A multidisciplinary review of machine learning for driver distraction prediction: Unveiling challenges
Abstract
Driver distraction remains a significant contributor to road accidents worldwide, presenting an urgent need for advanced prediction systems capable of addressing this critical public safety issue. Despite advancements in the integration of physiological signals such as electroencephalography (EEG), galvanic skin response (GSR), and heart rate variability (HRV) with machine learning algorithms, achieving reliable performance in real-world scenarios is still a challenge. The variability of individual responses to cognitive and visual distractions complicates the scalability of existing systems, limiting their applicability in dynamic driving contexts.This review examines the current state of distraction detection systems, focusing on the interplay between physiological signal processing and machine learning methodologies. While promising progress has been made with techniques such as hybrid deep learning architectures, LSTM-CNN models, these approaches often struggle to generalise across diverse populations and real-world environments. Challenges such as signal variability, overfitting, and the practical deployment of multimodal systems are explored, offering a critical evaluation of the limitations in existing frameworks.By synthesising insights from cognitive science, physiological analysis, and machine learning, this paper highlights the importance of developing adaptive and context-aware systems capable of addressing the complexities of distraction detection. Through a multidisciplinary approach, the review provides a foundation for future research aimed at bridging theoretical advancements and practical implementation, paving the way for more effective, scalable, and personalised solutions to enhance road safety.
RAS ID
83669
Document Type
Conference Proceeding
Date of Publication
1-1-2025
School
School of Arts and Humanities / School of Science
Copyright
subscription content
Publisher
IEEE
Identifier
Craig Speelman: https://orcid.org/0000-0001-8629-174X
Mike Johnstone: https://orcid.org/0000-0001-7192-7098
Recommended Citation
Aponso, A., Speelman, C., & Johnstone, M. (2025). A multidisciplinary review of machine learning for driver distraction prediction: Unveiling challenges. Retrieved from https://ro.ecu.edu.au/ecuworks2022-2026/6702
Comments
Aponso, A., Speelman, C., & Johnstone, M. (2025). A multidisciplinary review of machine learning for driver distraction prediction: Unveiling challenges. 2025 10th International Conference on Information and Network Technologies (ICINT), 30-35. https://doi.org/10.1109/ICINT65528.2025.11030911