Network analysis of problematic internet use and psychopathology: Anxiety as a central symptom and sleep reduction as a bridge symptom

In adolescent populations, depressive disorders exhibit bidirectional associations with problematic internet use (PIU), a comorbidity presenting heightened complexity among clinically diagnosed individuals. Previous research has been limited in its ability to resolve multidimensional interactions among psychopathological symptoms or capture how self-reinforcing symptom networks perpetuate comorbid persistence and PIU. This study aimed to establish examine the network structure of psychopathological symptoms and PIU among clinically depressed adolescents in China, identifying core and bridge symptoms. Psychopathological symptoms and PIU were assessed using the Symptom Checklist-90 (SCL-90) and the Revised Chen Internet Addiction Scale (CIAS-R), respectively. Network analysis was employed to model PIU, psychopathological symptoms, and comorbidity network structures. Anxiety emerged as the central hub node (expected influence, EI = 1.26) in the psychopathological network, demonstrating the strongest edge weight with somatization (0.34). The PIU network was dominated by “increasing online time” (EI = 1.13) and “tolerance” (EI = 1.07). Cross-network bridge symptom analysis identified “Psychoticism dimension (SCL-90; psychotic-like/schizotypal experiences rather than frank psychosis)”, “Hostility”, “Phobic Anxiety”, " Increasing Online Time” and " Less Sleep for Internet " (bridge EI > 1) as key bridge nodes, suggesting a closely interconnected pattern linking mood-related symptoms, PIU symptoms, and functional impairment-related complaints. Global connectivity strength was higher in females than males (13.89 vs. 12.35, p = 0.03). The comorbidity observed in clinically depressed adolescents s characterized by interconnected symptom networks. Effective intervention necessitates simultaneous targeting of core hubs, notably anxiety, and bridge nodes such as sleep reduction to disrupt network stability. These findings establish a network-informed basis for gender-differentiated approaches: emotion regulation protocols for females and behavioral prevention strategies for males. Not Applicable.