Gender differences in symptom interactions between problematic smartphone use and social anxiety in adolescents: a network analysis

BackgroundIncreased prevalence of problematic smartphone use (PSU) in adolescents, results in a cycle of interaction between PSU and social anxiety. However, it is still unknown whether PSU and social anxiety symptoms have interacted among adolescents and whether there are gender differences in these symptoms. Therefore, this study investigated the gender differences in the symptom interactions between PSU and social anxiety via symptom network analysis.MethodsThis study included 2918 adolescents (52.71% boys; Mage = 14.73, SDage = 1.39) from junior and senior high schools in China. The Mobile Phone Addiction Index and Social Anxiety Scale were used to evaluate symptomatology and networks. Network analysis and network comparison tests were used to determine the network structure, centrality, bridge symptoms and gender differences in the PSU-social anxiety network among adolescents.ResultsThe most influential symptoms were "productivity loss" and "afraid of negative evaluation". "Afraid of negative evaluation" was the bridge through which PSU was related to social anxiety. Gender differences were not found in network strength but occurred in network structure. Although girls reported more social anxiety, boys had a tighter network structure. The correlation between PSU and social anxiety was greater in boys than in girls. The "inability to control craving" was particularly critical for girls, while "feeling anxious and lost" was prominent for boys.ConclusionsThe current study highlights the symptom interactions between PSU and social anxiety among adolescents and the gender differences in network structures. Further intervention that targets "afraid of negative evaluation" may disassociate the interaction between PSU and social anxiety symptoms. In particular, changing girls' cognitive ability (e.g., inhibition) and boys' negative emotions are potentially effective means of intervention. The limitations of the cross-sectional design and data-driven methodology necessitate interpreting the results with caution.