Foundation Models for Cross-Domain EEG Analysis Application: A Survey
It addresses the problem of inconsistent and dispersed research in EEG analysis for neuroscientists and AI researchers, providing a systematic reference framework, but it is incremental as it surveys and categorizes existing work rather than proposing new methods.
This survey tackles the fragmented research landscape in EEG analysis by introducing the first comprehensive modality-oriented taxonomy for foundation models, organizing advances across EEG decoding, EEG-text, EEG-vision, EEG-audio, and multimodal frameworks to unify the field and accelerate practical applications.
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.