Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Semantic Processing in Anesthetized and Awake Patients
Poster Session E, Sunday, September 14, 11:00 am - 12:30 pm, Field House
This poster is part of the Sandbox Series.
Elizabeth Mickiewicz1, Melissa Franch1, Kalman Katlowitz1, James Belanger1,2, Assia Chericoni1, Ana Chavez1, Eleonora Bartoli1, Sarah Heilbronner1, Nicole Provenza1, Andrew Watrous1, Sameer Sheth1, Benjamin Hayden1,2; 1Baylor College of Medicine, 2Rice University
It has been shown that the meanings of words are encoded during language comprehension via both vectorial coding and prediction. More recently, we discovered that despite reduced perceptual processing during anesthesia, semantic processing persists though potentially different from semantic processing in awake patients. To address this gap in knowledge, we recorded hippocampal single neuron activity using Neuropixel and sEEG electrodes from anesthetized and awake patients, respectively, during naturalistic speech presentation. Here, we compared how dense (or sparse) the codes were in the two contexts, as it may be the case that reduced firing rates and less noise during anesthesia allow for a sparser code to encode word meanings compared to awake patients. We also examined whether semantic categories are decoded similarly as a function of word frequency between awake and anesthetized patients, despite overall lower firing rates in anesthetized patients. Additionally, we analyzed whether semantic encoding systematically varies along the length of the Neuropixel probe. We found that the ability of word embeddings to predict neural activity varies with anatomical location in the hippocampus. We expect that neurons physically closer to each other have more similar embedding predictor patterns. This research provides insights into how different states of consciousness affect linguistic processing in the human hippocampus, potentially highlighting aspects of semantic processing that are apparent in varying levels of awareness.
Topic Areas: Meaning: Lexical Semantics, Methods