A comparative and critical study of EEGNet for fNIRS-driven cognitive load classification
2026-06-15 • Machine Learning
Machine LearningArtificial IntelligenceHuman-Computer Interaction
AI summaryⓘ
The authors studied how to best classify mental effort using brain signals from fNIRS, a technique that measures blood flow in the brain. They tested different ways of cutting the data into time segments, extracting key features, and training a neural network model called EEGNet. They found that using overlapping data segments works well when testing on the same people, but not for new individuals, where non-overlapping segments performed better. The work shows that avoiding repeated information in the data helps the model generalize better across people. Also, fixed learning rates worked better than adaptive ones for training the model.
cognitive loadfunctional near-infrared spectroscopy (fNIRS)EEGNettemporal segmentationfeature extractionlearning ratesubject-independent evaluationprincipal component analysis (PCA)hemodynamic response
Authors
Mehshan Ahmed Khan, Houshyar Asadi, Li Zhang, Mohammad reza Chalak Qazani, Ghazal Bargshady, Stefanos gkikas, Christian arzate, Sam Oladazimi, Zoran Najdovsk, Lei Wei, Chee Peng Lim
Abstract
Accurately classifying cognitive load from functional near-infrared spectroscopy (fNIRS) signals remains a significant challenge due to temporal variability, inter-subject differences, and sensitivity to preprocessing choices. This study provides a comprehensive evaluation of EEGNet for fNIRS-based cognitive load classification by systematically examining the effects of temporal segmentation strategies (overlapping vs. non-overlapping), window lengths (10s, 20s, 30s), feature extraction methods (Analysis of Variance (ANOVA), Principal Component Analysis (PCA), Fast Independent Component Analysis (FastICA)), learning rate configurations (fixed and adaptive), and evaluation protocols (random split vs. subject-independent (SI)). Results from random-split experiments show that overlapping segmentation, combined with smaller fixed learning rates (0.01-0.001), yields the highest accuracies, due to temporal redundancy and dense sampling of hemodynamic transitions. However, SI evaluation reveals a substantial drop in accuracy, demonstrating limited generalization to unseen participants. Under SI evaluation, non-overlapping segmentation outperformed overlapping windows, with the best accuracy of 56.11% achieved using PCA features with a 20-second window and a 0.1 learning rate. These findings indicate that eliminating temporal redundancy helps the model learn more robust and generalizable representations of cognitive load across individuals. Although adaptive learning rate strategy improved training stability, it did not surpass the performance of optimally selected fixed learning rates. The study highlights the critical role of segmentation strategy and learning rate selection in improving model generalization and identifies methodological considerations essential for developing reliable, real-time, and SI cognitive load classification systems using fNIRS.