The Visual Scalability of Integrated and Multiple Views for High Resolution Displays

Many real world datasets involve the integration of spatial and multidimensional data. Users can gain insight into these datasets using information visualizations. One aspect of the visualization design space for integrating these data types is a choice of when to use a single integrated view and when to use multiple views. Because of the many tradeoffs involved with this design decision it is not always clear which design to use. Additionally, as the cost of display technologies continues to decrease high resolution displays are increasingly being used for visualization. In the past, the scalability of different visualization designs has been limited by the size and corresponding resolution of the display technology. As the technological limitations lessen and more information can be displayed, consideration of human limitations becomes increasingly critical.

The purpose of this research is to compare the different information visualization designs for integrating spatial and multidimensional data in terms of their visual scalability for high resolution displays. Toward this goal the design space was articulated and, to establish a baseline, user performance with an integrated view and multiple views was compared using low resolution displays. The proposed work deals specifically with visualizing geospatially-referenced multidimensional time-series data on high resolution displays. We consider (analytically and empirically) the visual scalability of integrated and multiple views. We also consider the benefits of combining these designs and adding a separate view of the multidimensional data on high resolution displays. The end product of this work will be design guidelines for visualizing datasets with spatial and multidimensional information on high resolution displays based on empirical evaluation of user performance.