Wellformedness Properties in Euler Diagrams: An Eye Tracking Study for Visualisation Evaluation
This incremental research provides insights for software developers on optimizing Euler diagram visualizations in fields like engineering and medicine.
The study investigated which of two wellformedness properties in Euler diagrams—duplicated curve labels or disconnected zones—most affects user comprehension, finding that duplicated curve labels slow down task performance and increase eye movements, though accuracy was not significantly impacted.
In the field of information visualisation, Euler diagrams are an important tool used in various application areas such as engineering, medicine and social analysis. To effectively use Euler diagrams, some of the wellformedness properties needs to be avoided, as they are considered to reduce user comprehension. From the previous empirical studies, we know some properties are swappable but there is no clear justification which property would be the best to use. In this paper, we considered two main wellformedness properties (duplicated curve labels and disconnected zones) to test which among the two affect user comprehension the most, based on the task performance (accuracy and response time), preference and eye movements of the users. Twelve participants performed three different types of tasks with nine diagrams of each property (so, in total eighteen diagrams) and the results showed that duplicated curve labels property slows down and trigger extra eye movements, causing delays for the tasks. Though there is no significant difference in the accuracy but the insights obtained from the response time, preference and eye movements will be useful for software developers on the optimal way to visualise Euler diagrams in real world application areas.