Comparing efficiency and cognitive load of uncertain data visualizations

The current information age is associated with increasingly growing amounts of data. Information systems are built to retrieve relevant information or to extract new information, for consumers as well as in an operational and scientific context. Here important decisions are frequently grounded on information systems results. Unfortunately, the data used are not always accurate and systems do not always work precisely. This paper includes an empirical user study about the effect of different uncertain data visualizations on users' efficiency (time, error rate) and the cognitive load. A table, a static graphic and an interactive graphic containing uncertain data were compared. The results of our online study (N=146) showed that there is a significant difference in task time between visualizations, while there are no significant differences in error rate and cognitive load. Based on this, we recommend a static graphic visualization for the interface of the Fish-4Knowledge system.