View Single Post
Old 12-01-2015, 06:56 PM
campbell's Avatar
campbell campbell is offline
Mary Pat Campbell
Join Date: Nov 2003
Location: NY
Studying for duolingo and coursera
Favorite beer: Murphy's Irish Stout
Posts: 91,001
Blog Entries: 6

Research in the field of information visualization is usually mediocre, often severely flawed, and only occasionally well done. In this article, I’ll critique a recently published research paper to illustrate some of the flaws that are common and to propose what we can do to improve the work. This is a paper by Michelle A. Borkin, Zoya Bylinskii, Nam Wook Kim, Constance May Bainbridge, Chelsea S. Yeh, Daniel Borkin, Hanspeter Pfister, and Aude Oliva, titled “Beyond Memorability: Visualization Recognition and Recall.” I’ve chosen this particular paper, not because it is exceptionally flawed, but because it has received a great deal of attention and the findings that it erroneously claims are potentially harmful to data visualization. This concerns me, and should concern us all.

Visualizations don’t need to be designed for memorability— they need to be designed for comprehension. For most visualizations, the comprehension that they provide need only last until the decision that it informs is made. Usually, that is only a matter of seconds. When the comprehension has lasting value, it should be stored in memory, not the visualization. It is true that a visualization can be designed to serve as a mnemonic mechanism to encode a particular message in a memorable way. Borkin’s study, however, merely addresses characteristics that make visualizations memorable, not whether or how those characteristics can be used to encode comprehension.

We’ve known for quite a while that particular characteristics of data visualizations tend to catch our attention and make them sticky. For example, novel images and those that trigger strong emotions tend to stick in memory, but rarely, if ever, in a way that supports comprehension. If I incorporate an image of a kitten into a data visualization, I can guarantee that a test subject would remember seeing that kitten if it is shown to her again a few minutes later. But how is that useful? Unless the visualization’s message is that kittens are cute and fun, nothing of consequence has been achieved.

4. The information visualization community is complacent.
Why is it that few people besides me are critiquing flawed information visualization research? I’m certainly not unique in my understanding of science. Several of my friends and colleagues do excellent research in the field. They’re aware of the same flaws I am, so why aren’t they speaking up? I suspect that for many, it is fear of recrimination or distaste for conflict that keeps them silent. I loathe recrimination and conflict as much as anyone. But I want information visualization to contribute as much as it can to the world, so I work to improve it. My lone voice, however, cannot turn the tide. Others who are respected in the field must speak up as well. Imagine what a difference we could make if we raised our voices in a chorus of guidance to elevate the work of our field. Oh, how I long for compatriots.
I hear ya, I hear ya.

LinkedIn Profile
Reply With Quote
Page generated in 0.08837 seconds with 9 queries