Foundation Models for Brain Signals: A Critical Review of Current Progress and Future Directions

2025-07-15

AI Summary

Supervised EEG encoders suffer from poor generalizability and heavy reliance on labeled data, motivating the emergence of universal self-supervised EEG foundation models (EEG-FMs). This paper presents the first systematic review of ten inaugural EEG-FMs, identifying critical bottlenecks: methodological homogeneity (predominantly Transformer-based masked reconstruction), fragmented evaluation protocols, and misalignment with real-world clinical deployment. To address these, we propose (1) a standardized evaluation framework and a trustworthy representation learning pipeline, and (2) interdisciplinary collaboration to co-develop domain-specific benchmarks, open-source toolkits, and application programming interfaces. We argue that unified, task-driven, and clinically interpretable validation paradigms are essential for enhancing EEG-FM utility. Our work establishes a methodological foundation for next-generation robust, scalable, and deployable EEG-AI systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems
• Machine Learning: Deep Neural Architectures and Foundation Models
• 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
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Patterns of electrical brain activity recorded via electroencephalography (EEG) offer immense value for scientific and clinical investigations. The inability of supervised EEG encoders to learn robust EEG patterns and their over-reliance on expensive signal annotations have sparked a transition towards general-purpose self-supervised EEG encoders, i.e., EEG foundation models (EEG-FMs), for robust and scalable EEG feature extraction. However, the real-world readiness of early EEG-FMs and the rubric for long-term research progress remain unclear. A systematic and comprehensive review of first-generation EEG-FMs is therefore necessary to understand the current state-of-the-art and identify key directions for future EEG-FMs. To that end, this study reviews 10 early EEG-FMs and presents a critical synthesis of their methodology, empirical findings, and outstanding research gaps. We find that most EEG-FMs adopt a sequence-based modeling scheme that relies on transformer-based backbones and the reconstruction of masked sequences for self-supervision. However, model evaluations remain heterogeneous and largely limited, making it challenging to assess their practical off-the-shelf utility. In addition to adopting standardized and realistic evaluations, future work should demonstrate more substantial scaling effects and make principled and trustworthy choices throughout the EEG representation learning pipeline. We believe that developing benchmarks, software tools, technical methodologies, and applications in collaboration with domain experts may further advance the translational utility and real-world adoption of EEG-FMs.

Problem

• Evaluating readiness and research progress of EEG foundation models
• Addressing limitations in self-supervised EEG encoder evaluations
• Identifying future directions for scalable EEG representation learning

Innovation

• Self-supervised EEG encoders for robust feature extraction
• Transformer-based backbones for sequence modeling
• Masked sequence reconstruction for self-supervision