Deficient prefrontal-amygdalar connectivity underlies inefficient face processing in adolescent major depressive disorder

Adolescence represents a critical developmental period where the prevalence of major depressive disorder (MDD) increases. Aberrant emotion processing is a core feature of adolescent MDD that has been associated with functional alterations within the prefrontal-amygdala circuitry. In this study, we tested cognitive and neural mechanisms of emotional face processing in adolescents with MDD utilizing a combination of computational modeling and neuroimaging. Thirty adolescents with MDD (age: M = 16.1 SD = 1.5; 14, 20 females) and 33 healthy controls (age: M = 16.2 SD = 1.9; 20 females) performed a dynamic face- and shape-matching task. A linear ballistic accumulator model was fit to the behavioral data to study differences in evidence accumulation. We used dynamic causal modeling (DCM) to study effective connectivity in the prefrontal-amygdala network to reveal the neural underpinnings of cognitive impairments while performing the task. Face processing efficiency was reduced in the MDD group and most pronounced for ambiguous faces with neutral emotional expressions. Critically, this reduction was related to increased deactivation of the subgenual anterior cingulate (sgACC). Connectivity analysis showed that MDD exhibited altered functional coupling in a distributed network spanning the fusiform face area–lateral prefrontal cortex–sgACC and the sgACC–amygdala pathway. Our results suggest that MDD is associated with impairments of processing nuanced facial expressions. Distributed dysfunctional coupling in the face processing network might result in inefficient evidence sampling and inappropriate emotional responses contributing to depressive symptomology. Our study provides novel insights in the characterization of brain function in adolescents with MDD that strongly emphasize their role of aberrant prefrontal-amygdala interactions during emotional face processing.