Training to Improve Drivers’ Behavior When Partial Driving Automation Fails [supporting datasets]
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 document serves as a metadata record and dataset description for the study "Training to Improve Drivers’ Behavior When Partial Driving Automation Fails," authored by Shannon C. Roberts and Yalda Ebadi of the University of Massachusetts Amherst. The research addresses the critical safety challenge posed by Level 2 (L2) partial driving automation systems. As these systems shift the driver’s role to a supervisory one, drivers must accurately perceive the system’s Operational Design Domain (ODD) limits and effectively regain control when the automation fails or reaches its boundaries. The study was motivated by the need to enhance driver situational awareness and takeover performance in these specific scenarios, funded by the U.S. Department of Transportation. The researchers designed and tested a PC-based training program intended to improve drivers’ situational awareness regarding ODD constraints and their performance during transfer-of-control situations. The experimental design compared the efficacy of this PC-based training against two control conditions: standard user manual training and placebo training. The study focused on measuring how effectively drivers could take back control when L2 systems encountered their operational limits. The associated dataset, preserved in the SAFER-SIM Dataverse within the Harvard Dataverse repository, contains two primary CSV files: `SS2-SART-Data.csv` and `SS2-Takeover-response.csv`. These files store tabular data related to the study’s metrics, likely including Situational Awareness Rating Technique (SART) scores and takeover response times, facilitating further analysis of the experimental outcomes. The findings indicated that the PC-based training program was significantly more effective than the alternative methods. Drivers who underwent the PC-based training demonstrated superior ability to take back control when L2 systems reached their ODD limits. Furthermore, these drivers exhibited higher levels of situational awareness compared to those who received only user manual instructions or placebo training. This evidence suggests that interactive, computer-based training can better prepare drivers for the supervisory demands of partial automation than traditional informational materials. The significance of this work lies in its contribution to the development of effective training protocols for automated vehicle users. By demonstrating that specific training interventions can improve critical safety behaviors—namely, timely and effective takeover and accurate situational awareness—the study provides actionable insights for vehicle manufacturers and regulatory bodies. As L2 systems become more prevalent, ensuring drivers can reliably manage the transition from automated to manual control is essential for road safety. The availability of the underlying data supports transparency and allows for further validation or extension of these findings within the field of transportation engineering and human factors research.
Key finding
Drivers given PC-based training regained control more effectively at L2 ODD limits and reported higher situational awareness than those given a user manual or placebo training.
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 (7 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 | — | — | — | 3 | 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.
- automation
- automation surprise
- takeover transitions
- driverless ads
- automation complacency bias
- situational 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).
- Methodological Resource: dataset resource, tool software
- Theoretical Contribution: conceptual framework