In modern information and search automated systems, the main role is played by the human-machine interface for decision-making based on the detection of objects of a given class in the information field. The human operator, as an element of such an interface, carefully reviews the provided information on the information field – the monitor screen, analyzing images of scenes, for example, pictures of territories, economic tables, abstract data, etc. Based on the received results of the analysis, the operator forms or chooses from a set of alternatives the appropriate decision, for which he mostly bears some responsibility. Such, let’s call them primary solutions, can be both final and the basis for further functions of this search engine. However, among the factors that in one way or another can negatively affect the quality of work of this interface, one of the first places is stress in the form of a collection of micro stresses. The main goal of this study is to develop a methodology for the organization of experimental selection of operator personnel based on the study of their behavior under the influence of micro-stresses. A human-machine interface model has been developed, which takes into account the change in the functional state of the human operator. The presented concept of the difficulty of detecting the object of attention contributed to the development of a special sequence of ordinary test images with stressor images included in it, and presented models of the flow of presenting test images to the recipient. With the help of descriptive statistics, the parameters of individual boxplot diagrams were determined and the recipient group was clustered. Overall, the proposed approach based on the example of the conducted grouping makes it possible to ensure the objectivity and efficiency of the professional selection of applicants for operator specialties. The developed methodology is of practical importance and can be used in systems of training, certification and professional selection of operator personnel.