Latour takes an anthropological looks at what gives visualizations their cognitive value and comprehension. After reviewing several anthropological, psychological, and business-related works, he found that Visualizations are most effective when they contain certain characteristics.
First, visualization must have elements of optical consistency. One of the most effective elements of optical consistency is perspective. Perspective is the reason why many graphs and, especially, maps seem incomplete or confusing without legends or scales. Our brains are nearly automatic when it comes to taking something we see in one picture, and comparing it with object in another picture as long as we have a baseline to do so.
Second, it must obey by the “visual culture” at the time of the visualization’s creation. Visual culture is an abstract requirement that essentially requires the photographer or artist to have elements in the photo or work that allows the observer to assessing its own worldly attributes to it. The work can be viewed at a future time, but still be understood to be a snapshot of a different time. The overall picture or message is still clear, regardless of when the picture is viewed.
Third, and related to the second requirement, a visualization is most effective when is can be understood relatively. The ability to publish visualizations have made this requirement easier to meet. The ability to publish makes visualization mobile (able to be view across a wider time and space) and immutable (able to remain unchanged over time).
After outlining what makes for the greatest mobile and immutable visualizations, Latour explores how the use visualizations help people understand otherwise overwhelmingly complex phenomena.
While anything can be re-imaged or re-visualized, Latour argues that consistency is key. A dissenter can go find various illustrations of his/her positions, but too many visualizations may actually harm his/her cause. Like scientific theories, visualizations are best understood when being conveyed in a consistent fashion. As a very simplified example, ‘bar graph’ issues can be become convoluted when too many start to use pie graphs to portray them. Spatial dynamics would be much more confusing displayed in a table rather than a map. Since visualization be produced and dispersed at low costs, consistency is key.
In addition, visualizations make otherwise complicated, 3 or more-dimensional phenomena into flat representations. When these issues are illustration sufficiently on a flat venue, greater comprehension and communication is achieved – especially when the visualization is coupled with a written text.
However, this requirement seems to be mainly useful for photography and art, and is of little importance to intelligence analysts. Latour’s exploration of visualizations makes intuitive sense, but there are little experimental citations in his writing. However, he does include plenty of anthropological and scientific research to guide his exploration. Until his intuitive points are proven wrong in an intelligence-setting experiment, analysts should follow his recommendations. Visualizations are a valuable modifier, if not a method.
Latour, B. (1983). Visualization and cognition: Drawing things together (pp. 1–33). Boston, MA: Harvard University. Retrieved from http://isites.harvard.edu/fs/docs/icb.topic1270717.files/Visualization%20and%20Cognition.pdf