An EEG dataset for understanding driving expertise from naturalistic urban road experiments.

Gong J; Yu Y; Cao Y; Yang R; Chang X; Tang H; Zheng X; Liu Y; You S; Zheng C; Zhou G · 2026 · PubMed Central

DOI: 10.1038/s41597-026-07223-1

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 introduces a comprehensive multimodal EEG dataset designed to investigate the neural signatures of driving expertise in naturalistic urban environments. The research addresses a critical gap in autonomous driving development: while expert human drivers demonstrate superior safety, comfort, and adaptability, the underlying cognitive and neural mechanisms enabling these skills remain poorly understood. Existing datasets are largely limited to simulated environments or focus on isolated driver states like fatigue, lacking real-world comparisons between expert and novice drivers. By decoding these neural differences, the authors aim to provide benchmarks for developing more human-like, intelligent autonomous driving algorithms. The study employed a between-subject design involving 20 groups of participants, comprising 10 expert professional drivers and 10 novice drivers, along with two passengers per trip. Data collection occurred on a fixed 5.7-kilometer urban route featuring 13 distinct driving conditions, such as turns, merges, and straightaways. The experimental setup integrated synchronized multimodal sensors, including a 64-channel EEG system for brain activity, Tobii Pro Glasses for eye tracking, and Empatica E4 wristbands for physiological measures like electrodermal activity and heart rate. Vehicle dynamics were recorded via CAN bus data (speed, acceleration, steering), while external traffic conditions were captured using a 360-degree panoramic camera and processed with YOLOv8 to identify traffic participants. To validate driving performance quality, the dataset uniquely includes physiological and subjective feedback from passengers. All participants completed pre- and post-experiment questionnaires and underwent semi-structured interviews using Auto-Confrontation Video Analysis to explore decision-making processes. The resulting dataset, hosted on Figshare, provides millisecond-level alignment for networked sensors and frame-level alignment for video recordings. It contains 20 complete experimental groups, organized by driving condition, driver expertise level, and participant group. The data encompasses raw EEG signals, vehicle chassis data, GNSS trajectories, traffic volume statistics, and multi-angle video recordings of driver behavior and the external environment. Notably, the dataset excludes female professional drivers due to recruitment constraints and omits three initial groups due to technical or weather-related issues. This dataset offers significant value for researchers studying the neural basis of driving expertise. By providing synchronized brain activity, vehicle dynamics, and external validation through passenger feedback in real-world settings, it enables the analysis of how expert drivers allocate cognitive resources and filter stimuli. These insights are intended to inform the design of driver training programs and guide the development of autonomous systems that can better navigate complex urban traffic with human-like intelligence and safety.

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-prismaquant on 2026-06-26; verification: verified.

Topics

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

• situational awareness

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: dataset resource, tool software