Evaluating Effects of Enhanced Autonomy Transparency on Trust, Dependence, and Human-Autonomy Team Performance over Time
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 study addresses the challenge of human-autonomy team performance by examining how enhanced autonomy transparency affects trust, dependence, and performance over time. Prior research often measured trust as a static "snapshot" at the end of experiments, overlooking the temporal dynamics of human-agent interaction. The authors aim to fill this gap by investigating how an "option-centric rationale display"—which explicitly shows the option space and the reasoning behind an autonomous agent’s recommendations—impacts these variables during repeated interactions. The researchers conducted a human-in-the-loop experiment using a simulation called "Treasure Hunter," adapted from the Wumpus world game. Participants acted as hunters navigating a map to find gold while avoiding pits and a wumpus, aided by an intelligent assistant that used propositional logic to recommend next moves. The study employed a within-subjects design with 28 participants (after excluding outliers and technical failures). Participants completed five maps with the rationale display and five without. The display detailed all possible next locations, the probabilities of outcomes (e.g., finding gold, hitting a pit), and expected scores, whereas the control condition showed only a red star indicating the recommendation. Trust and self-confidence were measured after each map using Likert scales, while recommendation acceptance and task scores were recorded automatically. Results indicated a significant interaction between the presence of the display and time regarding trust. When participants had access to the rationale display, their trust in the intelligent assistant increased significantly over trials and became more calibrated to the agent’s near-optimal capability. In contrast, trust remained stagnant without the display. The display also led to a significantly higher acceptance rate of the agent’s recommendations, although this effect did not change significantly over time, likely due to a ceiling effect where acceptance was already near 100%. Surprisingly, neither the display nor time had a significant effect on overall task performance scores. The authors attribute this lack of performance difference to a ceiling effect and the inherent stochasticity of the game, where even optimal strategies could occasionally yield unfavorable outcomes. The findings suggest that enhancing autonomy transparency through explicit rationale displays improves trust calibration and dependence behaviors over time, particularly in cognitively demanding tasks. While the study was limited to a simulated environment with a small action space and a highly capable agent, the authors argue that the display design can be generalized to larger action spaces, such as those involving exploration-exploitation trade-offs. The study highlights the importance of viewing trust as a dynamic variable and demonstrates that transparency mechanisms can help humans better understand and rely on autonomous systems during prolonged interactions.
Key finding
Providing an option-centric rationale display that explains an autonomous agent's decision-making logic significantly increases user trust and recommendation acceptance over time, though it does not necessarily improve overall task performance.
Methodology
simulator
Sample size: 28
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 openalex_abstract on 2026-05-08 (2 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | — | — | — | 1 | 2026-05-05 |
| archive | success | canonical_url | — | — | 12 | 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 | 2026-05-05 |
| 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: behavioral performance data
- Theoretical Contribution: theory or model