Computer-based training for deception detection: What users want?

Abstract

Training humans in detecting deception is as much a difficult and important problem as detecting deception itself. A computer-based deception detection training system, Agent99 Trainer, was built with a goal to train humans to understand deception and detect deception more accurately. Based on the previous studies, a newer version of this system was designed and implemented not only to overcome the limitations of the earlier system, but also to enhance it with additional useful features. In this paper, we present a usability study to test the design of this system from a users' perspective. The findings of this study, based on quantitative and qualitative data, demonstrate good usability of the training system, along with providing a better understanding of what users want from such a deception detection training system. © Springer-Verlag Berlin Heidelberg 2004.