T-SNN: Temporal Simplicial Neural Network for EEG Decoding

2026-09-27

AI Summary

This study addresses the limitation of existing methods in effectively modeling higher-order dynamic interactions among brain regions. To this end, we propose the Temporal Simplicial Neural Network (T-SNN), which represents EEG signals as sequences of evolving simplicial complexes. By integrating simplicial convolutions with recurrent update mechanisms for the first time, T-SNN explicitly captures higher-order neural interactions and their temporal evolution. Evaluated on the SEED-VII emotion recognition task, T-SNN significantly outperforms current state-of-the-art methods, and its performance can be further enhanced through multimodal fusion. Overall, this work establishes a novel paradigm for modeling dynamic higher-order brain networks.

Abstract

Decoding brain states requires models that capture both the evolution of neural activity and interactions among groups of brain regions. Existing EEG methods often treat recordings as multivariate time series or represent functional connectivity with pairwise graphs, leaving dynamic higher-order interactions largely unmodeled. We introduce the Temporal Simplicial Neural Network (T-SNN), which represents EEG recordings as sequences of evolving simplicial complexes. By combining simplicial convolutions with recurrent updates, T-SNN jointly learns higher-order interactions and their temporal evolution. On the seven-class SEED-VII emotion recognition task, T-SNN outperforms convolutional, recurrent, graph-based, and Transformer methods in both trial-wise and cross-subject evaluations. Incorporating eye-movement features further improves performance, demonstrating the framework's potential for multimodal brain-state decoding.

Problem

• EEG decoding
• higher-order interactions
• temporal evolution
• brain state decoding
• simplicial complexes

Innovation

• Temporal Simplicial Neural Network
• Simplicial Complexes
• Higher-order Interactions
• EEG Decoding
• Multimodal Fusion