Collaborative Research on Behavioral Models of Intertemporal Choice

尽管理性选择模型在广泛的人类活动中产生了一系列见解，但心理学研究表明，它在一些系统和重要的方面是不准确的。研究人员建议继续他们的议程，将心理学上关于人类行为的更现实的假设整合到跨期选择的正式经济模型中。大多数研究人员过去和未来的研究都围绕着一个既定的现象：尽管标准经济模型假设偏好是时间一致的，但有证据表明，人们有现在偏向的偏好，即一个人相对于较早的日期而不是较晚的日期的相对偏好随着较早日期的临近而变得更强。但是，除了研究当前偏好本身的含义之外，研究人员还对一个人是否意识到未来的自我控制问题的重要性特别感兴趣。他们的研究重点是假设一个人是成熟的——完全意识到未来的自我控制问题——或幼稚的——完全没有意识到未来的自我控制问题的不同含义。研究人员有两个正在进行的项目，他们计划在拨款期间完成。第一项研究考察了当下偏好对成瘾产品消费的影响。有自我控制问题的人倾向于过度消费成瘾产品，但这种行为受到对自我控制问题的复杂意识的影响。因为世故使人们对自己抵抗未来诱惑的能力感到悲观，因此觉得他们现在还不如上瘾，所以世故会加剧过度消费。即便如此，我们表明，在现实情况下，这种悲观效应可能会被一种激励效应所克服，在这种激励效应中，成熟使人们意识到当前的约束可能会促进未来的约束。在目前的工作中，研究人员正试图概括和扩展这一早期的分析，以探索诸如价格对消费的影响等问题。在第二个正在进行的项目中，研究者与卡内基梅隆大学的George Loewenstein合作，探索了人们在跨期选择领域中犯的第二个系统性错误的含义：人们倾向于低估改变他们未来偏好的因素的影响，因此将当前偏好投射到未来的自己身上。研究人员和他们的合著者开发了这种预测偏差的正式模型，并将该模型应用于几个说明性的经济应用。在继续的研究中，他们计划将他们的模型更深入地应用于成瘾和消费边际效用递减的错误预测等问题。研究人员还有几个新项目处于非常早期的阶段。研究人员计划在赠款期间进行这些项目的初步阶段。其中一篇探讨了长期项目的拖延问题。尽管大多数关于拖延症的研究都假设一个项目只需要一段时间的努力，但大多数现实世界的项目都需要一些时间来完成。这种区别对于有“现在偏向”偏好的人来说是至关重要的，因为一个人是否开始一个项目和是否完成它成了两个截然不同的问题。自我控制问题会导致人们开始但从不完成项目，并且，当面对多个正在进行的项目时，会选择错误的项目。在第二个项目中，研究人员计划在一个允许他们建立在现有研究中提出的一些观点的一般框架中分析当前偏见偏好。有一种简单的直觉，即成熟的影响如何取决于不同时刻的放纵对长期健康的影响。当沉迷于一个时期会让沉迷于另一个时期对一个人未来的幸福更有害时，对未来放纵的成熟会让一个人意识到当前放纵的高成本，从而减少现在的不当行为。当沉迷于一段时间使沉迷于另一段时间对一个人未来的幸福伤害更小时，成熟加剧了沉迷。初步分析确定了这种影响占主导地位的风格化环境，并深入了解了为什么它没有在一般情况下占主导地位。

Although the rational-choice model has yielded an array of insights across a broad range of human activities, research from psychology suggests that it is inaccurate in some systematic and important ways. The investigators propose to continue their agenda of integrating psychologically more realistic assump-tions about human behavior into formal economic models of intertemporal choice. Most of the investigators' past and proposed future research revolves around a single well-established phenomenon: Whereas the standard economic model assumes that preferences are time-consistent, evi-dence suggests people have present-biased preferences, wherein a person's relative preference for well-being at an earlier date over a later date gets stronger as the earlier date gets closer. But more than merely examining the implications of present-biased preferences per se, the investigators are particularly inter-ested in the importance of whether a person is aware of future self-control problems. Their research fo-cuses on the differing implications of assuming a person is sophisticated - fully aware of future self-control problems - or naive - fully unaware of future self-control problems. The investigators have two continuing projects that they plan to complete during the grant period. The first examines the implications of present-biased preferences for the consumption of addictive products. People with self-control problems tend to over-consume addictive products, but this behavior is affected by sophisticated awareness of self-control problems. Because sophistication makes people pessimistic about their ability to resist future temptations, and hence feel they might as well get addicted now, so-phistication can exacerbate over-consumption. Even so, we show that in realistic cases this pessimism effect is likely to be overcome by an incentive effect, wherein sophistication makes the person realize that current restraint may promote future restraint. In current work, the investigators are attempting to gener-alize and extend this earlier analysis, to explore such issues as the effects of price for consumption. In a second continuing project, joint with George Loewenstein of Carnegie-Mellon University, the investiga-tors explore the implications of a second systematic error that people make in the realm of intertemporal choice: People tend to underappreciate the influence of factors that change their future preferences, and hence to project current preferences on their future selves. The investigators and their co-author develop a formal model of such projection bias, and apply this model to several illustrative economic applications. In continuing research, they plan to apply their model in more depth to such things as addiction and mis-predictions of diminishing marginal utility of consumption. The investigators also have several new projects that are in the very early stages. The investigators plan to carry out the preliminary phases of these projects during the grant period. One explores procrasti-nation of long-term projects. Whereas most work on procrastination assumes that a project requires a sin-gle period of effort, most real-world projects take some duration to complete. This distinction is crucial for people with present-biased preferences, because whether a person starts a project and whether she finishes it become two distinct questions. Self-control problems can lead people to start but never finish projects, and, when faced with multiple ongoing projects, to choose the wrong projects to work on. In a second project, the investigators plan to analyze present-biased preferences in a general framework that allows them to build upon some of the points made in existing research. There is a simple intuition for how the effects of sophistication depend on how indulgence at different moments affect long-run well-being. When indulging in one period makes indulging in the other period more harmful to one's future well-being, sophistication about future indulgence leads a person to perceive high costs of current indul-gence and hence to misbehave less now. When indulging in one period makes indulging in the other pe-riod less harmful to one's future well-being, sophistication exacerbates indulgence. Preliminary analysis identifies a stylized setting in which this effect dominates, and yields insight on why it does not dominate in general.