Doctoral Dissertation Research in Political Science: A Computational Model of Political Attitude and Behavior

一个更全面的人类心理过程模型是可能的，也是必要的，以提高我们对普通公民政治判断和行为的理解。在这里，我开发了一个心理过程的计算模型，命名为John Q.Public，它融合了认知和情感机制，并整合了在线加工和基于记忆的模型。使用该模型，我将同时再现和解释公众舆论和投票行为中的众所周知的经验现象，这些现象只被单独和部分地从经验上解释。也就是说，我将展示众所周知的经验规律--如认知和态度启动效应、双过程、调查研究中的问题顺序和措辞效应以及在线加工--是由一套单一的、原则性的认知/情感机制产生的。研究还表明，政治判断的两种主要竞争性理论，即在线加工和基于记忆的模型，实际上是相互依存的，应该整合到一个更一般的理论中。这项研究还引入了一个关于战略行为的“完全”认知模型。具体地说，我把多个John Q.公众作为重复的社会困境博弈中的参与者，并试图以一种在认知上可行的方式解释在实验数据中观察到的合作动态。更广泛的影响：有助于扩大对普通公民政治判断和投票行为的理解。

A more comprehensive model of human mental process is possible and necessary to improve our understanding of ordinary citizens' political judgment and behavior. Here, I develop a computational model of mental process, named as John Q. Public, that incorporates both cognitive and affective mechanism, and integrates on-line processing and memory-based model. Using the model, I will simultaneously reproduce and explain the well-known empirical phenomena in public opinion and voting behaviors that have been only separately and partially accounted for empirically. That is, I will show that the well-known empirical regularities - such as cognitive and attitude priming effect, dual-process, question order and wording effects in survey research, and on-line processing - are generated by a single, principled set of cognitive/affective mechanisms. By doing so, the study also shows that the two major competitive theories of political judgments, on-line processing and memory-based model, are in fact interdependent and should be integrated into a more general theory. The study also introduces a 'fully' cognitive model of strategic behavior. Specifically, I put multiple John Q. Publics as players in a repeated social dilemma game and try to explain, in a 'cognitively feasible' way, the dynamics of cooperation observed in experimental data. Broader Impacts: Can help broaden understanding of ordinary citizens' political judgment and voting behavior.