Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.26434 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914602447536128 |
|---|---|
| author | Kommineni, Aditya Zhou, Emily Avramidis, Kleanthis Segaard, Simon Bock Münster, Jeppe Roden Hansen, Andreas Peter Juhl Medani, Takfarinas Feng, Tiantian Leahy, Richard Narayanan, Shrikanth |
| author_facet | Kommineni, Aditya Zhou, Emily Avramidis, Kleanthis Segaard, Simon Bock Münster, Jeppe Roden Hansen, Andreas Peter Juhl Medani, Takfarinas Feng, Tiantian Leahy, Richard Narayanan, Shrikanth |
| contents | EEG foundation models, pre-trained on large-scale unlabelled EEG data, have emerged as a promising direction towards learning generalizable EEG representations. Despite showing positive results in data-rich regimes, they often fail to outperform significantly smaller supervised models in low-resource settings compared to fully supervised models. We provide a mechanistic account of this shortcoming, attributing it to a fundamental mismatch between reconstruction-based pretext tasks and the idiosyncratic spectral structure of EEG signals, which decompose into distinct high-power aperiodic and low-power oscillatory components. Using controlled, synthetically-generated EEG inputs, we demonstrate that EEG foundation model embeddings are biased to capture the aperiodic components of the EEG signal while under-representing oscillatory components, particularly at higher frequencies. Additionally, linear probe evaluations on real-world BCI datasets further reveal that embeddings encode subject identity more strongly than task-relevant information, thereby reinforcing the low-frequency and aperiodic component bias in foundation model embeddings trained primarily on reconstruction based objectives. Together, these findings elucidate a failure mode in reconstruction based EEG foundation models and motivate future work to incorporate auxiliary losses explicitly targeting high-frequency oscillatory structure as a path toward more capable and generalizable EEG representations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_26434 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Aperiodic and Low-Frequency Spectral Bias in Reconstruction based EEG Foundation Models Kommineni, Aditya Zhou, Emily Avramidis, Kleanthis Segaard, Simon Bock Münster, Jeppe Roden Hansen, Andreas Peter Juhl Medani, Takfarinas Feng, Tiantian Leahy, Richard Narayanan, Shrikanth Machine Learning Artificial Intelligence EEG foundation models, pre-trained on large-scale unlabelled EEG data, have emerged as a promising direction towards learning generalizable EEG representations. Despite showing positive results in data-rich regimes, they often fail to outperform significantly smaller supervised models in low-resource settings compared to fully supervised models. We provide a mechanistic account of this shortcoming, attributing it to a fundamental mismatch between reconstruction-based pretext tasks and the idiosyncratic spectral structure of EEG signals, which decompose into distinct high-power aperiodic and low-power oscillatory components. Using controlled, synthetically-generated EEG inputs, we demonstrate that EEG foundation model embeddings are biased to capture the aperiodic components of the EEG signal while under-representing oscillatory components, particularly at higher frequencies. Additionally, linear probe evaluations on real-world BCI datasets further reveal that embeddings encode subject identity more strongly than task-relevant information, thereby reinforcing the low-frequency and aperiodic component bias in foundation model embeddings trained primarily on reconstruction based objectives. Together, these findings elucidate a failure mode in reconstruction based EEG foundation models and motivate future work to incorporate auxiliary losses explicitly targeting high-frequency oscillatory structure as a path toward more capable and generalizable EEG representations. |
| title | Aperiodic and Low-Frequency Spectral Bias in Reconstruction based EEG Foundation Models |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.26434 |