A Comparison of Instructor-Led vs. Web-based Training for Detecting Deception

Janna M. Crews, Jinwei Cao, Ming Lin, Jay F. Nunamaker, Judee K. Burgoon


Research has long recognized that many biases and shortcomings of humans severely limit their ability to accurately detect deception, and this may lead to great risks in government or military operations. One possible method to improve humans? deception detection ability is to train them to recognize cues of deception. To do this, we need to create effective training curricula and educational tools. This paper focuses on describing how we used existing research on deception detection to guide the design, development and evaluation of such a training curriculum. Research- authenticated cues of deception were selected, organized and presented as the core
of the training curriculum. Real-life examples and their analyses were created to illustrate the cues and provide learners with immediate feedback in detection practice.
Besides traditional instructor-led, lecture-based training, we also implemented this curriculum with a Web-based, learner-centered multimedia training system called
Agent99 Trainer. Experiments were conducted to study the effectiveness of our training curriculum and to compare the two training implementations. The initial results showed that our training curriculum significantly improved the accuracy of human deception
detection and the Agent99 Trainer system provided training as effective as the instructor-led, lecturebased training.

Full Text:


JSTEM. ISSN: 1557-5284