Using Trust in Automation to Enhance Driver-(Semi)Autonomous Vehicle Interaction and Improve Team Performance
DOI: 10.2139/ssrn.3859704
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 issue of trust miscalibration in driver-automated driving system (ADS) interactions, where drivers either undertrust or overtrust automation capabilities, jeopardizing safety and performance. The authors aim to model the dynamics of driver trust development during short-term interactions, specifically examining how risk factors such as false alarms and misses influence trust levels. By characterizing these dynamics, the study seeks to enable the design of ADSs that can adapt their behavior to maintain appropriate trust levels, thereby enhancing human-robot teaming effectiveness. To achieve this, the researchers developed a state-space model for trust dynamics using linear mixed effects models. The model treats driver trust as a state variable influenced by interaction events: true alarms, false alarms, and misses. Observations used to estimate trust included visual focus (measured via eye-tracking), non-driving-related task (NDRT) performance, and automation usage time. Data were collected from 80 participants in a driving simulator experiment involving straight and curvy road trials, each containing 12 interaction events. Participants self-reported their trust using Muir’s trust questionnaire, scaled from 1 to 100, while real-time behavioral metrics were recorded. The identified parameters were used to design a trust estimator based on a linear Kalman filter, capable of providing real-time estimates of trust levels. The results demonstrate that the proposed method can successfully track self-reported trust levels from the observed behavioral variables. However, the accuracy of the estimation is highly dependent on the precision of the initial trust estimate. In cases where the initial estimate was inaccurate, the estimator required more interactions to converge, as trust development is a slow process and the model’s noise terms—capturing individual behavioral differences—had large variances. These large variances caused the Kalman filter to behave conservatively, slowing convergence when observations were noisy. The study suggests that longer-term, individualized training processes could improve initial estimates and tune model parameters, thereby enhancing estimation accuracy. The significance of this work lies in its contribution to the development of adaptive ADSs that can monitor and adjust driver trust in real time. By avoiding trust miscalibration, such systems could improve safety and performance in semi-autonomous driving contexts. The findings underscore the importance of understanding trust dynamics in human-robot interactions and provide a foundational model for future research on context-adaptive trust management in automated vehicles.
Key finding
A state-space model using a linear Kalman filter can successfully estimate and track drivers' trust dynamics in automated driving systems based on observable behavioral metrics, though convergence is sensitive to initial estimates and individual noise.
Methodology
simulator
Sample size: 80
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 (2 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 3 | 2026-05-28 |
| archive | success | canonical_url | — | — | 5 | 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 | semantic_scholar | — | — | 2 | 2026-06-04 |
| 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
- 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).
- Empirical Findings: self report data
- Theoretical Contribution: computational model