Date of Award

1986

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Abstract

The functional region is usually conceptualized as a spatial partitioning that identifies communities of interest. Attempts at quantitative definition have mimicked analogous concepts in the social sciences, especially social clique identification. Although it is accepted that the functional region is based on human interaction, little attention has been paid to interaction modeling in the context of functional regions. This is partly because the spatial interaction model is not equipped to handle behavioral biases.;This research develops a model of spatial interaction that recognizes perceptual affinities or barriers between spatial units. Derived from the Luce and Tversky choice axioms, the Black/White Model relaxes the assumption of isotropism in landscapes, and offers an alternative conceptualization of the functional region. Four classes of the model are distinguished, to address the various conceptual structures traditionally associated with regions in general, and the functional region in particular. It is argued that the calibration of the Black/White model serves two purposes: first, it allows the model to be applied in a predictive capacity; secondly, it identifies regional memberships, and is thus in effect a mechanism for regionalization. A method of calibrating the Black/White model is developed. The technique is initially illustrated on simulated data, and is shown to retrieve embedded regional structures over a range of scenarios. A heuristic calibration method is developed for large problems. The methods are compared under conditions of controlled random error; their performance is documented with various simulated spatial and regional configurations.;The Black/White regionalization method is demonstrated on a data matrix of student migrations in the United States. A number of refinements in the heuristic are effected in the light of the initial results, and there are indications that the algorithm may have identified the optimal Black/White partitioning, given the data.;In conclusion, while further work remains to be done on the detection of complex regional structures, the Black/White model offers a method of binary partitioning, and a fresh outlook on regionalization and spatial interaction, with promising implications for both planning and academic enquiry.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.