Authors :
Presenting Author: Adam Khalef, BS – Emory University
Adam Dickey, MD, PhD – Baylor College of Medicine
Evan Brady, BS – Emory University School of Medicine
Taylor Shade, BS – Emory University School of Medicine
Rebecca Roth, BS – Emory University
Vex Hutton, BS – Emory University
Edward Valentin, PhD – Emory University School of Medicine
Vince Calhoun, PhD – Georgia State University
Stephan Hamann, PhD – Emory University
Daniel Drane, PhD – Emory University School of Medicine
Rationale:
Neuropsychological research in temporal lobe epilepsy (TLE) has long emphasized memory outcomes, but growing evidence indicates that socio-emotional abilities are also disrupted. These skills are critical for daily functioning, yet remain underexamined in surgical populations. We evaluated how TLE patients recognize complex emotional states from limited facial information, focusing on the eye region, to clarify the nature of socio-emotional deficits.
Methods:
This study analyzed socio-emotional data from a cohort of 60 patients with TLE who underwent a stereotactic laser amygdalohippocampotomy (SLAH) procedure and from 71 healthy controls (HC). The side of seizure onset was determined by EEG, neuroimaging (3T MRI, PET), and neuropsychological testing. A subset of patients also underwent intracranial monitoring to confirm seizure onset location. Social-emotional processing was evaluated using the Reading the Mind in the Eyes Test (RMET) alongside a depression questionnaire to evaluate potential covariate effects. Item-level accuracy was calculated for each participant and compared against normative valence ratings from Hudson et al. (2020). Group comparisons (TLE vs. controls; left vs. right TLE) were examined using independent-samples t-tests. Associations between item valence and accuracy were evaluated with Pearson correlations, computed separately for controls, the overall TLE cohort, and left and right TLE subgroups.Results:
Controls outperformed epilepsy patients on the RMET: 76% vs 62% accuracy (p< 0.001, unpaired t-test). Using Hudson et al.’s normative valence ratings as a continuous predictor, controls showed a significant coupling between valence and accuracy (r = 0.53, p=0.008), whereas the epilepsy group displayed a weakened association (r = 0.24, p=0.16). This relationship differed by laterality: right TLE patients retained a moderate and significant positive correlation (r = 0.36, p=0.03), while left TLE patients showed no relationship (r = –0.01, p=0.95), indicating a flat performance trend across the valence continuum. The mean Beck Depression Inventory scores were elevated in epilepsy patients vs controls (14.8 epilepsy vs 2.4 control, p< 0.0001, unpaired t-test) but did not differ between left (13.3) and right (15.9) groups (p = 0.31), suggesting that depressive symptoms do not explain the hemispheric dissociation.