{
  "abstract": "In a previous report we showed that a decoder can recover open-vocabulary word-level information from 19-channel EEG recorded during silent reading, by aligning short EEG windows with hidden-state embeddings of the presented word taken from a large language model (LLM). That analysis contrasted two of the model's 33 layers: the context-free input embedding and one mid-depth layer. Here we ask how EEG-decodable structure is distributed over the full depth of the language model. We trained one decoder per layer of Llama-3.1-8B, with and without a causal transformer over the sequence of EEG windows. We distinguish word-level decodability from a context-tracking component. Word-level decodability peaks in early layers and declines steadily towards the last layer. In contrast, the context-tracking component of the retrieval gain rises with depth, both with and without a causal transformer. Unexpectedly, a decoder that sees a single 0.5-s window at a time tracks the text passage too. Two new controls dissect this component. A target-geometry control, a synthetic predictor made of the true target plus noise calibrated to each decoder's word-level accuracy, shows that context tracking is not a by-product of word decoding and target geometry. A control that permutes predictions across runs from either the same or from different sessions shows that most of the observed context-tracking is not caused by an EEG recording session signature. Instead, context tracking mostly follows the text. A single EEG window therefore carries passage-level information that a contextual target can absorb. The results locate word identity in the early layers of the language model, discourse state in the deeper layers, and make the choice of target layer a design dial for EEG language decoders.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.23638",
  "author_total": 3,
  "authors": [
    {
      "name": "Ingo Marquardt"
    },
    {
      "name": "Anthilia Alchanat"
    },
    {
      "name": "Priyanka Jain"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dv8s21dqnpdnw.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.23638",
  "fulltext_url": null,
  "id": "dv8s21dqnpdnw",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.23638",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "q-bio.NC"
    ],
    [
      "openalex",
      "2805|EEG and Brain-Computer Interfaces"
    ]
  ],
  "openalex": "W7222961840",
  "pdf_url": "https://arxiv.org/pdf/2610.23638",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-16",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dv8s21dqnpdnw.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:30:38Z",
      "native_id": "2610.23638",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.23638"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.NC",
      "subfield": "neuroscience"
    }
  ],
  "title": "Deep LLM layers expose passage-level context in EEG during silent reading",
  "title_text": "Deep LLM layers expose passage-level context in EEG during silent reading",
  "type": "preprint",
  "updated_at": "2026-10-30T12:33:34Z",
  "url": "https://research.draconitic.com/paper/dv8s21dqnpdnw",
  "venue": "arXiv",
  "versions": []
}