Objective:

We determined whether pupil could serve as a reliable measure of arousal state to distinguish awake neural rhythms (e.g. hippocampal theta, gamma) from resting neural rhythms (e.g. hippocampal SPW-Rs) important for memory encoding and consolidation.

Background:

In rodents, hippocampal SPW-Rs (120-250 Hz) dominate offline, low-cholinergic states such as waking rest and sleep  (Zhang et al. 2021) and support memory consolidation. Recent human studies discovered that hippocampal ripple-frequency events (80-120 Hz) are associated with successful memory encoding and retrieval during wakefulness (Norman 2020, Vaz 2020, Sakon 2022, Henin 2023). However, distinguishing hippocampal ripple events from high frequency gamma activity (50–90 Hz) during wakefulness remains imprecise due to artifact, coarse electrode localization, and variable detection methods (Liu, Buzsaki Nat Comm 2023). We hypothesize that combining hippocampal iEEG recordings with dynamic measurements of arousal state (cholinergic state), such as pupil diameter, could help separate gamma and ripple activity.

Design/Methods:

Surgical epilepsy patients at NYU Langone Health with hippocampal depth electrodes participated in a meditative color dot task simulating quiet wakefulness. Pupil data was simultaneously collected via Tobii eye-tracking systems. Putative ripples were identified in hippocampal channels showing Power Spectral Density (PSD) peaks above 60 Hz and confirmed using ripmap, a Uniform Manifold Approximation and Projection (UMAP)-based waveform clustering toolbox. Pupil diameter changes were plotted over time and split into low and high cholinergic, based on top and bottom 25% of pupil size.

Results:

Conclusions:

Preliminary evidence suggests that pupil constriction correlates with ripple peaks, in agreement with rodent data. We will continue to analyze new subjects, with a planned enrollment of 10 patients.