Written by SOURCE: RESEARCHGATE
This 2026 systematic review by Molloy et al. highlights the growing role of machine-learning approaches in objectively assessing cognitive workload in aviation through physiological and behavioural data such as EEG, heart rate variability, and eye-tracking. For CRM and human-factors practitioners, the paper is significant because it demonstrates how advances in data analytics may allow workload to be monitored more accurately and potentially in real time, providing deeper insight into how workload influences situational awareness, decision-making, and crew coordination in complex operational environments. By identifying the strengths and limitations of current methods—particularly the benefits of multimodal data and the challenges of small experimental datasets—the study points toward future systems that could support adaptive training, workload management, and improved crew performance in high-reliability aviation teams
The study presents a systematic review of machine learning (ML) approaches used to measure and classify cognitive load in aviation. The authors reviewed 1,949 studies, ultimately selecting 43 relevant papers that applied machine-learning techniques to detect or predict cognitive workload using physiological, behavioural, or multimodal data.
The review shows that traditional approaches to measuring cognitive load—such as subjective questionnaires or performance measures—often lack precision and real-time applicability.
Machine learning offers a way to analyse complex physiological signals such as EEG, heart rate variability, and eye-tracking data to automatically classify pilot cognitive workload. Across the studies reviewed, ML models achieved widely varying classification accuracies, ranging from about 40% to nearly 99%, with higher accuracy generally achieved when multiple data sources (multimodal inputs) were combined rather than relying on a single physiological signal.
Reference:
Molloy, O., Eves, G., Vahidnia, S., & Shahin, M. (2026). Machine learning methods for cognitive load analysis and classification in aviation: A systematic review. International Journal of Human–Computer Interaction.https://doi.org/10.1080/10447318.2026.2632151