Foundation Models for Cross-Domain EEG Analysis Application: A Survey

2025-08-21

AI Summary

Current EEG foundation model research is fragmented, architecturally heterogeneous, and lacks a systematic taxonomy—hindering field-wide progress. To address this, we propose the first output-modality–driven classification framework, categorizing EEG foundation models into five classes: EEG decoding, EEG–text, EEG–vision, EEG–audio, and multimodal models—thereby establishing the first comprehensive taxonomy. Through a systematic literature review and cross-modal technical analysis, we identify three core challenges: cross-domain generalization, model interpretability, and practical deployment. We further construct an EEG foundation model research landscape that maps architectural paradigms, training strategies, and evaluation protocols. This structured taxonomy and landscape serve as a principled reference for methodological innovation, significantly enhancing the scalability, applicability, and real-world deployability of EEG-based intelligent analysis systems.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Electroencephalography (EEG) analysis stands at the forefront of neuroscience and artificial intelligence research, where foundation models are reshaping the traditional EEG analysis paradigm by leveraging their powerful representational capacity and cross-modal generalization. However, the rapid proliferation of these techniques has led to a fragmented research landscape, characterized by diverse model roles, inconsistent architectures, and a lack of systematic categorization. To bridge this gap, this study presents the first comprehensive modality-oriented taxonomy for foundation models in EEG analysis, systematically organizing research advances based on output modalities of the native EEG decoding, EEG-text, EEG-vision, EEG-audio, and broader multimodal frameworks. We rigorously analyze each category's research ideas, theoretical foundations, and architectural innovations, while highlighting open challenges such as model interpretability, cross-domain generalization, and real-world applicability in EEG-based systems. By unifying this dispersed field, our work not only provides a reference framework for future methodology development but accelerates the translation of EEG foundation models into scalable, interpretable, and online actionable solutions.

Problem

• Surveying fragmented EEG foundation model research landscape
• Establishing first modality-oriented taxonomy for EEG analysis
• Addressing interpretability and generalization challenges in EEG systems

Innovation

• Modality-oriented taxonomy for EEG foundation models
• Systematic categorization of EEG-text-vision-audio frameworks
• Addressing interpretability and cross-domain generalization challenges