![]() |
||||||||||||||||||||
|
Spring 2008 Colloquia |
|||||||||||||||||||
| Thursdays |
4:00-5:30 |
N119 Elliot Hall - campus map | ||||||||||||||||||
April 10 |
![]() |
|
||||||||||||||||||
Abstract At the morphology lab at the College of Architecture, Georgia Institute of Technology, we study the formal principles that govern the generation, intelligibility and function of built form. Particular emphasis is placed on the development of descriptions and measures of the properties of built form that are associated with its cultural and cognitive intelligibility and functions. This usually involves three steps: 1) a reading of the geometry of architectural plans; 2) a representation of different kinds of constituent geometrical entities, such as lines of movement, convex spaces or visibility polygons; 3) and a computation of the graph-theoretic properties that describe the relationships between these entities. I will discuss research and as yet unpublished findings which I believe are relevant to studies of spatial cognition, deliberately addressing the scales of building interiors and urban space. An analysis of behaviors at a Neural ICU at Emory Hospital, Atlanta, shows that as they move and interact, nurses automatically tune their position in the environment according to patterns of visibility targeted towards patient beds. Doctors, on the other hand, tune their positions according to patterns of general visibility. The results indicate that the reading of environment is influenced by organizational role but operates according to generic principles that have to do with patterns of occlusion: Whether we talk about lines of movement or lines of sight, the question “how many corners away things are” is more important than the question “how many steps away things are”, at least in the context of a relatively compact setting. An analysis of the distribution of retail frontage along the streets of the city of Buenos Aires shows that the density of retail frontage is driven not only by an inverse relation to distance from the city center (a traditional gravity model) but also by the density of the surrounding street network (a context dependent, configurational gravity model). However, the measure of density which best explains the distribution of retail is the length of streets in the surrounding environment that are accessible within two direction changes, rather than the length of streets that are accessible within a metric distance of walking. This implies a cognitive bias in the way in which we read the attraction potential of individual spaces, as a function of their fit in the surrounding network. Some of the implications of the finding are particularly interesting in the context of this colloquium. Traditional measures of street connectivity which are essentially based on metric properties (physical effort) have to be complemented by measures of connectivity which are sensitive to configuration (cognitive effort). From a broader point of view, the findings indicate that the spatial structure of streets is the underlying framework that governs the evolution of land uses, for reasons which, we believe, have to do with the relationship between urban intelligibility and urban functionality as well as with the practicalities of planning and design. Readings:
|
||||||||||||||||||||