Multi-Domain EEG Representation Learning with Orthogonal Mapping and Attention-based Fusion for Cognitive Load Classification

TL;DR

This work tackles cognitive load classification from EEG by proposing a multi-domain representation learning framework that jointly leverages raw time-series and frequency-domain multi-spectral topography maps. A dual-stream encoder architecture produces time- and frequency-domain embeddings, which are fused via an attention module into a shared space; an orthogonal embedding constraint is enforced with a total loss $\mathcal{L}_{total} = \mathcal{L}_{CE} + \beta \mathcal{L}_{OC}$ to maximize inter-class separation while preserving intra-class cohesion. The approach achieves state-of-the-art performance on CL-Drive and CLARE, demonstrating robust gains over single-domain baselines and strong resilience to noise, with real-time inference feasibility around 20 ms. These results advance EEG-based cognitive load detection in driving and MATB-II contexts by effectively exploiting complementary time- and frequency-domain information through a principled fusion and regularization scheme.

Abstract

We propose a new representation learning solution for the classification of cognitive load based on Electroencephalogram (EEG). Our method integrates both time and frequency domains by first passing the raw EEG signals through the convolutional encoder to obtain the time domain representations. Next, we measure the Power Spectral Density (PSD) for all five EEG frequency bands and generate the channel power values as 2D images referred to as multi-spectral topography maps. These multi-spectral topography maps are then fed to a separate encoder to obtain the representations in frequency domain. Our solution employs a multi-domain attention module that maps these domain-specific embeddings onto a shared embedding space to emphasize more on important inter-domain relationships to enhance the representations for cognitive load classification. Additionally, we incorporate an orthogonal projection constraint during the training of our method to effectively increase the inter-class distances while improving intra-class clustering. This enhancement allows efficient discrimination between different cognitive states and aids in better grouping of similar states within the feature space. We validate the effectiveness of our model through extensive experiments on two public EEG datasets, CL-Drive and CLARE for cognitive load classification. Our results demonstrate the superiority of our multi-domain approach over the traditional single-domain techniques. Moreover, we conduct ablation and sensitivity analyses to assess the impact of various components of our method. Finally, robustness experiments on different amounts of added noise demonstrate the stability of our method compared to other state-of-the-art solutions.

Figures (8)

• Figure 1: Overview of the orthogonality constraint used in our method, which improves inter-class separation and intra-class clustering for the fused embedding space.
• Figure 2: The overview of our proposed network. The diagram shows the multi-domain representation learning along with the attention module and the orthogonality constraint.
• Figure 3: EEG electrode locations according to the international 10-20 system.
• Figure 4: UMAP visualizations of the CL-Drive dataset under different model configurations.
• Figure 5: Accuracy and F1 scores for our method vs. baselines with increasing Gaussian noise.