MSCGC-KAN: Multi-scale Causal Graph Convolution and Kolmogorov-Arnold Feature Mapping for EEG Emotion Recognition

2026-05-26

AI Summary

This work addresses limitations in existing EEG-based emotion recognition approaches—specifically, insufficient multi-scale dynamic modeling, underutilization of functional channel connectivity, and the limited representational capacity of linear classification heads during fine-tuning of pretrained models. Building upon the CBraMod pretrained backbone, the authors propose a structured task head that integrates multi-scale causal graph convolution with Kolmogorov–Arnold nonlinear feature mapping to jointly enhance temporal modeling, learnable inter-channel connectivity, and discriminative power. Evaluated on the FACED and SEED-VII datasets, the method achieves balanced accuracies of 60.66% and 33.27%, respectively, outperforming the CBraMod+Linear baseline by 5.91 and 2.03 percentage points, thereby significantly improving sensitivity to emotion-relevant spatiotemporal patterns.

Abstract

Electroencephalogram (EEG)-based emotion recognition is an important affective computing task, and recent EEG foundation models provide useful generic representations for downstream adaptation. However, under the fine-tuning setting, three limitations remain prominent: insufficient modeling of multi-scale emotional dynamics, inadequate exploitation of inter-channel functional connectivity, and the limited expressive power of simple linear classification heads. To address these issues, this paper proposes a new EEG emotion recognition method, termed MSCGC-KAN, which introduces a structured task head composed of multi-scale causal graph convolution and Kolmogorov--Arnold feature mapping. Built on a pre-trained CBraMod backbone, MSCGC-KAN enhances downstream adaptation by jointly strengthening multi-scale temporal modeling, learnable inter-channel connectivity modeling, and nonlinear discriminative mapping within a compact task-specific head. This design preserves the representation advantage of the foundation model while making the classifier more sensitive to emotion-related spatiotemporal patterns. Extensive experiments are conducted on the public FACED and SEED-VII datasets. The proposed method achieves a balanced accuracy of 60.66\%, a Cohen's Kappa of 0.5525, and a weighted F1-score of 60.40\% on FACED, and obtains 33.27\%, 0.2223, and 33.64\%, respectively, on SEED-VII. Compared with the CBraMod+Linear baseline, the balanced accuracy is improved by 5.91 and 2.03 percentage points on the two datasets, respectively. These results indicate that structured task-head design is an effective way to improve EEG emotion recognition when fine-tuning pre-trained EEG models.

Problem

• EEG emotion recognition
• multi-scale emotional dynamics
• inter-channel functional connectivity
• linear classification head
• fine-tuning

Innovation

• Multi-scale Causal Graph Convolution
• Kolmogorov-Arnold Network
• EEG Emotion Recognition
• Structured Task Head
• Functional Connectivity Modeling