EEG-Xplain introduces a unified attribution framework to interpret EEG foundation models such as BIOT, LaBraM, and EEGMamba. The framework combines gradient, perturbation, and activation-based methods to analyze model behavior across spatial, temporal, and frequency dimensions, identifying critical channels, decision-relevant signal segments, and contributions of canonical EEG rhythms. It evaluates explanation reliability with population-level metrics and uses large language models to convert structured attributions into natural-language reports, demonstrating consistency with known neurophysiological markers on benchmark datasets.
arXiv:2608. 13676v1 Announce Type: new Abstract: Objective: Foundation models represent the next advancement in AI for EEG analysis; however current explainable AI techniques provide attribution scores in the time-channel input space, which is mismatched to clinical intuition about EEG.
The paper evaluates EEG foundation models (EEG‑FMs) beyond clean accuracy, testing six models and a supervised baseline across ten datasets. It finds that no single model excels in all robustness tests, that attribution methods reveal task‑appropriate brain region relevance, and that poor head‑only performance is largely due to pooling strategies rather than representation quality. The study also shows that fine‑tuning repurposes late blocks while early blocks already encode task information.
arXiv:2608. 04156v1 Announce Type: new Abstract: Electroencephalography (EEG) analysis extends beyond assigning predefined labels to recordings; it requires workflows connecting natural-language instructions, signal processing, quantitative evidence, and scientific interpretation.
arXiv:2609.36609v1 Announce Type: cross Abstract: Electroencephalography (EEG) analysis requires careful choices in preprocessing, statistical modeling, and machine learning because EEG signals are h...
arXiv:2608.24597v1 Announce Type: cross Abstract: Electroencephalography (EEG) is a widely used window into human brain function, but most EEG models remain tied to a one-dataset-one-model supervised...
The article reviews how inter- and intra-subject variability in EEG affects reliability and reproducibility across resting-state, ERP, and task-related/BCI paradigms. It finds that inter-subject differences are generally larger than within-subject fluctuations, with stability varying by feature—alpha-band measures are more reliable than higher-frequency or connectivity metrics, and P300 components show moderate-to-good reliability. The review catalogs biological, state-related, technical, and analytical sources of variability, discusses quantification methods such as ICC, CV, SNR, and generalizability theory, and offers design and reporting recommendations to manage and leverage variability for precision neuroscience and robust neurotechnology.
The paper investigates how EEG embeddings from signal processing and foundation models encode trait and state information. Using a hierarchy metric and kNN purity on a BCI dataset, it finds that most embeddings form a macro‑trait hierarchy with micro‑states inside, which hampers state discrimination. Normalizing embeddings reduces trait leakage and improves state classification accuracy by about 10%, and fine‑tuning shifts the hierarchy toward macro‑state organization, enhancing decoding performance.
arXiv:2604. 16926v2 Announce Type: replace-cross Abstract: Electroencephalography (EEG) foundation models have shown strong potential for learning generalizable representations from large-scale neural data, yet their clinical deployment is hindered by distribution shifts across clinical settings, devices, and populations.
arXiv:2606. 00815v1 Announce Type: new Abstract: Electroencephalography (EEG) supports a variety of brain-computer interface (BCI) tasks ranging from brain-state monitoring to human-LLM interactions.
Brain4FMs is a unified benchmark for evaluating brain foundation models (BFMs) on both scalp EEG and intracranial EEG (iEEG). It incorporates 17 representative models and 21 public datasets spanning clinical diagnosis, sleep staging, communication, and affective computing, and offers dataset-aware preprocessing, cross‑subject evaluation, and standardized downstream workflows. The benchmark highlights that no single BFM consistently outperforms others across all tasks, modalities, and adaptation protocols, prompting further exploratory analyses of model‑specific spatial, spectral, and discrete representations.
NeuroAtlas is the largest EEG benchmark to date, comprising 42 datasets and 260,000 hours of clinical EEG data across epilepsy, sleep medicine, brain age estimation, and brain‑computer interfaces. The study evaluates foundation models (FMs) for EEG against supervised baselines and generic time‑series FMs, finding that EEG‑specific FMs do not consistently outperform generic ones. It also demonstrates that standard machine‑learning metrics are inadequate for clinical relevance, advocating for task‑specific measures such as event‑level decision quality, hypnogram features, and brain‑age gap.
EEG-Captioner introduces a new EEG-to-language framework that grounds language generation in measurable EEG signal properties. The authors built a large dataset of 160,781 EEG segments annotated with question‑answer pairs targeting spectral, spatial, temporal, and lateral properties, and paired each segment with a feature‑grounded caption. Their Dual‑Domain EEG Encoder jointly models temporal waveforms and spectral‑spatial representations, achieving a macro‑averaged balanced accuracy of 88.81% on signal‑level QA and improving caption factual consistency from 82.06% to 83.46% compared to caption‑only models.