Automated or human: Which driver wins the race for the passengers’ trust? Examining passenger trust in human-driven and automated vehicles following a dangerous situation
DOI: 10.1016/j.chb.2024.108387
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates how passenger trust develops in automated vehicles (AVs) compared to human-driven taxis, particularly following a dangerous driving incident. The research is motivated by the critical need for "calibrated trust"—a balanced alignment between user confidence and system capability—to ensure the safe adoption of AV technology. While AVs offer significant societal benefits, their success depends on passengers trusting them appropriately, neither under-trusting (leading to disuse) nor over-trusting (leading to misuse). The authors aimed to determine if trust dynamics differ between human and automated drivers and whether trust can recover after a critical error. The researchers employed a between-subjects experimental design using a driving simulator. Forty-one participants were randomly assigned to either a "Human" group (simulating a ride in a human-driven taxi) or an "AV" group (simulating a ride in a self-driving vehicle). Participants viewed a 17-minute dash-cam video of a drive through Madrid. Trust was measured continuously via verbal ratings on a 1–7 Likert scale triggered by beep sounds every 45 seconds. At approximately 3 minutes and 46 seconds into the simulation, the vehicle ran a red light, and either a human or robotic voice apologized for the error. Dispositional (pre-simulation) and learned (post-simulation) trust were assessed using a modified Trust in Automation questionnaire. Self-esteem was also measured to test for correlations with trust levels. The results revealed that pre-simulation trust was significantly higher for human drivers than for AVs, but this difference disappeared after the simulation, indicating that learned trust converged between the two conditions. During the simulation, trust dynamics were comparable for both groups: trust levels dropped significantly immediately following the dangerous incident (running the red light) and remained low until approximately 11 time points into the drive, after which trust began to recover. Notably, the recovery process was unstable, with fluctuations persisting even after the initial recovery phase. Contrary to prior literature suggesting a link between personality traits and automation trust, this study found no significant relationship between self-esteem and trust in either condition. The findings suggest that passengers react similarly to errors regardless of whether the driver is human or automated, initiating a process of trust calibration based on performance. The comparable evolution of trust in both conditions implies that AVs can achieve trust levels similar to human drivers if they demonstrate consistent, error-free performance after a mistake. This highlights that trust in AVs is highly susceptible to driving style and specific incidents rather than being inherently lower due to the lack of a human driver. The study underscores the importance of understanding dynamic trust development to facilitate the safe integration of AVs into public transportation systems.
Key finding
Passenger trust in automated vehicles and human-driven taxis follows a comparable dynamic pattern, dropping after a dangerous incident and recovering thereafter, with initial differences in trust levels disappearing after the experience.
Methodology
simulator
Sample size: 41
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 | openalex | — | — | 9 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| enrich | success | openalex | — | — | 4 | 2026-07-02 |
| 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 | 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: self report data
- Methodological Resource: validation psychometrics