicle’s actual capabilities. Drivers who received high-expectation training were more likely to keep their hands off the steering wheel compared to those with low-expectation training, regardless of whether the vehicle had high or low lateral capability. This disengagement persisted immediately prior to the surprise events. However, experiencing the surprise event caused a significant shift in behavior; participants in all conditions increased their hands-on-wheel and eyes-on-road engagement after the incident. There were no significant effects of training or capability on response times to the surprise events. Notably, during the forward crash scenario, multiple drivers across all conditions did not brake, instead relying on the adaptive cruise control to slow the vehicle, reporting they wanted to "see what the car would do." Subjective trust ratings remained high and increased throughout the experiment, showing no statistically significant change before and after the surprise events. The study concludes that pre-conceived expectations significantly shape driver engagement with Level 2 automation, often leading to disengagement that is potentially detrimental to safety. Because Level 2 systems require the driver to remain attentive and ready to intervene, any disengagement poses a risk. The findings suggest that accurate communication of system capabilities and limitations is crucial; if drivers expect less than the system can do, they may underutilize it, but if they expect more, they may disengage. The research highlights the need for training and human-machine interfaces that foster "appropriate reliance," ensuring drivers understand the specific boundaries of automated features to maintain adequate engagement and readiness to respond to system failures.
Key finding
Driver training congruent with system capabilities reduced hands-off-wheel behavior on public roads compared to high-expectation training, while surprise events universally increased driver engagement across all conditions.
Methodology
mixed_methods
Sample size: 96
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 bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 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: conceptual framework, computational model