Hippocampal Features Supporting Episodic Memory and Simulation in Temporal Lobe Epilepsy
Abstract number :
1.342
Submission category :
5. Neuro Imaging / 5A. Structural Imaging
Year :
2025
Submission ID :
888
Source :
www.aesnet.org
Presentation date :
12/6/2025 12:00:00 AM
Published date :
Authors :
Presenting Author: Rebecca Roth, BA – Emory University
Victor Karpychev, PhD – Emory University
Jessica Decker, DO – Oregon Health & Science University
Greydon Gilmore, PhD – Emory University School of Medicine
Nimit Desai, MD – Medical University of South Carolina
Daniel Schacter, PhD – Harvard University
Leonardo Bonilha, MD, PhD – University of South Carolina
Ezequiel Gleichgerrcht, M.D., Ph.D. – Emory University
Rationale: Episodic memory (EM) involves recalling past experiences, while episodic simulation (ES) refers to imagining possible future or hypothetical events. These abilities are central to real-life decision-making, shaping future behavior through goal setting, problem-solving, and self-regulation. While previous studies have established EM and ES deficits in TLE and the neurobiological correlates in healthy individuals, no prior work has jointly examined EM and ES in TLE using advanced neuroimaging approaches, particularly leveraging fine-grained hippocampal segmentation.
Methods: The current study was approved by the Institutional Review Board at the Medical University of South Carolina according to the Declaration of Helsinki. 31 patients with unilateral TLE (n= 17 left, n=14 right) underwent 3T magnetic resonance imaging (MRI) as well as completed an episodic memory and simulation task. Task performance was evaluated as the number of vivid details produced while controlling for verbal output. We used HippUnfold, a novel pipeline that unfolds the hippocampus, segments into its subfields, and provides metrics including thickness and gyrification, to extract features of the hippocampus. We first examined hippocampal subfield volumes as related to EM and ES performance, and second, assessed hippocampal thickness and gyrification using vertex-wise regression with permutation-based clustering.
Results: Linear models revealed that left CA1 (t=2.82, p=0.01) and right CA2/3 (t=2.58, p=0.02) interacted with laterality to predict EM performance, thus subfield volumes were only predictive in right sided patients (Fig. 1). Across all patients, increased thickness in the left hippocampus predicted better episodic memory performance (13 clusters) while decreased thickness in the right hippocampus predicted better episodic memory performance (9 clusters). Further, increased gyrification of the left hippocampus predicted better simulation performance (5 clusters; Fig. 2).
Conclusions: Using fine-grained hippocampal segmentation, we examined hippocampal volumes and anatomostructural features supporting episodic memory and simulation. We found that specific subfield volume (left CA1, right CA2/3) predicted EM performance in right TLE patients. Additionally, higher thickness in the left hippocampus and lower thickness in the right hippocampus supported EM while gyrification of the left hippocampus supported ES. Taken together, our study elucidates hippocampal features that support EM and ES in TLE.
Funding: Medical University of South Carolina South Carolina Translational Research Institute Discovery Pilot Grant Award #2212 (2022-2023). EG was supported by awards UL1TR002378 and KL2TR002381 by the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH). Imaging data were obtained through NINDS R01NS110347.
Neuro Imaging