Two methods of identifying schizophrenia, Parkinson's disease (PD), and Huntington's disease (HD) are described. The methods are based on the analysis of the contingent negative variation (CNV), an event related potential (ERP) in the electroencephalogram. The first method involves spectrum analysis of the CNV and discriminant analysis of the Fourier harmonic frequency components. The other method involves the application of supervised learning artificial neural networks to the CNV features obtained in the time domain. Additionally, unsupervised artificial neural networks were used to presymptomatically assess the risk of HD. Sensitivities and specificities lie between 0.81 and 1.0 with low false positive rates (0 to 0.13) for differentiating between disease and normal data, and between disease data, dependent on disease and method. The preferred method for disease differentiation for accuracy and ease of application is the multilayer perceptron. Using Kohonen and ART networks for detecting abnormal CNVs in subjects at risk of HD (ARs) eight abnormals are identified in agreement with the prediction of risk derived from a published risk table. In addition, one of the abnormals has since developed symptomatic Huntington's disease. The recommended method is to combine the results of the Kohonen method with an ART2 and a modified ART1 network.