Context-Adaptive Management of Drivers’ Trust in Automated Vehicles
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the critical challenge of trust miscalibration in automated vehicles (AVs), specifically within SAE Level 3 systems where drivers must collaborate with automation. Trust miscalibration—defined as a mismatch between a driver’s trust and the AV’s actual capabilities—manifests as overtrust (leading to misuse) or undertrust (leading to disuse), both of which compromise safety and performance. While prior research has focused on estimating trust, this study introduces a novel framework for the active management and recalibration of driver trust in real time. The authors aim to identify miscalibrations and deploy context-adaptive communication strategies to align driver trust with system capabilities, thereby mitigating unsafe driving scenarios. The methodology integrates a Kalman filter-based trust estimator with a rule-based trust calibrator. The estimator infers driver trust levels from behavioral cues, including eye-tracking data (focus on non-driving-related tasks), ADS usage rates, and task performance. The calibrator compares these trust estimates against the AV’s current capability level, which varies based on road difficulty (straight, curvy, or dirt roads). When a mismatch is detected, the AV adjusts its verbal communication style to influence the driver’s situation awareness and risk perception. For instance, the system encourages drivers to focus on secondary tasks when undertrusting or issues harsh warnings demanding attention when overtrusting. The framework was validated through a user study with 40 participants operating an AV simulator while performing a visual search task. Participants completed trials with and without the trust calibrator active, allowing for a controlled assessment of the intervention’s effectiveness. The results demonstrate that the proposed framework effectively manages driver trust. The system successfully increased trust levels in undertrusting drivers and decreased trust levels in overtrusting drivers, bringing them into alignment with the AV’s capabilities. Quantitatively, the framework reduced the average time periods during which trust was miscalibrated by approximately 40%. Statistical analysis confirmed that the communication styles significantly influenced trust differences, validating the mechanism’s ability to recalibrate trust in real time. The study confirms that manipulating situation awareness and risk perception through targeted verbal messages is a viable strategy for trust calibration. The significance of this work lies in its contribution to human-robot interaction and automated vehicle safety. By providing a functional method for real-time trust calibration, the framework addresses a gap in existing literature that has largely neglected active trust manipulation. The findings suggest that context-adaptive communication can enhance the performance and safety of driver-AV teams by preventing the misuse and disuse associated with trust miscalibration. This approach offers a practical pathway for designing SAE Level 3 systems that can dynamically adapt to human factors, ensuring that drivers maintain an appropriate level of reliance on automation across varying driving conditions.
Key finding
The proposed context-adaptive trust management framework effectively recalibrated driver trust levels and reduced the average time periods of trust miscalibration by approximately 40%.
Methodology
simulator
Sample size: 40
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 2026-05-27.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-27 |
| archive | success | unpaywall | — | — | 2 | 2026-06-04 |
| 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 | 2026-05-27 |
| 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 | partial | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- trust calibration
- trust in automation foundations
- automation
- automation surprise
- acceptance adoption
- 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).
- Empirical Findings: self report data
- Methodological Resource: validation psychometrics
- Theoretical Contribution: computational model