Fig. 3: A closer look at interactions between PFC subregions and their relation to symptom severity.

Scatter plots from GEEs show the relationships between directed connectivity and symptom severity. Depression symptom severity was measured by the computerized adaptive test (CAT-DI: y-axes). The x-axes indicate directed connectivity between the subregions labeled in the legends. A Symptom severity was positively correlated with directed connectivity between the left OFC and dlPFC (p < 0.001). B Directed connectivity between the right orbitofrontal cortex (OFC) and dorsolateral prefrontal cortex (dlPFC) were not significantly linked to symptom severity. C There were no effects of directed connectivity between the left OFC and anterior cingulate cortex (ACC) on MDD severity (p > 0.10). D Conversely, directed connectivity from the right OFC to the right ACC was positively correlated with symptom severity (red line: p < 0.001). No negative correlation was observed in the reverse direction (blue line: p = 0.293). E CAT-DI scores positively correlated with directed connectivity from the left dlPFC to the left ACC (p = 0.008) and from ACC to dlPFC (p = 0.020). There was also a significant interaction between these directed connectivity pathways, where symptom severity increased when information flow between both regions increased simultaneously (p = 0.020). F There were no significant relationships between directed connectivity among the right dlPFC and ACC. These findings expand our understanding on the role of PFC subregions in depression. Regression lines were fit to the data using robust (bi-square) regression for visualization purposes. Dotted lines indicate 95% confidence bounds. Kernel density plots on the x- and y-axes display probability distributions of the corresponding variables. All results were false-discovery rate (FDR) corrected for multiple comparisons across hemispheres. OFC orbitofrontal cortex, dlPFC dorsolateral prefrontal cortex, ACC anterior cingulate cortex.