Extracting Unknown Crime Information by Psychophysiological Responses

本研究的目的是为了扩大隐藏信息测试作为知识检测工具的应用范围。日本警方使用CIT有两个目的。首先，通过证明嫌疑人对犯罪具有事实性知识来识别肇事者，其次，从嫌疑人那里引出新的信息来帮助调查人员，也称为搜索CIT （sCIT）。原则上，考生会被提出一个问题，然后给出多个合理的答案，同时记录生理指标。测试遵循一个简单的逻辑，如果你不知道犯罪，所有的答案都会产生相同的反应，但如果你知道真相，正确的答案会引起比其他答案更高的反应。在这个项目中，我的重点是扩展工商及科技局。具体来说，我打算检验刺激相似性在答案选择中的影响。我是通过检查与正确答案相似的答案是否会引起与其相似程度相关的更大反应来做到这一点的。此外，我还研究了传统的不同答案选择集是否可以应用于连续的刺激，如路线和地图。该项目的成果直接适用于日本目前使用的CIT实践。研究结果有助资讯科技从业员了解如何选择不同的答案，以最大限度地提高资讯科技的侦测准确度，亦有助从业员在传统的资讯科技难以应付的情况下使用资讯科技，例如利用城市地图找出失窃物品的位置。

The purpose of this research project is to broaden the application of the Concealed Information Test (CIT) as a knowledge detection tool. The CIT is used by the Japanese police for two purposes. First, to identify perpetrators by demonstrating that a suspect has factual knowledge about a crime, and second, to elicit novel information from a suspect to aid the investigators, also known as searching CIT (sCIT). In principle, the examinee is presented with a question followed by multiple plausible answers while physiological indices are recorded. The test follows a simple logic, if you have no knowledge of the crime all answers produce the same response, but if you know the truth, the correct answer will elicit an elevated response compared to the rest.In this project I focus on expanding the sCIT. Specifically, I intent to examine the effects of stimulus similarity among the answer alternatives. I do so by examining if answers similar to the correct answer elicit larger responses respective to their degree of similarity. Furthermore, I examine if the traditional set of distinct answer alternatives can be applied to continuous stimuli, such as routes and maps.The results of this project are directly applicable to the CIT practices used in Japan today. The findings could help CIT practitioners to better understand how answer alternatives should be selected to maximize the detection accuracy of the CIT and they may allow practitioners to use the sCIT in situations that are difficult for the traditional sCIT, such as finding the location of a stolen item with a map of the city.