Task complexity and user model attributes: An analysis of user model attributes for elderly drivers

Abstract

Modeling users in order to design appropriate interfaces and interactions or to simulate a specific user behavior is an ambitious task. When using user model attributes to design an interface as well as its interactions we focus tasks at different levels of complexity. In our work we address the appropriateness of physical, cognitive, behavioral, and psychological attributes and their relevancy for designing and describing tasks at such levels of complexity. We conducted a study that uses tasks of varying complexity levels that we relate to attributes in terms of the categorization previously described. A driving simulator together with a prototype of in-car controls that allows to perform primitive as well as complex tasks during a driving scenario represent the study context and the user interface for the participants who took part in three different scenarios, where they performed selected tasks that have been identified for the automotive area. Further additional workload tasks were used to induce stress and to investigate in the effect of cognitive, behavioral, and psychological attributes. First results show that the physical parameters address mainly primitive tasks. Regarding cognitive, behavioral and psychological parameters, tasks need to be addressed at a more complex level, which was supported by the results of the study. Concluding the relation of primitive tasks to cognitive, behavioral, and psychological attributes is not viable. © 2012 Springer-Verlag.