Abstracts

Propagation Patterns of Human Physiological Ripples Revealed by Hippocampal Microgrids

Abstract number : 2.191
Submission category : 3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year : 2024
Submission ID : 1054
Source : www.aesnet.org
Presentation date : 12/8/2024 12:00:00 AM
Published date :

Authors :
Presenting Author: Monica Chao, MD – University of California, San Francisco

Itzik Norman, PhD – University of California, San Francisco
Edward Chang, MD – University of California, San Francisco
Jonathan Kleen, MD, PhD – University of California, San Francisco

Rationale: Low frequency oscillations (1-15 Hz) propagate bidirectionally along the hippocampal longitudinal axis in humans, a spatiotemporal property that may represent a fundamental mechanism behind some of this structure’s diverse cognitive functions. Hippocampal physiological ripples in the 70-150 Hz range are markers of instantaneous memory recall and consolidation, and these have also been shown to propagate bidirectionally in rat models. However, it is unknown whether this spatiotemporal property translates to humans, much less its functional significance. As the anatomical sampling of depth electrodes may undermine the spatial detail required to evaluate these questions, here we leverage novel microgrid arrays conformed over the hippocampal surface to examine the two-dimensional spatiotemporal properties and related functional relevance of physiological ripples.

Methods: Six participants, including 5 with drug-resistant epilepsy, underwent excision of the antero-lateral temporal lobe with awake mapping during clinically-indicated surgery. A 32-channel thin-film microgrid array (4 x 8 layout, 2 mm spacing, 1.22 mm diameter contacts) was conformed to the ventricular hippocampal surface over the CA1 region and activity was sampled at 24 kHz. We detected ripples using standard threshold-based approaches on two patients that performed behavioral tasks, performed two-dimensional circular-linear regression to determine the direction and speed of propagating ripple cycles, and time-locked the data to behavioral tasks. We performed unsupervised ripple characterization using non-negative matrix factorization (NNMF) on space-time-frequency data features to identify inherent ripple sources along the hippocampal body, and investigated their physiologic relevance.

Results: Hippocampal ripple rates on microgrids increased during visual naming and word recall tasks. Interictal epileptiform spikes were detected using hybrid automated and manual methods and screened out of physiological analyses. We found that ripples propagated bidirectionally towards the septal (70%) and temporal (30%) poles, with a median wavelength of 30-50 mm and rapid traveling speed of 2000-4000 mm/s. Propagation directions overall were not modulated by behavioral tasks, though statistical power was limited by low trial numbers. NNMF on candidate ripples demonstrated two anatomically-distinct ripple sources. Both sources generated ripples that propagated towards the septal and temporal hippocampal poles, though again without a consistent link to behavior.

Conclusions: Our microgrid dataset allowed 2D spatial analysis of ripples for the first time in awake humans. Ripples were recorded from the hippocampal surface directly over the CA1, exhibiting spatiotemporal properties and behavioral modulation mirroring ripples recorded on depth electrodes, confirming their functional similarity. Further study with higher trial numbers are needed, though multiple anatomically-distinct sources of ripple generation with versatile propagation suggest potential functional properties relevant to the variance of cortical connections along the hippocampal longitudinal axis.

Funding: K23NS110920 (JKK)

Neurophysiology