KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 44 2026 04 10 Dynamic Bayesian Networks of Self-Esteem Fluctuations and Social Feedback Sensitivity Dynamic Bayesian Networks of Self-Esteem Fluctuations and Social Feedback Sensitivity 1 10 10.61838/kman.jayps.5315 EN Journal Article 2025 10 29 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The present study aimed to model the dynamics of state self-esteem fluctuations and social feedback sensitivity using Dynamic Bayesian Networks.</p> <p><strong>Methods and Materials: </strong>A quantitative intensive longitudinal design was implemented with 287 university students from Slovenia who completed a 21-day ecological momentary assessment protocol. Participants responded to five semi-random prompts per day assessing state self-esteem, perceived social feedback valence, and negative affect, yielding 26,874 momentary observations. Baseline measures included trait self-esteem, rejection sensitivity, depressive symptoms, and social anxiety. Data were person-mean centered and preprocessed to ensure stationarity. A multilevel vector autoregressive model was first estimated to identify preliminary lagged associations. Subsequently, Dynamic Bayesian Networks were constructed using a two-time-slice framework with hill-climbing structure learning and Bayesian Information Criterion scoring. Bootstrapped model averaging (1,000 resamples) ensured stability of directed edges. Hierarchical Bayesian extensions examined moderation by trait rejection sensitivity, and predictive accuracy was evaluated via 10-fold cross-validation.</p> <p><strong>Findings: </strong>State self-esteem demonstrated significant autoregressive stability (β ≈ 0.50, p &lt; 0.001). Positive social feedback prospectively predicted increases in subsequent state self-esteem (β ≈ 0.24, credible intervals excluding zero), whereas negative affect predicted decreases (β ≈ -0.20, p &lt; 0.001). Reciprocal effects between self-esteem and negative affect formed a significant feedback loop. Trait rejection sensitivity significantly moderated the feedback-to-self-esteem pathway, amplifying evaluative reactivity (posterior mean ≈ 0.15, credible intervals excluding zero). Bootstrapped inclusion probabilities for key edges exceeded 90%, and cross-validated predictive performance demonstrated satisfactory out-of-sample accuracy (RMSE ≈ 0.41).</p> <p><strong>Conclusion: </strong>Self-esteem operates as a dynamically regulated state embedded within a probabilistic network of social and affective processes, where social feedback functions as a central driver and individual differences in rejection sensitivity shape the magnitude of evaluative updating across time.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jayps/article/download/5315/9664