ORIGINAL RESEARCH article
Front. Neurosci.
Sec. Decision Neuroscience
Multimodal Sensor-based Response Mechanisms of Drivers Under Adverse Weather Conditions Using EEG Spectral Entropy and Prediction Entropy
- YT
Yi Tian 1
- JH
Jianping Hu 1,2
- HD
Hao Ding 1
- BY
Binhe Yang 1
- JH
Jialin Hu 1
- YL
Yuting Liu 1
- YD
Yu Ding 1
1. Xinjiang University, Urumqi, China
2. Xinjiang Key Laboratory of Green Construction and Smart Traffic Control of Transportation Infrastructure, Urumqi, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Abstract
Introduction: Whether rain, snow, and fog elicit qualitatively different driver-state dynamics remains unclear. Methods: Thirty licensed drivers completed a simulator experiment under clear, rain, snow, and fog conditions, during which subjective scales, electrocardiograms, electroencephalograms, and vehicle data were acquired simultaneously. Six response pathways were constructed to characterize subjective load, operational fluctuation, conservative control, physiological arousal, electroencephalographic response, and system uncertainty. C4 spectral entropy gauged neural complexity, class-wise optimized reliability fusion prediction entropy measured system uncertainty in multimodal sensor responses, and within-subject centering, together with distance correlation, revealed how information is structured across modalities. Results: Among out-of-fold predictions over 6456 windows, the class-wise optimized reliability fusion model achieved 91.9% accuracy and Macro-F1 of 0.919. All three adverse weather types significantly increased the subjective load, yet the dominant pathways differed. Rainfall increased cognitive load, snow led to synchronized increases in neural complexity and system uncertainty, and fog induced conservative compensation under visual restriction. Snow produced the strongest system-level response (class-wise optimized reliability fusion prediction entropy: r = 0.79, 95% confidence interval (CI): [0.62, 0.87]; C4 spectral entropy: r = 0.53, 95% CI [0.21, 0.79]). The four modalities formed an information-complementary structure. Discussion: This study elucidates the information dynamics of driving states under adverse weather conditions and demonstrates that rain, snow, and fog occupy distinct regions within the information space, thereby providing a sensor-informed foundation for driver-state monitoring and graded warning systems in intelligent vehicles.
Summary
Keywords
Adverse weather, cognitive workload, Complex System, driver state, Information complementarity, Multimodal sensors, neuroergonomics, prediction entropy
Received
25 June 2026
Accepted
07 August 2026
Copyright
© 2026 Tian, Hu, Ding, Yang, Hu, Liu and Ding. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jianping Hu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.