NeuroAct improves EEG signal analysis and model performance significantly
NeuroAct: Advancing EEG Analysis and Decoding via Dual-Loop Autonomous Research
Machine LearningArtificial Intelligence
Summary
Analyzing brain signals recorded by EEG is important for brain-computer interfaces and health monitoring, but current approaches don't clearly explain why some methods work better than others. The authors developed NeuroAct, a system that automatically tests and improves EEG analysis and prediction methods by learning from detailed feedback and reusing experience across tasks. Their system consistently improves accuracy on multiple EEG tasks and helps make better recommendations for future EEG research. NeuroAct offers a new way to advance brain data analysis through automated, evidence-driven exploration.
What this means in practice
- •For brain-computer interface developers: Automate EEG model testing and tuning to improve decoding accuracy for real-time control devices.
- •For sleep monitoring system engineers: Enhance EEG analysis pipelines for better detection and classification of sleep stages using adaptive model recommendations.
Authors
Wei Xiong, Jiangtong Li, Han Zhao, Linxiao Yang, Jie Li, Changjun Jiang
Abstract
EEG analysis and decoding underpin brain-computer interfaces, cognitive assessment, and sleep monitoring. Fixed pipelines and reward-driven search do not explicitly determine how signal structure, representations, predictor behavior, and training dynamics contribute to performance or generalization gaps, leaving intervention choice underdetermined. We introduce NeuroAct, an autonomous research agent framework for EEG analysis and decoding. NeuroAct implements dual-loop autonomous research across EEG foundation models, task-specific predictors, and model combinations. In an inner loop, research, development, and review roles use evidence on signal structure, representations, predictor behavior, and training dynamics to test hypotheses, select analyses or interventions, and refine candidates using performance, stability, and cost feedback. An outer loop distills outcomes into conditional, testable recommendations with task and model conditions, then validates and revises them on subsequent tasks.We evaluate four settings: (A) EEG General Analysis, (B) Cross-task Model Optimization, (C) Single-task Open Research, and (D) Collaborative Prediction. In Setting A, NeuroAct's 80.88 macro-average exceeds BrainAgent's 63.13 by 17.75 points. In Setting B, NeuroAct raises EEGPT's mean balanced accuracy (BA) by 7.34 points and HBN Pearson's (r) from 0.6894 to 0.7362; in Setting C, it improves ADFTD BA by 16.33 points; and in Setting D, it raises Workload BA from 79.69% to 85.02%. Experience reuse raises BCIC-IV-2a BA from 34.66% to 50.81% while reducing LLM tokens by 64.02%, indicating recommendations improve cross-task performance. NeuroAct enables evidence-guided EEG research improved by cross-task experience.