青少年手机成瘾的潜在类别及其干预靶点研究

Latent Classes of Problematic Mobile Phone Use in Adolescents and Targets for Intervention

摘要: 目的：分析青少年手机成瘾的潜在类别特征；用伊辛模型构建不同风险组的心理网络，明确手机成瘾与学业倦怠、心理韧性、自我控制等因素的复杂互动关系，找出关键干预靶点。方法：选取华北某城市1661名中学生，采用智能手机成瘾量表简版、青少年学习倦怠量表、心理韧性量表、自我控制量表开展调查。用Mplus做潜在剖面分析，划分成瘾风险类别；用R语言构建伊辛模型网络，分析节点中心性；通过NodeIdentifyR完成计算机模拟干预分析。结果：潜在剖面分析分出三类人群：低风险组(41.1%)、中等风险组(43.8%)、高风险组(15.2%)。自我控制越强，越可能属于低风险组；学业倦怠是高风险组的重要预测因素。网络分析显示，高风险组的网络连接最紧密；学业倦怠节点充当关键桥梁，连接自我控制与手机成瘾网络。中心性分析发现，“戒断症状”(SPA-2)和“健康习惯”(SC-2)是网络核心节点。模拟干预表明，干预“学业倦怠”(LB-2)与“健康习惯”(SC-2)，比直接干预成瘾症状更能降低整体网络风险。结论：青少年手机成瘾存在明显的类别差异。高风险群体的心理因素间有复杂的非线性互动。干预不能只控单一症状，而要做系统网络调节；优先改善学习倦怠、培养健康习惯，效果更好。

Abstract: Objective: This study aimed to identify the latent classes of mobile phone addiction (MPA) among adolescents and to construct psychological networks for different risk groups using the Ising model. The study sought to elucidate the complex interactions between MPA, academic burnout, resilience, and self-control, and to identify key targets for intervention. Methods: A total of 1661 middle school students from a city in Northern China participated in this study. Participants completed the Smartphone Addiction Scale Short Version (SAS-SV), the Adolescent Learning Burnout Scale, the Resilience Scale, and the Self-Control Scale. Latent Profile Analysis (LPA) was conducted using Mplus to categorize addiction risk levels. The Ising model was estimated using R to construct networks and analyze node centrality. Computer simulation intervention analyses were performed using the NodeIdentifyR package. Results: LPA identified three distinct classes: a low-risk group (41.1%), a moderate-risk group (43.8%), and a high-risk group (15.2%). Higher levels of self-control were associated with a higher likelihood of belonging to the low-risk group, while academic burnout emerged as a significant predictor for the high-risk group. Network analysis revealed that the high-risk group exhibited the densest network connectivity. Notably, the academic burnout node acted as a critical bridge connecting self-control with the mobile phone addiction network. Centrality analysis identified “Withdrawal Symptoms” (SPA-2) and “Healthy Habits” (SC-2) as the core nodes within the network. Simulation interventions indicated that targeting “Academic Burnout” (LB-2) and “Healthy Habits” (SC-2) was more effective in reducing overall network risk than directly intervening on addiction symptoms. Conclusion: Significant heterogeneity exists in the patterns of mobile phone addiction among adolescents. The psychological factors within the high-risk group are characterized by complex non-linear interactions. Intervention strategies should not merely focus on controlling isolated symptoms but rather on systematic network regulation. Prioritizing the alleviation of learning burnout and the cultivation of healthy habits yields superior intervention outcomes.