The Induction and Detection Method of Angry Driving: Evidences from EEG and Physiological Signals

Yan, Lixin; Wan, Ping; Qin, Lingqiao; Zhu, Dunyao · 2018 · Crossref

DOI: 10.1155/2018/3702795

archive: archived pipeline: cataloged

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical road safety issue of angry driving, which is a prevalent factor in traffic accidents, particularly in China where 60.7% of surveyed drivers report experiencing anger while driving. The research aims to develop effective methods for inducing driving anger in controlled settings and detecting it in real-world conditions using physiological and brain signals. The motivation stems from the limitations of traditional emotion induction methods, which often lack ecological validity, and the need for objective, continuous monitoring tools to distinguish between normal and angry driving states. The research employed a two-phase experimental design. The first phase utilized a driving simulator to test three specific anger-induction scenarios: frequent red light waiting, traffic congestion, and adjacent vehicle interference. Fifteen participants completed paired control and induced scenarios for each condition. Data collected included self-reported anger levels (using a 5-point scale), biosignals (blood volume pulse [BVP] and skin conductance [SC]), and EEG features (delta wave percentage [δ%] and beta wave percentage [β%]). The second phase involved 22 on-road experiments to validate detection methods in real-life driving. A Hidden Naïve Bayes (HNB) classifier was developed to detect angry driving based on four features: BVP, SC, δ%, and β%. Results from the simulator experiment confirmed that the developed scenarios effectively induced anger. The "adjacent vehicle interference" scenario yielded the highest emotional differentiation degree (87% hit rate for anger) and the highest mean anger level (2.625 on a 0–4 scale), significantly higher than control conditions (p < 0.05). Physiological analysis showed that BVP and SC increased significantly in induced scenarios compared to controls, while β% exhibited greater instability (higher standard deviation) during anger induction. In the on-road detection phase, the HNB classifier achieved an accuracy of 85.0% in distinguishing between normal and angry driving states based on the selected biosignal and EEG features. The study concludes that driving simulator-based induction is a safe and effective alternative to risky on-road experiments for exploring the causal links between anger, unsafe behavior, and accidents. The proposed detection model provides a theoretical foundation for developing real-time driving anger warning systems. By combining objective physiological markers with validated induction scenarios, the research offers a robust framework for future intelligent transportation systems aimed at mitigating road rage-related incidents.

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 qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

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

• anger road rage

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: physiological data
• Methodological Resource: validation psychometrics, tool software