Due to the wide variety of research methods on autism, the author sees the need for in- depth methodological reflection before planning own research on semantic and lexical competences in children with autism. For this purpose, using the method of the traditional literautre review she aims to discuss some theoretical problems and formulate conclusions for her own research. Con-cludes that studying linguistic competence in autism requires intensive conceptual work, including in-depth methodological reflection; practical skills within research methods and epistemological awareness. In the final part, she outlines a plan to investigating verb comprehension in autistic children.

The mechanism of interaction proposed in the research by Levinson and Holler (2014) in the cultural plane is the basis of intentional human behavior inscribed in the theory of mind. It is the basis for the development of communication competences. In order to understand the meaning of individual words, e.g. personal pronouns, the speaker and the recipient must place them in a social, conceptual and extra-linguistic context known to both the sender of the message and the recipient. Only mastering such factors in communication as: exchange of glances, pointing the finger, imitation and cooperation will allow a person with an autism spectrum to make progress in acquiring language and social communication. In connection with the above, I conducted a study the aim of which was to determine whether there are factors that build the "mechanism of interaction" in the communication of people on the autism spectrum? The research group consisted of 78 people (including 64 boys, 14 girls) with a diagnosed autism spectrum, aged 5–10 years. Method: observation, experiment and statistical description. Conclusions: The set of factors constituting the foundations of social commu-nication (known as the interaction mechanism) is disturbed in the autism spectrum. However, there were differences in their use in early childhood autism and Asperger's syndrome.