Human-Machine Interfaces and Vehicle Automation: The Effect of HMI Design on Driver Performance and Behavior

AAA Foundation for Traffic Safety · 2023 · AAA Foundation for Traffic Safety

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 investigates how different Human-Machine Interface (HMI) designs affect driver performance and behavior during takeover scenarios in automated vehicles. As driving automation becomes more prevalent, systems must effectively alert drivers to return attention to the road or assume control when automation limits are reached. The research, conducted under a cooperative agreement between the AAA Foundation for Traffic Safety and the SAFER-SIM University Transportation Center, aimed to evaluate driver responses to two distinct HMI configurations designed for various levels of automation. The primary goal was to determine how specific auditory, visual, and timing elements of warnings influence takeover time, driving performance, eye-glance behavior, and subjective trust. The methodology involved 54 participants aged 18 to 40 with valid U.S. driver’s licenses. Participants used a high-fidelity driving simulator featuring a 2013 Ford Fusion and a 330-degree field of view. The study employed a between-subjects design where drivers encountered four driving scenarios corresponding to automation Levels 0 through 3. In each scenario, the automation approached the edge of its operational design domain, requiring a transfer of control. During the Level 3 scenario, participants engaged in a non-driving-related task (watching a video) to simulate real-world distraction. Drivers were assigned to one of two HMI conditions: "Staged HMI," which provided a visual text warning followed by a non-descriptive auditory beep five seconds later; or "Simultaneous HMI," which presented concurrent visual (red textbox with pictogram) and auditory (imperative voice command) warnings that varied by automation level and provided explicit guidance on the required action. The results indicated that the Simultaneous HMI yielded nominally shorter takeover response times across all automation levels compared to the Staged HMI, though these differences did not reach statistical significance. However, driving performance metrics revealed that drivers using the Staged HMI exhibited greater variability in velocity and higher maximum deceleration after Level 1 takeover events, suggesting that the Simultaneous HMI facilitated smoother control transitions. Regarding eye-glance behavior, the Simultaneous HMI resulted in fewer off-road glances and fewer glances to the instrument cluster across all automation levels, with the effect reaching statistical significance only for Level 3. There were no significant differences in glance duration or the ratio of long glances. Additionally, the HMI design had no effect on subjective ratings of system usability or trust, indicating equivalent levels of user perception for both interfaces. The study concludes that while the Simultaneous HMI offered certain behavioral advantages, such as smoother vehicle control and reduced off-road attention, the benefits were not consistent across all measures. The findings highlight the importance of grounding HMI design in human factors principles. Specifically, providing direct, clear guidance through simultaneous multimodal alerts may improve driver response quality, whereas staged alerts may lead to more erratic driving behaviors during takeover. These results underscore the need for rigorous evaluation of HMI designs to ensure they effectively support safe transitions of control in automated driving systems.

Key finding

Drivers exposed to a simultaneous HMI design exhibited smoother control transitions with reduced velocity variability and fewer off-road glances compared to those using a staged HMI, although takeover response times did not differ significantly.

Methodology

simulator

Sample size: 54

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 bulk_ingest_aaa_foundation on 2026-05-23 (6 acquisition events logged).

StageOutcomeToolModelPromptAttemptsCompleted
discover success aaa_foundation 2 2026-05-23
archive success 1 2026-05-23
extract success cached 2 2026-06-10
clean success 1 2026-06-01
chunk success 1 2026-06-01
embed success 1 2026-06-02
enrich success 1 2026-05-23
promote success 1 2026-05-23
summarize success llm qwen3.6-27b-prismaquant summ-v5 3 2026-06-10
tag success vector_similarity 19 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.

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).