Understanding User Needs in Automated Vehicle Explanations: A Qualitative Approach
DOI: 10.1177/10711813251364801
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 users perceive and evaluate different types of explanations provided by automated vehicles (AVs), aiming to address the gap in understanding user-driven design preferences for AV transparency. While prior research has established that explanation content influences user outcomes, less is known about why these effects occur or how users interpret specific explanation types. The authors sought to answer two primary questions: how users perceive different explanation contents and what design improvements they would suggest to enhance understanding, interaction quality, and system acceptance. The researchers employed a qualitative approach using semi-structured interviews with 17 licensed drivers in the United States. Participants were randomly assigned to one of four conditions: no explanation, action-only ("what"), reasoning-only ("why"), or combined action and reasoning ("what + why"). Each participant viewed three short, first-person AV driving videos depicting various scenarios (urban, highway, rural) corresponding to their assigned condition. Following the video viewing, participants engaged in 40–50 minute interviews probing their understanding, emotional reactions, trust, and suggestions for improvement. Data were analyzed using thematic analysis in NVivo, with validity strengthened through triangulation and peer debriefing. The findings revealed distinct user responses across conditions. Participants in the no-explanation condition reported significant anxiety, confusion, and fear, often questioning the vehicle's reliability and environmental awareness. Action-only explanations improved predictability and transparency but left users uncertain about the AV's intent, sometimes increasing cognitive load as they tried to infer reasons. Reasoning-only explanations enhanced environmental awareness and confidence in the AV's logic but required users to mentally deduce the specific action, leading to increased cognitive effort and occasional uncertainty. The combined "what + why" explanations were generally preferred, offering the best balance of transparency and cognitive ease, though some participants noted a risk of information overload in complex environments. Participants provided actionable recommendations for improving AV explanations, emphasizing the need for multimodal delivery (combining audio with visual cues like maps or arrows) to reduce cognitive load. They also suggested including additional contextual details, such as potential consequences (e.g., travel time changes), spatial specificity, and system priorities (e.g., safety vs. efficiency). Furthermore, the voice characteristics of the AV, including tone, naturalness, and accent, significantly influenced user trust and comfort, with a preference for natural, conversational speech. The study concludes that effective AV explanations must be adaptable and context-sensitive, addressing both cognitive and emotional needs to foster long-term trust and acceptance.
Key finding
Combined action and reasoning explanations were perceived as the most effective format for enhancing user predictability and trust, although they carried a risk of information overload.
Methodology
survey
Sample size: 17
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 unpaywall on 2026-05-07 (4 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | — | — | — | 1 | 2026-05-06 |
| archive | success | canonical_url | — | — | 10 | 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 | openalex | — | — | 2 | 2026-05-08 |
| promote | success | — | — | — | 1 | 2026-05-06 |
| 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