We propose a constructive procedure for building a distribut-ed cognition model of a system out of contextual / ethnographic data. We then show how such a model can be conveniently used for studying, in a repeatable and justifiable way, if a system correctly implements selected user-centred design principles. Our approach thus complements user studies in that it enables reasoning about the situated use of a teamwork system even before direct user involvement. We have applied our procedure to a healthcare case study. In particular, we have re-analysed a well-known adverse incident that led to a fatality and for which a comprehensive investigation report is in the public domain. By reasoning about the distributed cognition model, we identified several issues that were not addressed in the incident report nor in other subsequent analyses.
10.1007/978-3-642-25364-5_10