EvoBrain: Continual Learning of EEG Foundation Models Across Heterogeneous BCI Tasks

2026-06-01

AI Summary

This work addresses the limited scalability and high resource overhead of traditional EEG decoding approaches, which rely on task-isolated models and struggle to enable knowledge transfer or unified decoding across heterogeneous brain–computer interface (BCI) tasks. To overcome this, the study reframes downstream adaptation as a cross-task continual learning problem and introduces EvoBrain, a novel framework incorporating neural-spectral task normalization (NSN), response affinity distillation (RAD), and a time-varying replay mechanism. These components collectively allow the model to dynamically learn new tasks while effectively preserving previously acquired knowledge. Evaluated across six heterogeneous BCI tasks, EvoBrain significantly outperforms existing methods, achieving efficient, low-forgetting unified decoding and advancing the vision of a single-model solution for universal brain signal interpretation.

Abstract

Electroencephalography (EEG) is the cornerstone of non-invasive brain-computer interfaces (BCIs), yet conventional decoding relies on fragmented, task-specific architectures that severely limit cross-task scalability. While EEG foundation models pre-trained on massive corpora promise universal brain decoding, current post-training depends on task-isolated fine-tuning. This static paradigm restricts knowledge transfer across heterogeneous tasks, hinders model scalability, and incurs computational and storage overheads that scale linearly with task count. To overcome these bottlenecks, we formulate downstream adaptation as a cross-task continual learning problem and propose EvoBrain, a dynamic, task-aware continual learning framework for unified EEG decoding. EvoBrain addresses the plasticity-stability trade-off via two complementary components: (1) Neuro-Spectral Task Normalization (NSN) aligns incoming tasks with historical statistics while recalibrating spectral responses to handle distributional and neuro-spectral shifts; and (2) Response-Affinity Distillation (RAD), combined with time-dependent replay, preserves old-task response geometry and promotes selective knowledge transfer between spectrally compatible tasks, effectively mitigating forgetting. Extensive evaluations across six distinct BCI tasks demonstrate that EvoBrain consistently surpasses state-of-the-art methods across diverse foundation backbones, optimally balancing plasticity and stability. To our knowledge, this work pioneers cross-task continual learning in the EEG domain, advancing the realization of a unified, one-for-all brain decoding system.

Problem

• continual learning
• EEG foundation models
• heterogeneous BCI tasks
• cross-task scalability
• knowledge transfer

Innovation

• continual learning
• EEG foundation models
• cross-task transfer
• neuro-spectral normalization
• response-affinity distillation