Recently on this blog we have talked about the cognitive mapping of a city, exploring how individuals see and record their own neighborhoods. As I write this I can’t help but recall Kevin Lynch’s The Image of the City. He analyzed the organization of cities and subsequently pointed out the way we navigate through these spaces. Lynch identified five elements: path, node, edge, landmark, and district, helping us form the spatial map of the city.
Half a century later, computer generated mapping tools like Geographical Information System (GIS) have become indispensible to planning at any level, whether local, regional or global. Similar to the GIS mapping system, the human mind posses a very sophisticated mapping system with many differentiated layers of maps of the city or neighborhood where we live. These mental maps are capable of guiding us to our destinations in the shortest and most pleasant way possible.
The human mind is capable of creating a large number of assorted maps, retained for different purposes and altered with new information every time they are used. These maps are for different modes of transportation but essentially work the same way whether the destination is a path to school, to a local grocery store, or to grandma’s house on Long Island. I think the most unusual aspect of these human maps is that the visual cues on the maps are tagged with emotional values. As a child I avoided a particular street on my way to school because of a nasty dog that would rush out of a yard without warning, and chase me half a block at break neck speed. I know many women avoid deserted streets out of fear of crime, and sometimes we just pick a path because we feel more comfortable because of better lighting or shaded trees.
This human mapping system is incredibly sophisticated, constantly evaluating and updating maps as we use them. I believe our mapping system is closely tied to a part of our brain which is responsible for our sense of safety and well-being. If we did not posses that ability to tag our mental maps with things which either threatened our safety or guided us to the safety of our homes, we as a species would long since disappeared and have been forgotten in the savannahs of Africa.
But, interesting as this is, let’s get back from the savannahs of our forebearers to our favorite city Paris.
What makes Paris look like Paris is not just the Eiffel Tower but a host of small architectural features like street signs, window gates, balcony’s and lamp posts which are the distinctive features of a Paris streetscape. The researchers at Carnegie Mellon University and Ecole Normale Sup´erieure, Paris tried to prove just that.
They conducted an experiment by geo-tagging thousands of images of the most distinctive architectural features of Paris by using Google Street view. The discriminative clustering approach of their software found representative images of Paris. Not surprisingly these are the same images that we have retained in our own mental maps of the Paris, and we intuitively recognize them as quintessentially Parisian, except that our own images of Paris are also tied to a great meal in St. Germain or a late night stroll along the banks of Seine.
The researcher at Carnegie Mellon also applied this approach of searching images from other cities to identify the distinctive styles of London and Prague. However, this method of finding distinctive features of a city came up short when applied to American cities. What does that say about the character of our cities?