Cluster Structure of Cortical Systems in Mammalian Brains

Abstract

Information about the numerous connections among the many cortical areas in mammalian brains has been published extensively. The organization of the cortical systems that these connections define, however, is still poorly understood. We analyzed this complex connectivity structure with the help of mathematical and computational tools in order to better understand cortical processing. Here we present the results of a number of different analyses that we used to decide whether cortical areas in the brains of the cat and the Macaque monkey can be grouped into separate clusters, based on the organization and strength of connections between the areas. We employed non-parametric cluster analysis together with multi-dimensional scaling, as well as a new computational tool based on the global stochastic optimization of interconnected structures. Our results revealed a well-defined clustered organization, which was particularly clear in the primate visual system. The identified clusters largely agreed with suspected functional sub-divisions of cortical regions. Our results provide a basis for modeling studies, for the interpretation of functional studies of the brain; as well as for predicting future anatomical experiments.