A Network Analysis of Cognitive Flexibility, Emotion Regulation, and Symptom Severity in Patients with Major Depression.

Abstract

This study aims to investigate the network structure linking cognitive flexibility, cognitive emotion regulation strategies, and depressive symptom severity in patients with major depressive disorder and identify the most central and bridge components of this network. This study was conducted using a descriptive and relational design. 188 patients diagnosed with major depression were included in the study. The BDI, CFI and CERQ were used as data collection tools. Network analysis revealed a structure containing 12 nodes and 33 active edges out of 66 possible edges (EBICglasso, γ = 0.50; resulting sparsity = 0.50). In terms of strength centrality, Positive Reappraisal (z-strength = 1.864), Refocus on Planning (z-strength = 1.176), and Acceptance (z-strength = 1.035) formed a statistically indistinguishable top-strength cluster, indicating that these three adaptive strategies were the most central nodes in the network. In terms of Expected Influence, the CFI Control subdimension (EI = - 1.821) and BDI (EI = - 1.741) showed the strongest negative values. Bootstrap stability analysis indicated modest stability for strength centrality (CS-coefficient ≈ 0.27) and low stability for betweenness and closeness, indicating that centrality findings should be interpreted as exploratory. Bridge centrality analyses revealed that BDI (0.580), CFI Control (0.570), and CFI Alternatives (0.540) were the strongest cross-domain connectors. Refocus on Planning and Alternatives exhibited positive bridge expected influence, indicating facilitative cross-domain effects, while Control, Catastrophizing, and BDI exhibited negative bridge expected influence, indicating a suppressive cross-domain effect. Cognitive restructuring interventions may show limited efficacy unless perceived loss of control and, to a lesser extent, catastrophizing which appear to function as suppressive hubs, are addressed first; however, given the cross-sectional design, this staged-intervention hypothesis remains speculative and requires longitudinal validation. Notably, the Control-BDI edge received weaker Bayesian support (Pr(H1) = 0.213) than the network's core adaptive-strategy connections, warranting caution in interpreting this specific association. These findings tentatively suggest, rather than establish that interventions aimed at restoring perceived control may represent a promising therapeutic target for enhancing adaptive emotion regulation in major depression.