A Dynamic Dual Process Model of Risky Decision Making
A Dynamic Dual Process Model of Risky Decision Making
复制标题
DOI：
10.1037/rev0000087
复制
发表时间：
2018-03-01
影响因子：
5.4
通讯作者：
Trueblood, Jennifer S.
中科院分区：
文献类型：
--
作者：
Diederich, Adele;Trueblood, Jennifer S.
判断和决策中的许多现象往往归因于两个推理系统的相互作用。虽然这些所谓的双重过程理论可以解释许多类型的行为，但它们很少被正式化为数学或计算模型。相反，双过程模型是典型的口头理论，难以最终评估或测试。在正式（即，数学）双过程模型已经被提出，但是它们没有定量地适合于实验数据，并且当涉及到两个系统的定时时，它们通常是沉默的。在目前的文章中，我们提出了一个动态的双过程模型框架的风险决策，提供了一个帐户的时间和相互作用的2个系统，可以解释选择和响应时间的数据。我们概述了模型的几个预测，包括两个系统的时间变化以及时间压力如何影响行为。该框架还使我们能够探索不同的假设偏好是如何构建的两个系统，以及动态的相互作用的两个系统。特别是，我们检查3种不同的可能的功能形式的2个系统和2种可能的方式，系统可以相互作用（同时或串行）。我们使用来自Guo，Trueblood和Diederich（2017）的风险决策数据将这些双过程模型与2个单过程模型进行比较。使用这些数据，我们发现其中一个双过程模型在考虑选择和响应时间方面显着优于其他模型。
Many phenomena in judgment and decision making are often attributed to the interaction of 2 systems of reasoning. Although these so-called dual process theories can explain many types of behavior, they are rarely formalized as mathematical or computational models. Rather, dual process models are typically verbal theories, which are difficult to conclusively evaluate or test. In the cases in which formal (i.e., mathematical) dual process models have been proposed, they have not been quantitatively fit to experimental data and are often silent when it comes to the timing of the 2 systems. In the current article, we present a dynamic dual process model framework of risky decision making that provides an account of the timing and interaction of the 2 systems and can explain both choice and response-time data. We outline several predictions of the model, including how changes in the timing of the 2 systems as well as time pressure can influence behavior. The framework also allows us to explore different assumptions about how preferences are constructed by the 2 systems as well as the dynamic interaction of the 2 systems. In particular, we examine 3 different possible functional forms of the 2 systems and 2 possible ways the systems can interact (simultaneously or serially). We compare these dual process models with 2 single process models using risky decision making data from Guo, Trueblood, and Diederich (2017). Using this data, we find that 1 of the dual process models significantly outperforms the other models in accounting for both choices and response times.