By Prof. Dr. Adrian Schwaninger
X-ray screening of passenger bags is an essential component of airport security. X-ray screeners have to decide within seconds whether an X-ray image of a passenger bag is harmless or whether it might contain a prohibited item and has, therefore, to be sent to secondary search. Several factors influence detection performance of X-ray screeners. Figure 1 shows a slightly adapted version of a model published ten years ago . On the left, determinants of detection performance are depicted: selection tests, computer-based training (CBT), threat image projection (TIP) and practical tests. Supervision and quality control are important to ensure that these processes and tools are implemented in a way that enables screeners to achieve and maintain good detection performance results. I have pleasure in reviewing the different aspects of this model from a scientific and practical point of view and adding my opinions on where we stand in Europe regarding the quality of implementation of these processes and tools.
Selection Tests are Important but Often Underestimated
X-ray screening is a demanding task in which prohibited items have to be detected even if they are depicted from an unusual viewpoint, superimposed by other objects or placed in cluttered bags (Figure 2). Research has shown that there are large differences between people regarding visual-cognitive abilities that are needed to cope with such image-based factors (e.g. ). Using reliable, valid and standardised selection tests is essential to ensure that only people who have the visual-cognitive abilities and aptitudes needed in X-ray screening are employed as screeners [3, 4]. Supervision and quality control on this matter varies substantially across different European countries. In some of them, a scientifically validated object recognition test is mandatory in addition to other requirements for pre-employment assessment. In many other countries, this is not the case and requirements for employment as a screener are much lower, which provides a first explanation for the variation in detection performance across European airports.
Computer-based Training is Essential but Not Always Effective
In order to learn which items are prohibited, what they look like in X-ray images, and to keep up to date with new and emerging threats, computer-based training (CBT) can be a very effective and efficient tool [5, 6, 7]. However, not every CBT is effective. For example, in a study conducted by Koller, Hardmeier, Michel, and Schwaninger , training with an individually adaptive CBT containing a large library of prohibited items (X-Ray Tutor, XRT) was compared to two other CBT systems that were not adaptive and had smaller image libraries. Detection performance was measured with an X-ray competency assessment test  before and after three and six months of training (about 20 minutes per week). While large increases in detection performance were observed for the XRT training group, there were only small training effects for the other two CBT systems. Figure 3 illustrates the results broken up by prohibited items category for the comparison between XRT and one other CBT.
According to EU regulation, for classroom and/or CBT, at least six hours every six months are mandatory for X-ray screeners. While most, if not all, airports in Europe are compliant with this requirement, there are differences regarding its implementation.