Perceptrons, like more complicated backpropagation artificial neural networks, are typically trained to minimize mean square error (MSE). In computer-aided diagnosis (CAD) applications, model performance is usually evaluated according to other more clinically relevant measures from receiver operating characteristic (ROC) analysis. The purpose of this study was to investigate the relationship between MSE and the area (Az) under the ROC curve and the partial ROC area (0:90Az) under the high sensitivity portion of the ROC curve. A perceptron was used to predict whether or not breast lesions were malignant based on two mammographic findings and patient age. For each performance measure, the error surface in weight space was visualized. Comparison of the surfaces indicated that minimizing MSE tended to maximize Az, but not 0:90Az. If it is important to maximize 0:90Az, then predictive models trained to minimize MSE may provide inferior solutions.