From a cognitive point of view, literary genres do not exist regardless of their perception; they should be embodied in the reader’s mind. The purpose of cognitive genre study is to consider genre as a form of mental representation. According to our hypothesis, this representation combines figurative and propositional ways of encoding information. There is no doubt that the reconstruction of the mental representation of the genre in the mind of the reader requires an experimental study. This article publishes the results of a pilot experiment conducted among university students of pedagogical specialties of the Faculty of Psychological, Pedagogical and Special Education of Saratov University and school students of Saratov. The experimental data show that the representation of the genre in the mind of the reader can be based on the cluster model which suggests that the readers define a number of features by which they identify the genre. The prototype model of the genre is no less relevant: belonging of a particular work to a particular genre is determined in the mind of the reader by its similarity with the best genre model – a standard (or a prototype). The number of such standards is not very large. As a rule, this status is acquired by works that are included in the school literary canon. In the reader’s mind, genre classes are not formations with clear boundaries; rather, they are more likely to be built on the principle of “family resemblance”. The cognitive model of a genre in the mind of the reader has certain features of a field. Conducting cognitive genre research, i.e. identifying ways of representing the genre in the mind of a reader requires the development of a special questionnaire that takes into account the level of the ordinary (naive) reader who is not a philologist.