DescriptionData visualizations are prevalent in scientific simulation, medicine, physical therapy, product design, manufacturing, weather predictions, simulations, and engineering. They are also used in information and scientific visualization for WWW exploration, document searching, and education. This pervasiveness makes it important to understand how we comprehend these visualizations, what comprehension difficulties can be expected in diverse populations, and which visual properties increase comprehension difficulty. This will enable us to create training methods that help people gain basic comprehension strategies and analytic skills for visualizations.
The goal of this dissertation is to define an approach that can be used to determine the factors that make a visualization difficult to comprehend by certain individuals. This information is then used to test if training using this information helps individuals develop new and workable strategies for visualization analysis.
In this dissertation we present an approach that is designed as a series of steps designed (1) to determine what cognitive abilities are correlated with comprehension of the visualization, (2) to identify visual properties that make a basic visualization difficult to comprehend and (3) to measure the effect of basic incremental training using these visual properties.