The Impact of Dynamically Changing Information on Decision Processes

在日常生活中，人们遇到了许多情况，在这些情况下，他们必须做出决定，可用信息随着时间的推移而变化。在繁忙的高速公路上改变车道的简单行为就是一个例子。虽然一开始，车道看起来很空旷，但下一刻，另一辆车可能已经从另一边开进来了。为了避免冲突，决策者必须能够分析和回应新的信息。过去，许多关于人类决策的研究都集中在信息恒定不变的情况下的决策。本项目采用实验和计算相结合的建模方法来研究动态信息对决策的影响。这项工作与日常生活相关，并为社会提供了潜在的好处。过去的研究表明，人们的决定往往对环境很敏感(例如，增加一个更糟糕的选项会让另一个选项看起来比以前更好)。广告商可以利用这种效应来推广对人们可能有益也可能有害的产品。这个项目调查信息的动态呈现如何影响上下文效应，并可能消除这些决策的偏差。该项目还通过促进数学心理学中的性别平等，支持增加数量社会科学的多样性。这个项目的研究目标是扩大我们对人们如何整合动态变化的信息的理解。讨论了三个关键问题：(1)人们适应新信息的速度有多快？(2)新信息是如何整合到决策过程中的？(3)决策者如何应对不同范式中变化的信息是否有共性？或者，新信息领域的适应和整合是特定的吗？为了解决这些关键问题，我们将研究一系列认知模型，这些模型编码了关于潜在决策过程的不同假设。将开发新的近似贝叶斯计算方法，将这些认知模型与数据相结合，并比较编码不同假设的模型。从经验上讲，该项目涉及一系列新的实验，这些实验将对了解人们在不断变化的信息下如何做出决定做出重大贡献。这些实验跨越了广泛的范式，从简单的知觉任务到涉及多种选择的高级优先选择任务。总而言之，它们将为不断变化的信息如何影响不同领域的决策提供有价值的见解。

In the course of everyday life, people encounter numerous situations in which they must make decisions where the available information changes over time. The simple act of changing lanes on a busy highway is an example. While at first, the lane looks clear, the next moment another car may have pulled in from the other side. In order to avoid a collision, the decision maker must be able to analyze and respond to the new information. In the past, much of the research on human decision-making has focused on decision-making when information is constant and unchanging. This project takes a joint experimental and computational modeling approach to study the influence of dynamic information on decisions. The work has relevance in day to day life and offers potential benefits for society. Past research has shown that people's decisions are often sensitive to context (e.g., adding a worse option can make another option look better than before). Advertisers can exploit such effects to promote products that may or may not be good for people. This project investigates how dynamic presentations of information can influence context effects and possibly de-bias these decisions. This project also supports increasing diversity in the quantitative social sciences through the promotion of gender equality in mathematical psychology. The research goal of this project is to expand our understanding of how people integrate dynamically changing information. Three key questions are addressed: (1) How quickly do people adapt to new information? (2) How is new information integrated into the decision process? (3) Are there commonalities in how decision makers respond to changed information across different paradigms? Or, is adaptation and integration of new information domain specific? To address these key questions, we will examine a range of cognitive models that encode different hypotheses about underlying decision processes. New approximate Bayesian computation methodologies will be developed to integrate these cognitive models with data and compare models encoding different hypotheses. Empirically, the project involves a series of new experiments that will contribute significantly to what is known about how people make decisions under changing information. These experiments span a wide range of paradigms from simple perceptual tasks to high-level preferential choice tasks involving multiple alternatives. Collectively, they will provide valuable insight into how changing information affects decisions across different domains.