Repeatability of facial emotion processing over 12 weeks in healthy participants

Few studies have assessed influences of mood state on facial emotion processing changes. We examined if the repeatability of fMRI facial emotional processing in healthy participants (HCs) is affected by mood state changes and hypothesized that fMRI activation would be stable but may be influenced by mood state fluctuations. In a multi-site study, thirty-two HCs underwent emotion faces task (EFT) fMRI and completed a post-scan facial emotion rating and the Profile of Mood States (POMS) ∼ 12 weeks apart. FMRI data were processed using AFNI software. POMS and behavioral data were analyzed using repeated measures ANOVA and the sign test, as appropriate. Paired-samples t-tests, covarying for site, assessed fMRI responses for facial emotions (Happy, Fearful, Sad, Neutral) and Arousal. Intraclass correlation coefficient (ICC) examined repeatability of POMS and fMRI responses; regression analyses examined relationships between mood changes and fMRI activations. There were no significant differences between visits in POMS, EFT performance or post-scan ratings accuracy. POMS scores were stable (ICC ≥ 0.74). FMRI activation exhibited repeatability that was fair or better (ICC ≥ 0.4). EFT activation changes were not significantly related to mood state changes. Our results provide evidence for acceptable EFT fMRI test-retest reliability over 12 weeks, without significant influence of mood state variability.