Redundant Coding Can Improve Segmentation in Multiclass Displays
Data visualizations allow viewers to compare one dataset to
another. The visual marks that represent these datasets, or classes,
are visually distinguished from one another by salient visual
feature differences, such as color or shape. A designer of a graph
or map might encode one class of marks as either red, or circular,
and another class as either green, or square. One common
technique is to combine these cues in a redundant fashion,
encoding one class as red and circular, and the other as green and
square, under the assumption that a larger difference (via multiple
differing features) should help. Despite the ubiquity of this
practice, we know of no empirical demonstration that reveals
evidence of a potential benefit. Across two experiments, we
demonstrate that redundant coding can improve visual
segmentation of a simulated dataset in a crowded display
(Experiment 1) and that redundant coding also leads to stronger
visual grouping of elements (Experiment 2).
Images and movies
BibTex references
@InProceedings{NGF15, author = "Nothelfer, Christine and Gleicher, Michael and Franconeri, Steve", title = "Redundant Coding Can Improve Segmentation in Multiclass Displays", booktitle = "IEEE Visualization Poster Proceedings", month = "Oct", year = "2015", url = "http://graphics.cs.wisc.edu/Papers/2015/NGF15" }