Reveals that frozen LLMs contain person-specific 'neural signatures' that can predict individual brain activity.
March 24, 2026
Original Paper
Riding Brainwaves in LLM Space: Understanding Activation Patterns Using Individual Neural Signatures
arXiv · 2603.21847
The Takeaway
The study finds that individual EEG patterns are encoded as stable directions in the hidden layers of standard LLMs (Qwen/LLaMA). This suggests LLMs can be personalized to individual cognition or even used as backends for high-fidelity neural interfaces without retraining the base model.
From the abstract
Consumer-grade EEG is entering everyday devices, from earbuds to headbands, raising the question of whether language models can be adapted to individual neural responses. We test this by asking whether frozen LLM representations encode person-specific EEG signals, directions in activation space that predict one person's brain activity but not another's. Using word-level EEG from 30 participants reading naturalistic sentences (ZuCo corpus), we train a separate linear probe for each person, mappin