One of the main contributions of Lynch's work was transforming a messy and idiosyncratic mental image into a representation which makes it possible to compare the mental images of different spaces. He believed that the contents of city images could be classified into five types of elements: paths, edges, districts, nodes and landmarks (1960). A highly imageable (apparent, legible, or visible) city is one with distinct and identifiable elements which are connected in a meaningful way. Imageability as a quality of the environment can be analysed by reading mental images as a combination of paths, edges, districts, nodes and landmarks, a unified procedure he developed for cities like Boston and Los Angeles, and applied later to other places (Appleyard, 1973).

Despite its universality, this approach has limitations. All the five elements of Lynch's mental image (paths, nodes, landmarks, edges and districts) refer to physical elements or perceptible objects, suggesting that the form is what defines the mental image. This view does not take into account the influence of functional or symbolic meaning and the way people actively interact with environment. According to Yadav, what Lynch studied was an “implicitly passive or responsive mode of perception” (1987, p. 6) which separated physical form from meaning.

The role of mental images beyond navigation was also not clear in Lynch’s work. How are mental images connected to human behaviour? Why is a strong mental image better than a weak one? Lynch argued that mental images are as important for well-being as they are for way-finding, but establishing this connection was not the aim of his work:

The study may have analysed the nature of the way finding image accurately enough. But it only assumed its importance and never demonstrated it. What do people care if they have a vivid image of their locality? (Lynch, 1984, p. 154).