Defining the Questions: A Research Agenda for Nontraditional Authentication in Arms Control

Many traditional authentication techniques have been based on hardware solutions. Thus authentication of measurement system hardware has been considered in terms of physical inspection and destructive analysis. Software authentication has implied hash function analysis or authentication tools such as Rose. Continuity of knowledge is maintained through TIDs and cameras. Although there is ongoing progress improving all of these authentication methods, there has been little discussion of the impact of human factors on authentication. These issues include sleight-of-hand substitutions, monitor perception vs. reality, and visual diversions. Since monitor confidence in a measurement system depends on the product of their confidences in each authentication element, it is important to investigate the impact of human factors. This paper will present an initial effort to identify the impact of human factors on traditional authentication approaches in safeguards that are especially relevant to arms control verification. This will include a survey of the literature and direct engagement with non-traditional experts in areas like psychology and human factors. Based on the identification of problem areas, potential research areas will be identified and a possible research agenda will be developed.