Essay questions

  1. What do we know about the role of perception in visualization?
  2. Why is it important?

Discussion

Ehrenberg (1975) has discussed that the primary goal of a graph is not to convey numbers, as tables would do better in describing numbers. Graphs are meant to be read and analyzed by extracting quantitative and qualitative information in it, by organizing extracted information, it is possible to see patterns and structure that were not directly visible by using other technics of data analysis. A graph is accurate than another if it leads to human judgements that come closer to the actual encoded quantities(1). A human being’s nervous system has a process that converts low-level information like length, angle, shape, size, position.. to high-level information such as objects, meaning and purpose… then it filters inputs which further modulate perception. and this where Graphical perception become handy as it deals with the interaction of the human visual system and statistical graphics.

Human being perception plays a major role for decoding graphs appropriately, a good understanding of perception can significantly improve both the quality and quality of information a graph is conveying. For example, one of the key findings from Cleveland and McGill’s paper was that some elementary perceptual tasks are more accurate than other i.e. Position along a scale is more accurate than length, while length is more accurate than area or volume in graph analysis[1]. These elementary tasks are showed in Figure 1.

Visualizations is a technique that researchers, presenters and other individuals use to communicate information with their audience. A presenter or writer want to convey a message and wish that their audience clearly perceive the message as it was conveyed to them. In the research conducted by Cleveland and McGill, they have noticed that people had different perception and this affect the way they interpret the same graphics or pictures. This finding that was discovered 3 decades ago, has been also noticed and confirmed[2] by Jeffrey Heer and Michael Bostock in their paper on using crowdsourcing on Amazon’s Mechanical Turk.

In 1932, Frederick Craxton did a graphic comparisons by bars, squares, circles and cubes and he found that bar are more effectives in communicating comparative information more than squares, circles and cubes, while circles and squares were the same and cubs being the worst[3].

Figure 2. Croxton found that participants were more accurate when they compared the sizes of bars of unequal lengths. Squares and circles were not much different. All three shapes fared better than cubes.

Figure 2. Croxton found that participants were more accurate when they compared the sizes of bars of unequal lengths. Squares and circles were not much different. All three shapes fared better than cubes.

Haroz, Kosara and Franconeri, conducted a study using pictographs instead of generic shapes to represent data in charts. They found that people remembered the data which were presented using discrete shapes ( generic circles or pictographs) compared to data represented using single bars[4]. While using pictographs as replacement for text on axis caused many errors, figure 3.

Figure 3. Haroz, Kosara and Franconeri found that replacing generic stacked shapes with stacked pictographs did not result in a decrease in accuracy for participants.

Figure 3. Haroz, Kosara and Franconeri found that replacing generic stacked shapes with stacked pictographs did not result in a decrease in accuracy for participants.

All visualizations are not that perfect to clearly present/convey information to the viewers so that they can accurately process it; perception is very important because it helps us to select the most effective way we can present our dataset using good visualization which will contribute to avoiding misunderstanding and confusion.

In conclusion, perception affects how audience(viewers, readers…) receive and process data, and in turn affects how a visualization should be done. Understanding how perception changes to different dataset presentation is critical for us to choose appropriate visual tools to use while presenting data effectively and letting our audience understand our visualization accurately.

Reference

  1. Graphical Perception: Theory, Experimentation, and the Application to the Development of Graphical Methods. William Cleveland, Robert McGill, 1984.
  2. Crowdsourcing Graphical Perception: Using Mechanical Turk to Assess Visualization Design. Jeffrey Heer and Michael Bostock, 2010.
  3. Graphic comparisons by bars, squares, circles and cubes. Frederick Croxton, 1932.
  4. ISOTYPE Visualization - Working Memory, Performance, and Engagement with Pictographs. Steve Haroz, Robert Kosara and Steven Franconeri, 2015