Interest towards cognitive mapping in psychology and geography was a response to scientific positivism, an attempt to explain human spatial activities by means of human cognitive processes (Kitchin, 1994). The behavioural scientists believed that "an understanding of spatial knowledge necessarily leads to an understanding of people's spatial behaviour" (Kitchin & Freundschuh, 2000, p. 4). They saw cognitive maps as mental devices used to structure the complexity of the environment into a meaningful pattern which can guide behaviour:
In effect, a cognitive map is a mental devise and store which helps to simplify, code and order the endlessly complex world of human interaction with the environment (Walmsley et al., 1990, as cited in Kitchin, 1994).
Cognitive maps, mental images and similar constructs are "complex, highly selective, abstract and generalised" (Downs & Stea, 1973, p.18). They are a result of abstraction, which is similar to the map-making process when a highly complex environment is reduced to a concise representation (or a map), with the goal to help people navigate space.
From a behavioural and cognitive perspective, cognitive maps are merely wayfinding aids that help people in their spatial problem-solving tasks. They are organised in a hierarchical way, ordering the environment by means of 'environmental cues' (Golledge, Richardson & Gale 1987, p. 216). These cues are significant physical elements which we connect in our minds as we learn how to navigate between them. Over time, we learn more about a set of cues and connecting paths and our spatial knowledge becomes more like a real map.
But the analogy with maps is a misleading one. Our internally stored information about the surrounding environment often lacks the coherence of a geographically precise map where all elements are organised by means of spatial relationships. Cognitive maps are often characterised as "incomplete, distorted, schematised, and augmented" (ibid) and even "simplified and idiosyncratic" (Kitchin, 1994, p. 4). Researchers connect the errors in cognitive maps to the limited processing capacity of our brains which cause distortions while organising a vast amount of discontinuous environmental experience, a process also known as cognitive economising (Appleyard, 1973).
If we shift our attention from the mental processes behind cognitive mapping to the mental images themselves, we can see that different environments produce images of different clarity or cohesion. Some spaces are easier to translate into a concise mental representation, while for others we need some time to understand how they are organised. Moreover, in some spaces discovering new areas seems easy, while in others there is a constant feeling of being lost. Mental representations of spaces provide insights not only into the cognitive processes behind the perception of space, but also into the qualities of the environment that make it easier for an observer to structure and navigate space.
Analysing urban spaces through mental representations was pioneered as a method by Lynch (1960; 1984). He suggested that well-designed environments evoke stronger cognitive images, making it easier for an observer to understand the structure of the space. He called this quality imageability (also known as legibility or visibility), "that quality in a physical object which gives it a high probability of evoking strong image in any given observer" (1960, p. 9). He referred to cognitive maps as mental images and studied them through interviews, sketch maps and surveillance on foot by trained observers.