EP 63. Altered functional connectivity of action and emotion regulation networks in Schizophrenia

Abstract

Emotion regulation (ER) and action regulation (AR) have been shown to rely on two broad networks ( Langner et al., 2014 ). While ER and AR networks (ERN/ARN) are mostly distinct from each other, there are also overlapping regions supporting a common underlying cognitive control mechanism. Several studies in schizophrenic patients (SZ) provide evidence for deficits in ER and AR ( van der Meer et al., 2014 , Minzenberg et al., 2009 ). Given these dysfunctions, we examined functional connectivity (FC) aberrations within these networks in SZ and associations with behavior. We used data from 60 SZ and 51 healthy controls (HC) obtained from the COBRE study ( COBRE ). ER ability was assessed with MSCEIT™ (Managing Emotions) questionnaire. AR was assessed using the following tasks: simple and complex working memory span, identical-pairs continuous performance task (CPT-IP). Resting-state fMRI data were preprocessed and de-noised ( COBRE ; Salimi-Khorshidi et al., 2014 ). For both ERN and ARN, FC between each pair of network nodes was determined by intercorrelating the eigenvariate of each seed region’s BOLD time series. Confounds of age, gender and movement were removed as were outliers in behavioural measures and FC. For each network, permutation tests were used to test for group differences in FC between all those nodes whose connectivity was of at least medium effect size (r ⩾ .24) in either or both groups. Pearson correlation analyses were performed to investigate whether ER abilities and behavioral measures of AR were associated with FC between ERN and ARN nodes, respectively. Results revealed significantly reduced scores in SZ (vs. HC) for ER and AR abilities. FC analysis showed several significant (FWE-corrected) group differences (SZ vs. HC) for both networks: For the ERN, FC between left amygdala (lAmy) and posteromedial prefrontal cortex (pmPFC) as well as lAmy and left Inferior frontal gyrus was decreased in SZ. For the ARN, FC of left dorsolateral prefrontal cortex (lDLPFC) with left inferior temporal gyrus (lITG), right intraparietal sulcus, right dorsal premotor cortex (rPMd), of left insula with right temporo-parietal junction (rTPJ) and rPMd, as well as of right insula with rTPJ was decreased in SZ. Furthermore, FC between lAmy and pmPFC in the ERN as well as lDLPFC and lITG in the ARN was positively correlated with CPT-IP scores. Our results indicate hypoconnectivities within the ERN and ARN in SZ and associations of attention with FC in both networks. In particular, the FC of the ARN’s lDLPFC node appears to play a major role in SZ. This could indicate a disturbance of neural mechanisms of top-down modulation, required for appropriate AR. Hypoconnectivity between lAmy and frontal regions within the ERN suggests aberrant cognitive control processes in ER, in line with previous research ( van der Meer et al., 2014 ). The positive correlation of CPT-IP performance with FC between lAmy and pmPFC within the ERN and with FC between lDLPFC and lITG within the ARN, suggests that successfully sustaining attention may depend on the efficient cognitive control of both emotion and action.