INTRODUCTION: The autism spectrum disorder known in the literature as a pervasive developmental disorder is making doctors and scientists very concerned with its increasing prevalence and incidence, both in Croatia and worldwide. Despite the early symptoms occurring between 12 and 18 months of life, an autism spectrum disorder is usually detected after the child's lack of speech production at the age of 3 to 4 years. When treating patients with autism spectrum disorder, apart from detailed (hetero) anamnesis and physical examination, electroencephalography is also used as a method for detecting changes in brain electrical activity. The electrical activity of the brain may be normal, but it also may show borderline or pathological changes, as we have shown that in our research. AIM: Our research goal was to examine the purpose of electroencephalography in diagnosing and monitoring children with autism spectrum disorder. MATERIAL AND METHODS: Through a retrospective analysis of the history of the disease and all available medical documentation as well as the electroencephalographic reports described by the neuropediatricians in the Clinical Hospital Center of Rijeka, the Pediatric Clinic, the Department of Neurology, and Pediatric Psychiatry, we included in the research of 72 children who match the given criteria. The data are represented by the total number and percentage share, and the measured variables are presented through descriptive indicators. RESULTS: Of the total number of children in the study (n = 72), 45 children had an EEG report. A statistically significant less EEG report was done in a girl (p = 0.00001). Electroencephalographic reports in 71.1% of children (n = 32) were normal - 25 male and 7 female children. Electroencephalographic reports of children were done during wakefulness and sleep. Three children were sleep-induced with melatonin while one child had a shorter sleep deprivation. In 36 children, the EEG was registered in sleep while at 9 in the wakefulness. The EEG report in the wakefulness in all children was normal and in sleeping was 23 normal, 6 border and 7 pathologic EEG reports. Statistically significant more pathological and border EEG reports were found during EEG sleep regression, indicating a greater sensitivity of such a method of registering in children with autism spectrum disorder. CONCLUSION: From the above results we can see that most children with autism spectrum disorder have a neat and normal EEG report. Despite the importance of using this method in the primary treatment of children with autism spectrum disorder due to the wide differential diagnosis, we argue that for a clear clinical picture of the autism spectrum disorder this method is not necessary. Furthermore, more research needs to be conducted on a larger number of subjects and further monitoring and education on autism spectrum disorder is also needed in order to facilitate detection and early diagnosis of these disorders.