Abstracts

Stress affects the neural response to and recognition of emotional facial expressions in left temporal lobe epilepsy

Abstract number : 2.217
Submission category : 5. Neuro Imaging
Year : 2011
Submission ID : 14950
Source : www.aesnet.org
Presentation date : 12/2/2011 12:00:00 AM
Published date : Oct 4, 2011, 07:57 AM

Authors :
J. B. Allendorfer, H. Heyse, J. P. Szaflarski

Rationale: Compared to healthy individuals, epilepsy patients report higher levels of psychological stress and may have alterations in emotional processing. However, the role of stress on emotional processing in temporal lobe epilepsy is unclear. Our objective was to assess the how stress affects neural and behavioral responsiveness to emotional faces in LTLE patients.Methods: 5 LTLE patients who believe that stress is an important factor in their seizure control (+S), 5 LTLE patients who do not (-S), and 7 healthy controls (HC) were included. During fMRI (TR=3s), subjects were presented a pseudo-randomized series of faces with different expressions (happy, fearful, sad, neutral; 30 each) and instructed to press 1 for male and 2 for female. After scanning, subjects were asked to indicate the emotion for each face. FMRI data pre-processing (image coregistration, spatial normalization), single-subject and group statistical modeling, and visualization were performed using AFNI. Subject s fMRI response to emotion (happy, fearful, sad) was determined using GLM (contrast with neutral faces). 1-sample t-tests were performed to assess emotion-related activation in each group, followed by group comparisons (+S vs. S; +S vs. HC; S vs. HC) to assess activation differences for each emotion; accuracy on post-scan rating of facial expressions was determined.Results: The 3 groups showed differential patterns of activation for each emotion (happy, fearful, sad) with LTLE groups showing the greatest activation for fearful. +S showed the greatest activation in the fusiform gyrus and hippocampus/parahippocampus. Comparison of LTLE groups showed: increased frontal, temporal and occipital activation for S than +S for fearful; increased frontal and striatal activation for S than +S for sad, and increased posterior cingulate activation for +S than S; increased frontal, posterior cingulate and occipital activation for S than +S for happy, and increased temporal and left insular activation for +S than S. Comparison of S with HC showed overall greater activation in S than HC for all emotions. Comparison of +S with HC showed increased activation for +S in left parahippocampus for happy and fearful, and increased BA10 activation for sad; HC showed increased activation for all emotions in medial frontal regions, in addition to increased activation in the caudate for sad. Accuracy on rating emotional faces was consistently less accurate for +S compared to HC and S subjects. +S and S subjects showed a trend towards a significant difference in rating accuracy for neutral faces (p=0.064) and faces overall (p=0.064).Conclusions: LTLE and HC groups recruit different brain regions for processing emotional faces, with the greatest activation differences occurring between +S and S patients. The differences in emotional face processing between +S and S patients may underlie the decreased ability of S patients to correctly recognize types of emotion. These results suggest stress may exert neuromodulatory effects on emotional processing in LTLE patients.
Neuroimaging