A feedforward neural network with one hidden layer is applied to the problem of brainstem auditory evoked potential classification. Network performances were tested separately both on subject-dependent samples (drawn from the same subjects from which the training set was derived) and on subject-independent samples (drawn from subjects from which no data were included in the training set), and compared. The results indicate that, while increasing the training set size improves performance, human-selected training sets give better results than randomly selected sets. Different encoding schemes used for representing the signal yield varying rates of correct recognition. Although the networks were overly complex and trained in the memorization mode, they show some feature extracting and generalization capabilities.