A network analysis of the relationship between temperament characteristics and problem behaviors in preadolescents

Temperament, defined as relatively stable, innate tendencies in emotional reactivity and self-regulation, is a robust correlate of childhood problem behaviors. However, while most studies focus on isolated symptoms, few have examined how specific temperamental traits simultaneously relate to the complex interactions between internalizing and externalizing systems. Using network analysis, this study explored the relationships among temperamental traits, internalizing problems (depressive symptoms), and externalizing problems (aggressive behavior) based on self-report data from 642 preadolescents. We estimated a Gaussian Graphical Model via the EBICglasso approach, treating 12 temperament, 5 depression, and 4 aggression dimensions as nodes, and subsequently calculated their Expected Influence (EI) and Bridge Expected Influence (BEI). Results revealed that negative self-esteem emerged as the most central node, followed by displaced aggression and negative mood (Depressive Symptoms). In terms of bridge centrality, depressive mood (Temperament) exhibited the highest bridge expected influence, emerging as the primary bridge node connecting the internalizing and externalizing systems, followed by hostility and displaced aggression, which served as significant bridge hubs connecting the temperament, depression, and aggression communities. Collectively, our findings provide an initial characterization of the conditional association patterns among these variables.