IntroductionBoredom proneness is a relatively stable dispositional vulnerability associated with adolescents’ problematic social media use (PSMU). However, it remains unclear whether this association reflects stable between-person differences, within-person temporal dynamics, or both. This study examined the reciprocal longitudinal associations between boredom proneness and PSMU among Chinese adolescents.MethodsFour-wave data were collected from 2020 to 2022 from 2,163 adolescents aged 12–19 years at baseline (mean age = 14.54 years, SD = 1.49; 47.4% female) attending four secondary schools in Guangzhou, China. A random-intercept cross-lagged panel model (RI-CLPM) was used to separate stable between-person differences from occasion-specific within-person deviations while controlling for gender, age, paternal education, and maternal education. Ethical approval and written informed consent were obtained.ResultsAt the between-person level, the stable components of boredom proneness and PSMU were positively associated (B = 0.195, p < 0.001). At the within-person level, an asymmetric prospective pattern emerged. Higher-than-expected boredom-proneness scores predicted higher-than-expected PSMU only from T1 to T2 (B = 0.073, p < 0.05), whereas higher-than-expected PSMU consistently predicted higher-than-expected boredom-proneness scores across all three adjacent intervals (B = 0.074, 0.108, and 0.103; all p < 0.05). Both constructs also showed significant autoregressive effects.DiscussionThe association between boredom proneness and PSMU comprised both stable between-person co-occurrence and asymmetric within-person temporal dynamics. PSMU consistently preceded subsequent elevations in boredom-proneness scores, whereas the reverse pathway was interval-specific. These findings do not establish causality or indicate that PSMU causes momentary state boredom or enduring personality change. They may help inform time-sensitive prevention and support strategies by distinguishing stable vulnerability from occasion-specific variation. Because all waves occurred during the COVID-19 pandemic, replication in post-pandemic cohorts is needed.