Abstract
Adolescents' depressive social withdrawal is a relevant concern for mental health professionals, and it is widespread among community teenagers in form of subclinical symptoms. Different studies suggest that insecure attachment representations increase the adolescents' likelihood to show symptoms of withdrawal (e.g., loneliness). This study explored the effect of the general attachment internal working model (IWM) and the independent and cumulative effects of the specific attachment representations of parents-in terms of secure base/safe haven-and peers on adolescents' withdrawal. Additionally, the mediation of peer attachment on the effect of parental attachment on symptoms was explored. All analyses were conducted controlling for the difference between teenagers living with parents together or divorced/separated, as children of divorcees are considered more exposed to stressors. Ninety-one adolescents aged 12-17 years old were assessed online during the COVID pandemic period, employing the Youth Self-Report to assess withdrawal and the Friends and Family Interview to assess attachment-general IWM and attachment-specific representations. Results show no influence of parents together/separated or of the general IWM on withdrawal, but higher parent secure base/safe haven and peer attachment cumulatively predicted 10-21% less withdrawal. Moreover, more positive peer attachment mediated 61% of the effect of the parental secure attachment on withdrawal, revealing an indirect effect of parental attachment on withdrawal through peer attachment. In conclusion, both parents and peers are influential on adolescent mental health, and fostering positive peer relationships can buffer the effect of dysfunctional family relationships on teenagers' withdrawal.