DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI

2026-01-05

arXiv.org

AI Summary

This work addresses the limited generalization of existing EEG foundation models under frozen-probe settings, which stems from their failure to incorporate biophysical mechanisms of neural activity. To bridge this gap, we propose DeeperBrain—the first EEG foundation model that explicitly integrates neuroscientific priors. Our approach models volume conduction effects through 3D geometry-aware spatial channel encoding and captures neural dynamics using oscillatory and exponential basis functions for temporal encoding. We further introduce a dual-objective pretraining strategy combining Masked EEG Reconstruction (MER) and Neural Dynamics Statistics Prediction (NSP). DeeperBrain achieves state-of-the-art performance under both end-to-end fine-tuning and frozen-probe protocols, demonstrating that neuroscience-informed representations can attain the intrinsic universality and interpretability required for general-purpose brain–computer interfaces.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Deep Neural Architectures and Foundation Models
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Electroencephalography (EEG) foundation models hold significant promise for universal Brain-Computer Interfaces (BCIs). However, existing approaches often rely on end-to-end fine-tuning and exhibit limited efficacy under frozen-probing protocols, lacking the intrinsic universality required for broad generalization. This limitation stems from adapting general-purpose sequence architectures that overlook the biophysical and dynamical principles of neural activity. To bridge this gap, we propose DeeperBrain, a neuro-grounded foundation model integrating domain-specific inductive biases into its model design and learning objectives. Architecturally, DeeperBrain incorporates a volume conduction-aware channel encoding to model spatial mixing via 3D geometry, and a neurodynamics-aware temporal encoding capturing slow adaptations using oscillatory and exponential bases. For pretraining, we introduce a dual-objective strategy combining Masked EEG Reconstruction (MER) for local fidelity and Neurodynamics Statistics Prediction (NSP). NSP enforces alignment with macroscopic brain states by predicting interpretable order parameters, including spectral power, functional connectivity, cross-frequency coupling, and dynamic complexity. Extensive experiments demonstrate that DeeperBrain achieves state-of-the-art or highly competitive performance under end-to-end fine-tuning. Crucially, it maintains superior efficacy under a rigorous frozen-probing protocol, verifying that embedding neuroscientific first principles endows learned representations with the intrinsic universality essential for universal BCI. The code will be publicly available.

Problem

• EEG foundation model
• universal BCI
• neurodynamics
• frozen-probing
• biophysical principles

Innovation

• neuro-grounded
• volume conduction-aware encoding
• neurodynamics-aware temporal modeling
• frozen-probing generalization
• EEG foundation model