Lightweight EEG model improves cognitive workload detection under limits
FOCUS: A Lightweight Mamba-2 Model for EEG-Based Cognitive Workload Classification
Artificial Intelligence
Summary
Figuring out how hard someone is thinking can help keep people safe in risky jobs. The paper presents a small and fast computer model named FOCUS that looks at brain signals to tell if someone is under low or high mental load. Instead of learning from scratch, this model uses fixed patterns from common brain wave types and works well even with little data. It shows a good balance between accuracy and speed in tests with eleven people doing different tasks.
What this means in practice
- •For wearable device developers: Integrate an efficient EEG-based workload classifier in low-power brain-monitoring wearables for continuous mental state tracking.
- •For ergonomics teams: Use lightweight models to monitor workers’ cognitive load in real time for safer task allocation in high-risk environments.
Authors
Ioannis Kyprakis, Stefanos Gkikas, Chariklia Chatzaki, Zongyi Jiang, Randy Gomez, Raul Fernandez Rojas
Abstract
Cognitive workload estimation from physiological signals is increasingly relevant in high-risk environments, where continuous and unobtrusive monitoring can inform adaptive systems and safety interventions. This paper presents FOCUS (Frequency-band Optimized Cognitive-load Ultralight SSM), a compact Mamba-2-based model for EEG-based binary workload classification under constrained data conditions. Rather than learning frequency sensitivity from data, FOCUS encodes preprocessed multi-channel EEG using a fixed FFT decomposition that covers the five canonical EEG bands. The proposed model combines depthwise-separable convolution and a weight-shared encoder-bottleneck Mamba-2 pair, with 0.04 million parameters and 0.11 GFLOPs. Evaluated in a pilot study with 11 participants under leave-one-subject-out cross-validation across three cognitively distinct tasks, FOCUS achieves 74.75% accuracy on the full dataset at 0.08 ms GPU model-forward latency. In the weight-sharing ablation, the shared configuration achieves 4.55 percentage points higher accuracy on the full dataset than the non-shared variant while reducing the SSM parameter budget; however, this difference does not reach statistical significance under a two-sided Wilcoxon signed-rank test ($p=0.083$). Overall, the selected FOCUS configuration provides a favorable trade-off between classification performance and computational cost within this controlled pilot setting.