心理科学

心理科学 ›› 2024, Vol. 47 ›› Issue (6): 1381-1391.DOI: 10.16719/j.cnki.1671-6981.20240610

• 发展与教育 • 上一篇    下一篇

童年晚期抑郁症状网络的演化及症状间的纵向关系*

陈嘉慧1, 任萍**1,2, 吕沐华1, 李添**1   

  1. 1北京师范大学中国基础教育质量监测协同创新中心,北京,100875;
    2北京师范大学认知神经科学与学习国家重点实验室,北京,100875
  • 出版日期:2024-11-20 发布日期:2024-12-24
  • 通讯作者: **任萍,E-mail: renping@bnu.edu.cn;李添,E-mail: litian@bnu.edu.com
  • 基金资助:
    *本研究得到科技创新2030（2021ZD0200500）、中央高校基本科研业务费专项资金（1243300003）和认知神经科学与学习国家重点实验室开放课题基金（CNLZD2203）的资助

Temporal Change in the Depression Network and Longitudinal Network Associations between Depressive Symptoms during Late Childhood

Chen Jiahui1,2, Ren Ping1,2, Lyu Muhua1, Li Tian1   

  1. 1Collaborative Innovation Center of Assessment for Basic Education Quality, Beijing Normal University, Beijing, 100875;
    2State Key Laboratory of Cognitice Neurosecience and Learning, Beijing Normal University, Beijing, 100875
  • Online:2024-11-20 Published:2024-12-24

PDF

摘要： 为揭示童年晚期抑郁症状网络的演化和症状间的纵向关系,对3042名小学中高年级儿童进行了一年三次的追踪测量。通过三个时间点的偏相关网络分析发现,易激惹和自我憎恨在抑郁症状网络中均呈现较高的中心性。交叉滞后网络分析表明,自我憎恨是影响网络中其他症状的核心症状;而易激惹更容易被其他症状影响;孤独感和负面身体意象对抑郁症状网络的影响存在时间特异性。这些结果揭示了童年晚期学生抑郁症状网络的核心症状及症状之间的纵向发展特征,为儿童青少年抑郁的早期防范、干预和诊疗提供了实证依据。

关键词: 抑郁症状, 童年晚期, 正则化偏相关分析, 交叉滞后网络分析, 症状学

Abstract: Depression is one of the most prevalent mental health problems among school-aged students. The psychopathology network theory conceptualizes depression as a network system of interconnected symptoms. Yet, information on the central symptoms, the structure of depressive symptoms, and the longitudinal associations between depressive symptoms is still limited among Chinese students in late childhood. Thus, using three waves of data from this group, the present study aimed to explore the structure and change of the depression network, as well as the longitudinal associations among depressive symptoms through the network analysis.
A total of 3042 Chinese 4th grade students (50.6% male, Mage = 9.36 years old, SD = 0.51 years old) were included in this study.Depressive symptoms were assessed using the short version of the Children’s Depression Inventory (CDI-S) at three time points, spaced six months apart (Time 1 (T1): November 2021, Time 2 (T2): May 2022, and Time 3 (T3): November 2022).The data were analyzed in SPSS 24.0 and R 4.2.2.For the regularized partial correlation network, the Graphical Gaussian Model (GGM) estimated the structure of the depression network at three time points.Strength was used in this study to quantify the role of each node.Regarding the cross-lagged panel network, a regression model using a series of nodes logistic regression was used to calculate auto-regressive effects (a node at T1 predicted itself at T2) and cross-lagged effects (a node at T1 predicted another node at T2).Centrality indices, specifically in-expected influence centrality and out-expected influence centrality, were used to differentiate the effects that a node predicting other nodes and being predicted by others.Additionally, network comparison tests (i.e., a network structure invariance test, a global strength invariance test, and an edge strength invariance test) were performed to assess the differences in network structure and core symptoms across three time points.
The regularized partial correlation network analysis showed that self-hatred consistently exhibited the highest strength values over time, marking it as a stable central symptom within the depression network.In addition, sadness exhibited the second-highest strength values at T1.In contrast, irritability had showed the second-highest strength values at T2 and T3, highlighting its escalating significance in the network over time.Network comparison tests highlighted that the network structure at T2 and T3 differed from that at T1.The global strength of the depressive symptoms network at T2 and T3 was stronger than the network at T1, suggesting a strengthening connectivity among symptoms over time.Furthermore, cross-lagged panel network analysis also showed that self-hatred was the overall essential influential symptom, which could give rise to other depressive symptoms and, conversely, be exacerbated by other depressive symptoms over time.The study also observed temporal shifts in symptom centrality.Specifically, loneliness displayed the highest out-expected influence centrality on the T1→T2 network, with strong association with T2 self-hatred and T2 friendlessness.Negative body image had the highest out-expected influence centrality on the T2→T3 network, with strong association with T3 self-hatred.Moreover, irritability consistently presented the highest in-expected influence centrality across both T1→T2 and T2→T3 networks, marking it as a prominent outcome within the depression network.
The current study enhances the knowledge of children’s depression symptomatology through the longitudinal network analysis.By combining regularized partial correlation network analysis and cross-lagged panel network analysis, the findings corroborate that self-hatred and irritability consistently emerge as core symptoms at all time points, while other highly central symptom vary across time points.Consequently, it is imperative to prioritize the prevention and intervention of children’s depression by focusing on central symptoms, namely self-hatred and irritability.Meanwhile, time-specific strategies targeting the central symptoms could prove instrumental in preventing the onset and escalation of depression in children.

Key words: depressive symptoms, late childhood, regularized partial correlation network, cross-lagged panel network analysis, symptomatology