Abstracts

Rationale: Patients with epilepsy are at a higher risk of developing psychiatric comorbidities, such as depression and anxiety (Mula et al., 2021). Functional neuroimaging studies in patients with primary psychiatric diagnoses suggest abnormality in brain networks overlapping with those involved in mesial temporal epilepsy, such as the hippocampus and amygdala. A promising tool to regulate dysfunctional neurological circuits is targeted electrical brain stimulation (EBS). The human amygdala is ideal for neuromodulation because it is heavily implicated in various cognitive functions, such as affective valence and arousal (Bujarski et al., 2022).



This study has two objectives: first, examine whether EBS of the basolateral nuclei of the human amygdala can modulate emotion perception. Second, investigate the brain networks involved in emotion perception.


Methods:

We recorded intracranial local field potentials in nine patients with refractory focal epilepsy patients using stereo-electroencephalography (SEEG). Patients without major psychiatric diagnoses were enrolled.

Patients viewed emotionally salient pictures during a behavioral task and rated them on valence and arousal scales (1-7 Likert-type) in a randomized block design. Pictures were selected from the OASIS dataset and balanced across valence (negative and positive) and arousal (low and high) categories (Kurdi et al., 2017). During each block, we presented ten images; the second half of the images in each block were presented with EBS at 200 Hz (200us, biphasic pulses for 3 seconds). The amplitude of the EBS was determined for each patient to avoid triggering after-discharges or ictal activity (range: 0.2 to 2 mA).

Results:

To examine the effects of EBS on behavior, patients were divided into low (0-0.7 mA), medium (0.7-1.4 mA), and high stimulation (1.4-2 mA) groups. We found a statistical difference across stimulation groups for positive valence images (ANOVA: p< 0.001), where individuals in the medium stimulation amplitude reported the lowest mean valence ratings. Similarly, we found that for low arousal images, patients in the medium stimulation amplitude reported the lowest mean arousal ratings (ANOVA: p< 0.001).