Aberrant hippocampal-cortical connectivity and network coupling in facial emotion recognition-based subtypes of depression.

BACKGROUND

Facial emotion recognition (FER), an essential component of emotion processing that influences social functioning and interpersonal relationship satisfaction, plays a key role in major depressive disorder (MDD). This study aimed to investigate the heterogeneity of FER performance within individuals with MDD and its relationship with whole-brain functional connectivity (FC) using multivariate methods.

METHODS

A total of 202 patients with MDD and 202 healthy controls (HCs) were included in the study and completed FER assessments. Among them, 158 patients with MDD and 128 HCs underwent resting-state functional magnetic resonance imaging (rs-fMRI) scanning. Data-driven clustering was employed to identify FER-based clusters within the MDD group based on recognition performance. Group differences in FC were explored at both the edge and large-scale network levels. Partial least squares (PLS) correlation analysis was then applied to investigate the association between whole-brain FC patterns and FER performance.

RESULTS

Clustering analysis identified three MDD subgroups characterized by progressively decreasing FER performance, accompanied by a corresponding stepwise reduction in overall network connectivity strength. Comparisons between subgroups highlighted the crucial involvement of hippocampal and prefrontal regions, as well as subcortical and visual systems. The PLS results revealed a distinctive FC pattern associated with FER performance.

CONCLUSIONS

Our findings suggest that multilevel neural alterations, including disrupted connectivity within the hippocampal-prefrontal-limbic circuitry and abnormal coupling between large-scale information integration and sensory-motor networks, may collectively impair affective information processing and contribute to individual differences in FER observed among individuals with MDD.