The term 'neural network' is generally used to describe populations of interconnected neurons. These neurons communicate with each other via synapses and form the building blocks of the entire cortex. Modeling these neural networks is an integral part of translating the results of sensory testing to neurological diagnoses. One of the most important models we have developed is for the coritical minicolumn. The structure and density of cortical minicolumns have been linked to neurological disorders such as autism. Using computer models of these minicolumns, researchers can introduce controlled operational anomalies to see which models match minicolumner patterns those found in the cortex of a person afflicted with neurological disorder in question. The models provide a way for researchers to 'check' their sensory testing results and make sure that the inferred underlying neurological cause of the symptoms could theoretically account for all of the observable variables.