Visualization of Uncertainty in Context Aware Mobile Applications

Rukzio, E., Hamard, J., Noda, C., and De Luca, A. 2006. Visualization of uncertainty in context aware mobile applications. In Proceedings of the 8th Conference on Human-Computer interaction with Mobile Devices and Services (Helsinki, Finland, September 12 – 15, 2006). MobileHCI ’06, vol. 159. ACM Press, New York, NY, 247-250.

Unlike previous research that argue that a context-aware system usability can be improved by displaying the uncertainty to the user, this study shows that the user needs slightly more time and produces slightly more errors when the confidence of the system is visualized.

The user study evaluating the effect of the visualization of uncertainty consisted of three phases which were conducted by every participant: preliminary interview, the test with the prototype and a post interview. The test consisted of mobile web forms pre-filled with the user’s personal information. The input forms were mapped with colors with similar meanings as the traffic lights. As a result many users mentioned that they were distracted by the colors. Most participants stated that they did not take the visualized probabilities into account. Therefore, the authors conclude that in general the visualization of uncertainty in context-aware system is still questionable.

Slides of Enrico’s presentation at MobileHCI 2006.

Relation to my thesis: This study is very close to a type of user study on showing/not showing uncertainty I had in mind. However, I would probably apply it into a stronger contextual setting implying location awareness in a real-world environment. I would also question the way that the confidence is visualized. Moreover, the type of the task might have an impact on the usefulness of uncertainty visualization and the usability of the system. The authors of this study are working on the situations in which the confidence should be visualized or not in order to develop guidelines defining when the uncertainty should be shown and how it should look like. Can’t wait to see the results…