Mapping family affect to adolescent short‐video addiction: A cross‐lagged panel network analysis

Abstract Short video platforms have rapidly become central to adolescents' digital lives, yet empirical work on short‐video addiction (SVA) is largely cross‐sectional and rarely embeds family dynamics within established theoretical frameworks. Most adolescent viewing occurs at home, making familial emotional environments potentially critical, yet understudied, drivers of SVA. To address this gap, the current study adopted a developmental cascade framework, using a cross‐lagged panel network approach to examine temporal associations among family ecological risks (family economic strain, parental negative mood, childhood emotional neglect), trait vulnerability (neuroticism), affective states (depressive symptoms, perceived stress), cognitive beliefs (fixed mindset, relative deprivation), and SVA. Two survey waves (6‐month interval) involving 807 Chinese middle school students ( M age = 13.57 years; SD age = 0.76; range: 11–15 years; 49.7% female) were conducted. Network analysis revealed robust predictability ( R 2 average = .32), with T1 variables explaining 24% of the variance in T2 SVA. Depressive symptoms and parental negative mood emerged as key risk drivers with the highest out‐expected influence, while perceived stress functioned as a vulnerability hub accumulating upstream influences with the highest in‐expected influence. Across waves, depressive symptoms and perceived stress predicted higher SVA. SVA also showed prospective associations with higher later fixed mindset and relative deprivation. Gender‐stratified networks revealed broadly similar structures, and supplementary serial mediation analyses provided convergent support for indirect cascade pathways. These findings are consistent with a developmental cascade account and suggest that effective prevention may complement time‐limit approaches with integrated strategies targeting family emotional climate, adolescents' stress regulation, and maladaptive cognitions.