Participants' speed-accuracy trade-off behavior in high-stress situations in simulator studies

Usai, Marcel; Schäfer, Philip; Herzberger, Nicolas Daniel; Flemisch, Frank · 2024 · Crossref

DOI: 10.54941/ahfe1004470

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 investigates the phenomenon of speed-accuracy trade-offs in human decision-making within high-stress, time-critical environments. The authors address the problem that participants in simulator studies often prioritize reaction speed over accuracy, ignoring explicit visual or auditory cues provided by the system. This behavior can lead to preventable task failures, creating a mismatch between the system’s required safety standards and the user’s actual behavior. The study aims to describe this phenomenon and raise questions about how human-machine systems can be designed to balance reaction time and quality, and whether AI co-systems can detect and intervene when humans make incorrect trade-offs. The research presents findings from two distinct simulator studies. The first study involved 24 participants in a driving simulator testing takeover requests (TOR) in highly automated driving scenarios. Participants faced two critical situations: an unregulated four-way crossing and a breakdown vehicle on a highway. They were divided into three groups based on different TOR interaction designs, with the third group receiving extensive visual cues, such as red "walls" indicating danger and green arrows indicating safe lanes. The second study involved eight participants in a virtual reality (VR) warehouse environment performing a search-and-collect task under time pressure. This study used a within-subject design to test the impact of visual (pulsing red overlay) and auditory (warning sounds) cues triggered when participants strayed from a predefined path. In the driving simulator, results showed that while the most advanced design (Design 3) generally reduced failures, three participants still collided with traffic by changing lanes to the left despite clear visual warnings against it. These participants either ignored, failed to recognize, or misunderstood the cues within the critical timeframe. In the VR study, statistical analysis revealed no significant difference in time spent off-path between the baseline and conditioning scenarios. However, a significant difference was found between the conditioning and control scenarios, indicating that participants adjusted their behavior after experiencing the warnings. Interviews revealed varied perceptions; while some participants recognized the warnings and corrected their path, others prioritized speed over comfort or failed to discern the cause of the warnings. The authors conclude that users frequently prioritize execution speed over accuracy, sometimes consciously and sometimes unconsciously, even when clear guidance is provided. This highlights a critical challenge in human-AI cooperation: designing systems that account for individual differences in stress responses. The paper suggests that early integration of human users into the design process is essential to identify unexpected reactions. Furthermore, it proposes that AI co-systems could potentially detect incorrect speed-accuracy trade-offs in real-time by analyzing user behavior patterns, such as gaze paths, and respond with escalated warnings or adaptive interaction designs. The findings underscore the need to balance safety with user freedom and to develop systems that can dynamically adjust to human limitations under pressure.

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.

StageOutcomeToolModelPromptAttemptsCompleted
discover success Crossref 1 2026-06-06
archive success canonical_url 1 2026-06-09
extract success cached 2 2026-06-09
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
promote success 1 2026-06-06
summarize success llm qwen3.6-27b-prismaquant summ-v5 1 2026-06-09
tag success vector_similarity 15 2026-06-11
verify success 1 2026-06-09

Summary generated by qwen3.6-27b-prismaquant on 2026-06-09; 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).