Similarity may be safer: the effect of similarity between Speech-based takeover request style and driver personality on self-driving takeover performance

Ma, Keer; Wu, Jianfeng; Lin, Yanxi; Li, Zihan; Guo, Songyang; Jiao, Dongfang; Yu, Shihan · 2025 · Crossref

DOI: 10.65927/wewq5422

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 the similarity between speech-based takeover request (TOR) styles and driver personality traits affects takeover performance in Level 3 automated driving. While prior research has examined TOR modality and timing, little is known about how speech style interacts with driver personality, particularly regarding the "similarity–attraction" effect. The authors hypothesized that TORs matching a driver’s dominant-submissive personality tendencies would improve attention redirection and takeover stability. The study also examined how scenario urgency (low vs. high) interacts with these personality similarities, addressing inconsistent findings in previous literature regarding urgency’s impact on performance. The researchers conducted a driving simulator experiment with 49 participants using a 2 × 2 within-subjects design. Independent variables were personality similarity (similar vs. dissimilar speech style) and scenario urgency (low: road construction; high: traffic accident). Participants were classified as dominant or submissive based on the IAS-R personality scale. Speech-based TORs were synthesized using text-to-speech technology, manipulating fundamental frequency to create dominant (82.3 Hz) and submissive (168.3 Hz) voices. To control for confounding variables, the takeover time budget was fixed at 7 seconds across all conditions. Participants performed non-driving-related tasks (playing the game "2048") during automated driving. Metrics included attention redirection time, takeover time, longitudinal and lateral control stability, eye-movement data, and subjective evaluations. Results indicated that personality similarity significantly reduced attention redirection time, with participants responding faster when the TOR style matched their personality tendencies. However, similarity did not affect the overall takeover initiation time. Scenario urgency significantly influenced takeover speed, with high-urgency scenarios eliciting faster responses than low-urgency ones. Crucially, while high urgency generally reduced takeover stability, personality-similar TORs mitigated this deterioration, leading to more stable vehicle control. Subjective evaluations also favored similar speech styles, reporting higher usefulness and satisfaction. The findings suggest that aligning speech-based TOR styles with driver personality traits can enhance safety by accelerating attention redirection and maintaining control stability, particularly in high-urgency situations. This supports the integration of personality-matching interfaces in automated driving systems to optimize human-machine interaction. The study contributes to the field by validating the "similarity–attraction" effect in automated driving contexts and highlighting the importance of considering individual differences in HMI design to improve takeover safety and user experience.

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.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• takeover transitions
• temporal
• risk taking
• automation

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
• Methodological Resource: measurement protocol
• Theoretical Contribution: conceptual framework