Comparing two interface tools in performing visual analytics tasks

In visual analytics, menu systems are commonly adopted as supporting tools because of the complex nature of data. However, it is still unknown how much the interaction implicit to the interface impacts the performance of visual analysis. To show the effectiveness of two interface tools, one a floating text-based menu (Floating Menu) and the other a more interactive iconic tool (Interactive-Icon), we evaluated the use and human performance of both tools within one highly interactive visual analytics system. We asked participants to answer similarly constructed, straightforward questions in a genomic visualization, first with one tool, and then the other. During task performance we tracked completion times, task errors, and captured coarse-grained interactive behaviors. Based on the participants accuracy, speed, behaviors and post-task qualitative feedback, we observed that although the Interactive-Icon tool supports continuous interactions, task-oriented user evaluation did not find a significant difference between the two tools because there is a familiarity effect on the performance of solving the task questions with using Floating-Menu interface tool.