Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Riding Brainwaves in LLM Space: Understanding Activation Patterns Using Individual Neural Signatures

Ajan Subramanian, Sumukh Bettadapura, Rohan Sathish

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, mapping hidden states from a frozen Qwen 2.5 7B to that individual's EEG power. Person-specific probes outperform a single population probe on every EEG feature tested; for high-gamma power, the person-specific probe achieves rho = 0.183, a ninefold improvement over the population probe (rho = 0.020, p < 10^-4). A negative control, fixation count, shows no person-specific advantage (p = 0.360); fixation count reflects word length and frequency rather than individual cognition. The individual directions are temporally stable (split-half cosine = 0.824), non-transferable across people (self rho = 0.369 vs. other rho = 0.143, p < 10^-19), and distinct from the shared population signal: person-specific probes retain predictive power after the population component is removed. The person-specific signal concentrates in the model's deep layers, rising consistently with depth and peaking at Layer 24 of 28. The results are consistent across architectures (LLaMA 3.1 8B) and survive word-level confound controls. Frozen language models contain stable, person-specific neural directions in their deep layers, providing a geometric foundation for EEG-driven personalization.

Riding Brainwaves in LLM Space: Understanding Activation Patterns Using Individual Neural Signatures

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, mapping hidden states from a frozen Qwen 2.5 7B to that individual's EEG power. Person-specific probes outperform a single population probe on every EEG feature tested; for high-gamma power, the person-specific probe achieves rho = 0.183, a ninefold improvement over the population probe (rho = 0.020, p < 10^-4). A negative control, fixation count, shows no person-specific advantage (p = 0.360); fixation count reflects word length and frequency rather than individual cognition. The individual directions are temporally stable (split-half cosine = 0.824), non-transferable across people (self rho = 0.369 vs. other rho = 0.143, p < 10^-19), and distinct from the shared population signal: person-specific probes retain predictive power after the population component is removed. The person-specific signal concentrates in the model's deep layers, rising consistently with depth and peaking at Layer 24 of 28. The results are consistent across architectures (LLaMA 3.1 8B) and survive word-level confound controls. Frozen language models contain stable, person-specific neural directions in their deep layers, providing a geometric foundation for EEG-driven personalization.
Paper Structure (28 sections, 2 equations, 3 figures, 4 tables)

This paper contains 28 sections, 2 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Representation engineering.
  4. Neuro-linguistic alignment.
  5. Precision neuroimaging.
  6. Word-level EEG prediction.
  7. Experimental Setup
  8. Data
  9. Hidden-State Extraction
  10. Probing Framework
  11. Confound Features
  12. Results
  13. Person-Specific Probes Outperform Population Probes
  14. Characterization: Genuine Individual Signatures
  15. Cross-person transfer.
  16. ...and 13 more sections

Figures (3)

  • Figure 1: (a) Different readers produce distinct neural signatures for the same word; each person-specific probe learns a different direction in the LM's activation space ("pop" = population). (b) Probing pipeline: words pass through a frozen LM and PCA; 30 person-specific Ridge probes are compared against a single population probe. The person-specific probe achieves $\bar{\rho} = 0.183$ vs. $0.020$ for the population probe on TRT_g2 (high-$\gamma$, Layer 24).
  • Figure 2: Layer-wise probe performance on Qwen 2.5 7B (ZuCo, TRT_g2). Person-specific probes (solid blue, mean $\rho$ across 30 participants) increase with depth, peaking at Layer 24. The population probe (dashed orange) remains near zero throughout. Layer 28 shows a slight decline relative to Layer 24.
  • Figure 3: Cross-person transfer matrix (Qwen 2.5 7B, mean pupil size, Layer 24, PCA 50). Each cell $(i, j)$ shows Spearman $\rho$ when participant $i$'s probe predicts participant $j$'s mean pupil size. The diagonal (self $\rho = 0.369$) is consistently higher than the off-diagonal (other $\rho = 0.143$), confirming person-specificity. LLaMA 3.1 8B shows the same pattern (Table \ref{['tab:llama-results']}).