To make maps that work, we must depict categories using methods that match the structures of human mental categorization. - Alan MacEachren

Robinson (1952) pointed out some limits to approaching map symbolization and design from a purely artistic viewpoint, as he suggested was the guiding perspective at the time. Maps, like buildings that are designed primarily for artistic impact, are often not functional... Robinson (1952) argued that treating maps as art can lead to "arbitrary and capricious" decisions. He saw only two alternatives: either standardize everything so that no confusion can result about the meaning of symbols, or study and analyze characteristics of perception as they apply to maps so that symbolization and design decisions can be based on "objective" rules... Robinson's dissertation, then, signaled the beginning of a more objective approach to map symbolization and design based on testing the effectiveness of alternatives, an approach that followed the positivist model of physical science. In his dissertation, Robinson cited several aspects of cartographic method for which he felt more objective guidelines were required (e.g., lettering, color, and map design). He also suggested that this objective look at cartographic methods should begin by considering the limitations of human perception. One goal he proposed was identification of the "least practical differences" in map symbols (e.g., the smallest difference in lettering size that would be noticeable to most readers).

The fact that map is a fuzzy and radial, rather than a precisely defined, category is important because what a viewer interprets a display to be will influence her expectations about the display and how she interacts with it.

Understanding how maps work and why maps work (or do not work) as representations in their own right and as prompts to further representations, and what it means for a map to work, are critical issues as we embark on a visual information age.