Exploring the relative importance of number of edge crossings and size of crossing angles: A quantitative perspective

Recent research has indicated that human graph reading performance can be affected by the size of crossing angles. The aesthetic of crossing angles is closely related to another aesthetic factor: edge crossings. Although the number of edge crossings has been previously identified as the most important aesthetic, its relative impact on human graph reading, compared to the size of crossing angles, has not been investigated. In this paper, we present an exploratory user study investigating the relative importance between crossing number and crossing angle. This study also aims to further examine the effects of crossing number and crossing angle not only on task performance measured as response time and accuracy, but also on cognitive load and visualization efficiency. The experimental results reinforce the previous findings that the two aesthetics each significantly affect performance of human graph reading. Further, in terms of the relative importance, the study demonstrates that given the current setting of the user study, the number of edge crossings is relatively more important than the size of crossing angles. To be more specific, crossing number and crossing angle together explain about 40% of the variance in response time, mental effort and visualization efficiency, with about 83% of the explained variance being attributed to crossing number. In regard to response accuracy, crossing number and crossing angle together explain about 14% of the variance, with a slightly larger portion of the explained variance being attributed to crossing number.