Skip to Main Content
Pervasive computing research is mainly focused on technology as such, while more human-centric issues are being addressed only recently. One such issue is trust and its management is getting crucial for further and wider acceptance of pervasive computing solutions. In IT domain trust was first addressed some fifteen years ago, but the suggested approaches were actually addressing security. After a few years, more advanced and rational agents assuming methodologies emerged that were based on Bayesian statistics. These were followed by Dempster-Shafer theory of evidence and its derivatives, e.g. subjective algebra, together with some game theoretic attempts. However, trust is a manifestation of (frequently non-rational) reasoning and assessment processes; it has to be treated in line with this fact and adequately supported in interactive computing environments. This requires a commonly accepted and agreed methodology for trust research to enable scientifically sound evaluations of proposed methodologies, which is the first motivation behind this paper - in this area we are lacking experimental verification of developed trust management technologies to check how well they meet users' mental models and expectations. On this basis a cognitive ergonomics focused methodology, called qualitative assessment dynamics (QAD), has been developed, being the second motivation behind this paper. QAD complements existing trust management methodologies.