Correction: Exploring the role of repetitive negative thinking in the transdiagnostic context of depression and anxiety in children

BMC Psychology (2025) 13:902

10.1186/s40359-025-03169-y

Following publication of the original article, the authors wish to correct an error in the “Results” section of the Abstract. Due to an oversight during the final proofreading stage, there was an inaccuracy in the description of the mediation effect results.

The original “Results” section of the Abstract was as follows:

“RNT and its factors were significantly positively correlated with depressive and anxiety symptoms (r = 0.56–0.68, p < 0.001) and mediated 12.94% of their bidirectional relationship (95% CI, 10.60%-15.27%). Network analysis revealed that RNT’s core features exhibited the highest bridge betweenness and bridge expected influence, indicating a critical mediating role in the co-occurrence of symptoms. The random forest model showed optimal predictive performance (AUC = 0.90, recall = 0.95), supporting its applicability for early screening.”

While the current (correct) “Results” section of the Abstract is as follows:

“RNT and its factors were significantly positively correlated with depressive and anxiety symptoms (r = 0.56–0.68, p < 0.001) and mediated 12.94% and 12.73% of their bidirectional relationship (with depression and anxiety as independent variables, respectively), with the same 95% CI: 8.78%–18.43% (p < 0.001). Network analysis revealed that RNT’s core features exhibited the highest bridge betweenness and bridge expected influence, indicating a critical mediating role in the co-occurrence of symptoms. The random forest model showed optimal predictive performance (AUC = 0.90, recall = 0.95), supporting its applicability for early screening.”

These corrections clarify the specific mediation effect values for the bidirectional relationship when considering depression and anxiety as independent variables, respectively, along with the accurate 95% confidence interval. They do not affect the main conclusions of the article regarding the positive correlation between RNT and symptoms, the critical mediating role of RNT’s core features, or the predictive performance of the random forest model.

All authors have reviewed and approved the corrections. Correspondence regarding this correction should be addressed to Xi Luo, Email: luorosi@126.com.

The authors thank you for reading this erratum and apologize for any inconvenience caused.