ASPEN: Spectral-Temporal Fusion for Cross-Subject Brain Decoding

Megan Lee, Seung Ha Hwang, Inhyeok Choi, Shreyas Darade, Mengchun Zhang, Kateryna Shapovalenko · 2026

Read ASPEN: Spectral-Temporal Fusion for Cross-Subject Brain Decoding by Megan Lee, Seung Ha Hwang, Inhyeok Choi, Shreyas Darade, Mengchun Zhang, Kateryna Shapovalenko (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-16147.txt.

Description

Megan Lee leads a team of six researchers improving brain-computer interfaces that must work for new users unseen in training. Cross-subject generalization in EEG-based interfaces remains hard. Neural signals vary from person to person. The team asks whether spectral representations give more stable features for transfer than temporal waveforms. Correlation analyses across three EEG paradigms, SSVEP, P300, and motor imagery, give a clear answer. Spectral features are consistently more similar across subjects than temporal signals.

That finding motivates ASPEN, a hybrid design that fuses spectral and temporal streams multiplicatively. Features propagate only when the two streams agree. Across six benchmark data sets, ASPEN dynamically finds the best spectral-temporal balance for each paradigm. It tops unseen-subject accuracy on three of six data sets and stays competitive on the rest. Multiplicative fusion thus supports effective cross-subject generalization.

Catalog record

CREATORMegan Lee, Seung Ha Hwang, Inhyeok Choi, Shreyas Darade, Mengchun Zhang, Kateryna Shapovalenko
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-16147

Browse the full library · About the Tower of Babel