A new approach to testing mental attitudes, psychological characteristics and the level of competence of students is proposed. This approach is based on the use of an electronic textbook coupled with a diagnostic neural network. Pairing is ensured by the fact that the text of the textbook is supplied with hyperlinks, which de facto form the first layer of the diagnostic neural network. Each of these links is associated with a formal neuron, and the fact of following the link is associated with the formation of a logical unit at the output of such a neuron, the absence of a transition is associated with a logical zero. Next, the standard procedure for training and using neural networks is used, and the training sample is formed using psychological tests. A specific example of an electronic textbook associated with a neural network is considered, which is built on the basis of a textbook on the history and philosophy of science, officially recommended by the Ministry of Science and Higher Education of the Republic for all specialties of the magistracy. The results of the primary testing of students using the proposed technology are presented. It is shown that this method can be used not only for testing individuals, but also for diagnosing social groups.