LUCID: Frozen Unsupervised Diffusion Features for EEG Decoding

EEG foundation models pretrain on large corpora to learn general representations that transfer to downstream decoding tasks. Yet they are typically evaluated after full fine-tuning, entangling what pretraining encodes with what task labels supply. To separate the two, we present $\text{LUCID}$ (Latent Unsupervised Characterisation from Internal Diffusion): an unsupervised diffusion model trained from scratch on each dataset, then frozen, with its internal activations read by a small linear probe. Two properties of EEG shape $\text{LUCID}$'s denoiser. Along the time axis, the signal is densely sampled and autocorrelated, with rhythms spanning tens of milliseconds to seconds; patching or tokenizing it, as most foundation models do, coarsens these dynamics. Across electrodes, channel power and inter-channel coupling shift from window to window – a nonstationary covariance that no fixed spatial operator can follow. We therefore build a time-preserving denoiser from structured long convolutions and, within each block, introduce $\text{CoSMO}$, a spatial-modulation operator that adapts to each window's own covariance. We evaluate $\text{LUCID}$ on ten EEG decoding tasks across twelve datasets against seven task-specific supervised networks, a diffusion-based representation model, and six fully fine-tuned foundation models. On five of the twelve datasets, its frozen features tie the best of these foundation models. Re-evaluating each foundation model frozen, with only its head trained, reveals a pattern: $\text{LUCID}$ ties where these models gain little from fine-tuning and trails where they gain most. It is also label-efficient, matching its full-label accuracy with a quarter of the labels on half of the datasets and keeping at least four fifths of it on the rest. These findings argue for evaluating EEG representations frozen, not only after full fine-tuning. Beyond decoding, $\text{LUCID}$ generates synthetic EEGs that capture much of the spectral, temporal, and spatial structure of real recordings. Our source code is made available.