跨期决策中的时间偏好：实验室测量与现实行为预测的综述

Time Preferences in Intertemporal Decision-Making: A Review of Laboratory Measures and Real-World Behavioral Prediction

摘要: 跨期决策研究通常通过折现函数刻画个体的时间偏好，并据此解释其行为差异。然而，既有研究显示，实验室中测得的跨期偏好在现实行为中的预测作用并不稳定。文章围绕这一问题，对折现函数的发展、实验测量范式及相关实证研究进行了梳理。总体来看，折现模型能够较好描述实验情境中的选择行为，但其与现实行为之间的关联普遍较弱，且在不同研究情境下表现出明显差异。进一步分析表明，这种差异不仅与时间偏好本身的多维性质有关，也受到实验任务中时间表征方式以及研究设计的影响。相较于单一折现参数，个体在真实情境中的跨期行为更可能由多种心理过程共同决定。基于此，文章认为，未来研究需要更加关注时间体验与情境因素在决策中的作用，并在研究设计上加强对现实行为的预测导向，以推动跨期决策研究从实验解释走向现实应用。

Abstract: Intertemporal decision-making research typically characterizes individual time preferences using discount functions and relies on these measures to explain behavioral differences. However, existing evidence suggests that time preferences measured in laboratory settings do not consistently predict real-world behavior. Addressing this issue, the present review synthesizes the development of discounting models, experimental paradigms, and empirical findings on real-world outcomes. Overall, while discounting models effectively describe choice behavior in controlled experimental contexts, their associations with real-world behavior are generally weak and vary substantially across settings. Further analysis indicates that this variability reflects not only the multidimensional nature of time preferences, but also differences in how time is represented in experimental tasks and how behavior is measured in empirical studies. Rather than being fully captured by a single discounting parameter, intertemporal behavior in naturalistic contexts is likely shaped by multiple underlying processes. Based on these observations, future research should place greater emphasis on temporal experience and contextual factors in decision-making, and adopt research designs that prioritize predictive validity, thereby facilitating the translation of laboratory findings to real-world applications.