Effective connectivity during faces processing in major depression: Distinguishing markers of pathology, risk, and resilience

BackgroundAberrant brain connectivity during emotional processing, especially within the fronto-limbic pathway, is one of the hallmarks of major depressive disorder (MDD). However, a lack of systematic approaches in previous studies made it difficult to determine whether a specific alteration in brain connectivity reflects a cause, correlate, or effect of the disorder. The current study aimed to investigate neural mechanisms that correspond to disease, risk and resilience in major depression during implicit processing of emotion cues. MethodsForty-eight patients with MDD, 49 first-degree relatives of patients with MDD and 103 healthy controls performed a face-matching task during functional magnetic resonance imaging. We used dynamic causal modelling to estimate task-dependent effective connectivity at the subject level. Parametric empirical Bayes was then performed to quantify group differences in effective connectivity. ResultsDepressive pathology was associated with decreased effective connectivity from the left amygdala and left dorsolateral prefrontal cortex to the right fusiform gyrus, whereas familial risk for depression corresponded to decreased connectivity from the right orbitofrontal cortex to the left insula and from the left orbitofrontal cortex to the right fusiform gyrus. Resilience for depression was related to increased connectivity from the anterior cingulate cortex to the left dorsolateral prefrontal cortex. ConclusionsOur results suggest that the depressive state alters top-down control of higher visual regions during the processing of emotional faces, whereas increased connectivity within the cognitive control network promotes resilience to depression.