Driving behavior model considering driver's over-trust in driving automation system

Liu, Hailong; Hiraoka, Toshihiro · 2019 · Unknown

DOI: 10.1145/3349263.3351525

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 addresses the critical safety issue of driver over-trust in Driving Automation Systems (DAS), particularly within SAE Levels 1–3 where human supervision remains required. As DAS functionality improves, drivers often experience degraded skills and situation awareness, leading to over-trust—a state where a driver trusts the system despite its inability to handle specific driving tasks. This phenomenon, illustrated by fatal accidents involving Level 2 systems, stems from misconceptions regarding the system’s foundation, purpose, process, or performance. The study aims to define over-trust, hypothesize its occurrence conditions, and propose a driving behavior model that incorporates an over-trust prevention Human-Machine Interface (HMI). The authors conceptualize trust as a psychological activity and reliance as its behavioral manifestation. They define over-trust through two conditions: the driver trusts the DAS, and the DAS cannot respond to the driving task. Because over-trust is unobservable in real-time by either the driver or the system, the paper proposes a theoretical framework based on mental models and risk homeostasis theory. The authors extend a manual driving behavior model to include DAS interactions. In this model, drivers form mental models of the DAS through experience, comparing predicted system states with actual feedback from the HMI to adjust their trust levels. Risk perception is determined by combining perceived environmental hazards, self-recognized driving skills, and trust in the DAS. The study introduces a comprehensive driving behavior model for vehicles equipped with DAS and an over-trust prevention HMI. This model integrates an "over-trust inference model" that estimates the driver’s trust state and predicts whether the DAS can handle the current situation. By monitoring driver physiological and motion data alongside DAS operational states, the system can infer over-trust before dangerous events occur. The proposed HMI provides reference results to correct driver misconceptions, thereby adjusting the trust state to appropriate levels. The model posits that preventing over-trust requires reducing misconceptions about the four dimensions of trust (foundation, purpose, process, performance) through effective information provision. The significance of this work lies in providing a structured theoretical basis for designing HMIs that mitigate over-trust in semi-autonomous vehicles. By linking mental models, trust dynamics, and risk homeostasis, the paper offers a mechanism for real-time estimation and prevention of over-trust. This approach addresses the limitations of current DAS, where drivers often fail to recognize system boundaries, leading to dangerous reliance. The proposed model suggests that integrating trust estimation and corrective HMI feedback can enhance safety by ensuring drivers maintain appropriate vigilance and understanding of system capabilities, ultimately supporting the safe deployment of Levels 1–3 automation.

Key finding

The study proposes a theoretical driving behavior model that integrates driver trust, mental models, and risk homeostasis to explain and prevent over-trust in driving automation systems.

Methodology

theoretical

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 author_sweep_intake on 2026-05-28.

StageOutcomeToolModelPromptAttemptsCompleted
discover success author_sweep 2 2026-05-28
archive success canonical_url 1 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 1 2026-05-28
promote success 1 2026-06-04
summarize success llm qwen3.6-27b-prismaquant summ-v5 2 2026-06-10
tag success vector_similarity 15 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).