Age-related differences in theta oscillations across development: insights from intracranial EEG and brain structure-function relationships

Poster Session F - Tuesday, April 16, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Zachariah Cross¹ (zachariah.cross@northwestern.edu), Samantha Gray¹, Adam Dede¹, Qin Yin^2,17, Parisa Vahidi², Elias Rau³, Christopher Cyr¹, Ania Holubecki¹, Eishi Asano², Jack Lin⁴, Olivia Kim McManus⁵, Shifteh Sattar⁵, Ignacio Saez^4,6, Fady Girgis^4,7, David King-Stephens^8,9, Peter Weber Weber⁹, Kenneth Laxer⁹, Stephan Schuele¹, Joshua Rosenow¹, Joyce Wu^1,10, Sandi Lam^1,10, Jeffrey Raskin^1,10, Kurtis Auguste^11,12, Edward Chang¹¹, Ammar Shaikhoun¹³, Peter Brunner¹⁴, Jarod Roland^14,, Rodrigo Braga¹, Robert Knight¹⁶, Noa Ofen^2,17; ¹Northwestern University, ²Wayne State University, ³Ruhr University Bochum, ⁴University of California, Davis, ⁵University of California, San Diego, and Rady Children’s Hospital, ⁶Ichan School of Medicine at Mount Sinai, ⁷University of Calgary, ⁸California Pacific Medical Center, ⁹Yale University, ¹⁰Ann & Robert H. Lurie Children’s Hospital of Chicago, ¹¹University of California, San Francisco, ¹²UCSF Benioff Children’s Hospital, ¹³Ohio State University and Nationwide Children’s Hospital, ¹⁴Washington University in St. Louis, ¹⁵St. Louis Children’s Hospital, ¹⁶University of California, Berkeley, ¹⁷University of Texas at Dallas

Neural oscillations change across the human lifespan, concurrent with changes in cognition and brain structure, emphasizing their importance in understanding brain development. Scalp-EEG is the primary tool in developmental neuroscience research. However, it cannot access memory-related brain regions, such as hippocampus. Here, we isolated slow (~1.5 – 4.5 Hz) and fast theta (~4.5 – 8 Hz) frequencies in a developmental intracranial EEG (iEEG) cohort (n = 83, n channels = 4618; Mage = 16.40 years, range = 5.93 – 54 years, 51 males). Using task-based (i.e., attending to to-be-remembered visual stimuli) and task-free recordings, we quantified age-related differences in slow and fast theta frequencies in primary sensory, limbic, and association cortices. Mixed-effects regressions revealed that in precentral gyrus, slow theta slowed with increasing age during task-free but not task-based states (p =.002), while the inverse was observed for fast theta (p =.03). In MTL, slow theta slowed with age irrespective of task state (p =.001). In dorsolateral prefrontal cortex, fast theta slowed with age irrespective of task state (p =.005). In relating structure-to-function, for both slow and fast theta in MTL, we observed significant interactions between gray matter volume and task state. Yet, despite the importance of MTL and prefrontal theta oscillations to memory, we found no significant relationship in these regions between age and theta frequencies on recognition accuracy, suggesting age-related variability in theta frequencies reflects a natural feature of development. These findings indicate that theta varies across development, differs by task state, and relates to brain structure.

Topic Area: LONG-TERM MEMORY: Development & aging