Abstract: With the increasing use of internet devices, the prevalence of problematic internet use (PIU) has steadily increased. PIU is a clinical psychological disorder resulting from excessive internet use, with global prevalence rates ranging from 12.6% to 67.5%. PIU increases the risk of academic difficulties, negative emotions, sleep disturbances, and aggressive behaviors. It is also associated with a disregard for long-term future considerations. Future thinking, a psychological process that involves planning and predicting future scenarios, is closely linked to the development of PIU. It encompasses psychological experiences of future scenarios, self-perception, and cognitive processing of future values and emotions. A deficiency in future thinking may lead individuals to overlook the long-term consequences of internet use, thereby increasing the risk of excessive use. The future-oriented cognition model suggests that clear future expectations lead to activities like simulation, prediction, and planning, which can enhance long-term consideration and self-control. Thus, having clear expectations about future events can enhance individuals'consideration of long-term impacts, making them more likely to reflect on the potential academic, social, and health-related risks of PIU, and promoting goal-directed behavior and self-control to mitigate current excessive internet use. However, current research on the relationship between future thinking and PIU remains limited, with most existing studies focusing on the association between individual variables and PIU. The development of PIU involves complex interactions among multiple aspects of future thinking. Therefore, understanding the interplay between different components of future thinking and PIU is crucial for developing effective interventions to mitigate PIU.
This study analyzed data from 2,685 undergraduate students across eight universities in China. The cognitive components of future thinking (i.e., episodic future thinking, future self-continuity, and delay discounting), the emotional component (i.e., anticipatory pleasure), and the individual difference component (i.e., time perspective) were measured, along with the degree of PIU. Network analysis was employed to examine the complex relationships between these variables, and simulation intervention algorithms were used to identify potential intervention targets. Spearman correlations were first calculated to generate a heat map of the correlation matrix. Network analysis was then performed with 18 factors, including PIU sub-dimensions, as nodes, using a regularized partial correlation glasso network model. Finally, an Ising network model was constructed to simulate the effects of node changes on PIU and to identify key intervention targets. The NodeIdentifyR algorithm simulated 5,000 participants, applying two types of interventions to each node in the network: (1) Alleviating intervention, which reduced the node’s original parameter values by two standard deviations to simulate improved performance, and (2) Aggravating intervention, which increased the node’s original parameter values by two standard deviations to simulate worsened performance. The “sum score” indicator was used to assess the overall effect of each intervention on the three sub-symptoms of PIU.
The results indicated that (1) Multiple dimensions of future thinking (i.e., episodic future thinking, future self-continuity, delay discounting, and anticipatory pleasure) and future positivity were significantly negatively correlated with PIU and acted as protective factors. (2) In the network model, time perspective was identified as a distal factor for PIU, with future thinking serving as a mediating factor between time perspective and PIU. (3) Clarity of anticipation was found to be the most effective intervention target for preventing the worsening of PIU symptoms, while vividness of the future self was identified as the best intervention target for improving PIU symptoms.
These findings highlight the importance of understanding the complex interactions that contribute to PIU. By focusing on the most influential ones, as identified by simulation algorithms, interventions can be more targeted and effective. Addressing key factors such as the clarity of future expectations and vividness of future self-image can help individuals better understand the long-term effects of internet use, leading to improved self-control and reduced impulsive behaviors. Training people to think more clearly about future scenarios and strengthening their connection to their future selves can help reduce PIU symptoms and prevent excessive internet use in the long term.

Key words: problematic internet use, future thinking, time perspective, network analysis methods, simulation intervention

摘要： 缺乏未来的远见与思考是问题性网络使用（problematic internet use, PIU）发生发展的重要原因之一。而未来思维与PIU的相互作用尚不清楚。研究使用网络分析方法探讨未来思维的认知（未来情景思维、未来自我连续性和延迟折扣）、情感（预期愉悦）和个体差异成分（时间洞察力）与PIU的相互作用,并采用模拟算法识别PIU的最佳干预靶点。通过对2685名被试调查分析后发现：（1）未来思维的认知和情感成分以及未来积极洞察力与PIU呈显著负相关。（2）时间洞察力是PIU的远端因素,而未来思维的认知与情感成分在两者之间起中介作用。（3）预期清晰性是预防PIU恶化的最佳干预靶点,而未来自我生动性是促使PIU改善的最佳干预靶点。研究结果对于理解PIU的心理机制和制定干预策略具有重要意义。

关键词: 问题性网络使用, 未来思维, 时间洞察力, 网络分析方法, 模拟干预