Adaptive Temporal Dynamics for Personalized Emotion Recognition: A Liquid Neural Network Approach

TL;DR

This paper introduces a novel liquid neural network framework for EEG-based emotion recognition, leveraging temporal dynamics, multimodal data, and attention mechanisms to improve accuracy and interpretability in personalized emotion detection.

Contribution

It is the first to apply liquid neural networks with learnable time constants to EEG emotion recognition, integrating multimodal features and attention-guided fusion for enhanced performance.

Findings

Achieved 95.45% accuracy on PhyMER dataset, surpassing previous methods.

Provided interpretability through temporal attention analysis.

Demonstrated self-organization of neurons into functional groups for emotion modeling.

Abstract

Emotion recognition from physiological signals remains challenging due to their non-stationary, noisy, and subject-dependent characteristics. This work presents, to the best of our knowledge, the first comprehensive application of liquid neural networks for EEG-based emotion recognition. The proposed multimodal framework combines convolutional feature extraction, liquid neural networks with learnable time constants, and attention-guided fusion to model temporal EEG dynamics with complementary peripheral physiological and personality features. Dedicated subnetworks are used to process EEG features and auxiliary modalities, and a shared autoencoder-based fusion module is used to learn discriminative latent representations before classification. Subject-dependent experiments conducted on the PhyMER dataset across seven emotional classes achieve an accuracy of 95.45%, surpassing previously reported results. Furthermore, temporal attention analysis provides interpretable insights into emotion-specific temporal relevance, and t-SNE visualizations demonstrate enhanced class separability, highlighting the effectiveness of the proposed approach. Finally, statistical analysis of temporal dynamics confirms that the network self-organizes into distinct functional groups with specialized fast and slow neurons, proving it independently tunes learnable time constants and memory dominance to effectively capture complex emotion artifacts.