Learning aligned EEG representations with subject-specific encoders

[Submitted on 15 Jun 2026 (v1), last revised 10 Aug 2026 (this version, v2)]

View PDF HTML (experimental)

Abstract:Cross-subject EEG decoding promises more training data, but it also exposes neural networks to strong inter-subject distribution shifts. We study whether task supervision and architecture alone can learn subject-aligned representations. We replace a shared EEG encoder with subject-specific encoders followed by a common classifier, and compare this hybrid model with standard EEGNet, AttentionBaseNet, and CTNet baselines with Euclidean Alignment (EA) on four motor-imagery datasets. EA improves shared encoders by recentering subject covariances, but the hybrid encoder largely internalises this role: validation-loss curves and latent-distance analyses change little when EA is removed. Subject-specific heads increase class distinctiveness and place each subject close to its own latent manifold, improving most subjects while leaving a method-sensitive subset. These results support subject-specific encoders as a learned alignment mechanism for EEG decoding and identify head selection for unseen subjects as the remaining bottleneck.

Submission history

From: Bruno Aristimunha [view email]
[v1] Mon, 15 Jun 2026 09:31:56 UTC (2,546 KB)
[v2] Mon, 10 Aug 2026 10:46:39 UTC (2,566 KB)