摘要: |
This report includes a review of the results of experiments conducted by researchers at the University of Central Florida under a TSA research grant. This research supports the Transportation Security Laboratory's Human Factors Long-Term Research program. The purpose of this research is to develop a theoretical understanding of the cognitive processes involved in the various security tasks. The research described in this report is most relevant to the X-ray screener search task. The first experiment examined the validity of training stimuli devised from X-ray images generated at the William J. Hughes Technical Center. A majority of these stimuli were modified using Adobe Photoshop post-processing techniques to devise test images consisting of overlapping and non-overlapping targets and distracters. The results showed significant differences across the test item categories which validated the stimuli as created by the UCF researchers. The second and third experiments examined the impact of clutter and target-distracter similarity in the training environment. Results indicated incremental improvements in detection accuracy when training manipulations are varied. Specifically, it appears that training and exposure to stimuli improves trainees' ability to correctly clear bags that do not contain threats or suspicious items. A group of trainees that was trained for only 20 minutes with difficult distracters in cluttered bags did not reduce their false-positive rate by almost 50 percent, when compared to untrained study participants. In addition, the groups trained with distracters that were highly similar to the targets produced the greatest performance improvements. These results, as well as this area of research, have important implications for the improvement of X-ray screener training. |