Emotion Processing in Schizophrenia: Insights From a Brain Imaging Study Comparing Patients, Siblings, and Healthy Controls

Abnormal facial emotion recognition has been proposed as a potential endophenotype for schizophrenia, as both patients with schizophrenia and their healthy relatives show difficulties in recognizing negative facial emotions. In psychiatric disorders, brain functioning is considered highly informative for endophenotype research. However, recent studies have raised questions about our current understanding of the neural correlates of emotion recognition across the schizophrenia spectrum, pointing out two major limitations in previous research. First, individual fMRI studies and meta-analyses have predominantly used neutral stimuli as a comparator in emotional tasks. Yet, recent evidence indicates that neutral stimuli are not perceived as truly neutral by patients with schizophrenia or their first-degree relatives, thereby calling into question the interpretability of earlier findings. Second, few studies have explored brain connectivity in response to negative emotional faces in healthy relatives of patients, even though emotions are processed by a complex network of interconnected brain structures. This study aims to investigate the neural mechanisms underlying emotion processing in patients with schizophrenia and their siblings, focusing on brain activation and functional connectivity during the perception of negatively valenced facial expressions. We employed a more ecologically valid task compared with previous studies, incorporating emotional faces within an emotionally charged context. By employing a well-matched control condition and examining connectivity patterns within emotion-processing networks, we seek to address the limitations of prior research. 118 participants (37 patients with schizophrenia, 39 siblings, and 42 healthy controls) underwent functional magnetic resonance imaging (fMRI) scanning while performing an emotional task involving negatively valenced faces and control conditions. Behavioral performance was assessed using the Balanced Integration Score (BIS) to evaluate speed-accuracy tradeoffs. fMRI data were analyzed for brain activation, using a complementary approach based on frequentist and Bayesian statistics, as well as for functional connectivity, using generalized psychophysiological interaction (gPPI) analysis with the right and left amygdala as seed regions. Behavioral analyses revealed significant group differences, with both patients and their siblings displaying lower BIS scores compared with controls. Exploratory between-group fMRI analyses revealed that, compared with controls, siblings exhibited decreased brain responses to negative faces in a cluster encompassing the right superior temporal gyrus and the right postcentral gyrus. Bayesian analyses showed that, as compared with controls, patients seem to display similarly impaired activation in the right postcentral gyrus. In most of the rest of the brain, including the amygdala, Bayesian analyses indicated an absence of group differences for both patients and siblings compared with controls. Functional connectivity analyses revealed that siblings had stronger task-related connectivity between the right amygdala and the right cuneus compared with controls. Our findings provide evidence that while the amygdala appears to respond to negative faces similarly among controls, siblings, and patients, an endophenotypic pattern may be present in the right postcentral gyrus. Furthermore, these results suggest that activation-based analyses may not fully capture the neural abnormalities associated with emotion processing in relatives of patients with schizophrenia. We hypothesize that the increased connectivity between the right amygdala and the right cuneus may reflect a compensatory mechanism in siblings. CLINICAL REGISTRATION: https://clinicaltrials.gov/study/NCT02834208?term=schizoimagen&rank=1.