Temporal dynamics of memory encoding and retrieval in the human medial temporal lobe and prefrontal cortex

Poster Session E - Monday, April 15, 2024, 2:30 – 4:30 pm EDT, Sheraton Hall ABC

Adam Dede¹ (adam.osman.dede@gmail.com), Zachariah Cross¹, Samantha Gray¹, Qin Yin², Parisa Vahidi², Eishi Asano², Stephan Schuele¹, Joshua Rosenow¹, Joyce Wu^1,3, Sandi Lam^1,3, Jeffrey Raskin^1,3, Jack Lin⁴, Olivia Kim-McManus⁵, Shifteh Sattar⁵, Ammar Shaikhouni⁶, David King-Stephens^7,8, Peter Weber⁷, Kenneth Laxer⁷, Peter Brunner⁹, Jarod Roland^9,10, Saez Ignacio^11,12, Fady Girgis^11,13, Robert Knight¹⁴, Noa Ofen², Lisa Johnson¹; ¹Northwestern University, ²Wayne State University, ³The Ann & Robert H. Lurie Children's Hospital of Chicago, ⁴University of California Davis, ⁵University of California San Diegoand UCSD Rady Children's Hospital, ⁶Ohio State University, ⁷California Pacific Medical Center, ⁸Yale University, ⁹Washington University in St. Louis, ¹⁰St. Louis Children's Hospital, ¹¹University of California Davis, ¹²Ichan School of Medicine at Mount Sinai, ¹³University of Calgary, ¹⁴University of California Berkeley

The medial temporal lobe and prefrontal cortex coordinate their activity in the service of memory. We used intracranial recordings from 36 neurosurgical patients (22 male, 19 ± 4.7 years old) to investigate the relative timing of activity in these regions during encoding and recognition testing of natural scene images. From an initial set of 605 channels, we selected only those with task-related high frequency broadband (HFB; 70-150 Hz) responses--which index neuronal population activity--yielding 181 dorsolateral prefrontal cortex channels, 30 anterior cingulate cortex channels, 30 polar prefrontal cortex channels, 67 parahippocampal and rhinal cortex (PHRC) channels, and 29 hippocampal (HC) channels. Linear mixed-effects modeling revealed a three-way interaction between hit/miss, region, and encoding/retrieval in predicting the latency of peak HFB activity. During successful encoding, PHRC was active first, followed by HC and then frontal regions. By contrast, during unsuccessful encoding, there was no difference in the timing of PHRC and HC activity. The HC’s HFB power also differentiated successful over unsuccessful encoding at the time of its peak activity. During successful retrieval, HC was active first, followed by PHRC and frontal regions in tight succession. By contrast, during unsuccessful retrieval, there was no difference in the timing of PHRC and HC activity, and frontal regions were more spread out in time. These results suggest that during encoding, HC receives information from PHRC, functioning as an extension of perception. During retrieval, HC functions as part of a top-down circuit involving the prefrontal cortex.

Topic Area: LONG-TERM MEMORY: Episodic