Modular associations are structures in complex networks that are defined based on the communication density between the network elements. The difference in these structures in a complex network of human brain signals (EEG) can be used as a factor in the diagnosis of diseases. In this study, with the focus on modular associations, attempts to achieve the differences between a complex two-group of network of Normal Case (NC) and Autistic of Spectrum Disorder (ASD). Eventually, using real EEG signals, the tested groups, with no use of the pre-processing signaling, have an accuracy of 88.37% in detecting Autism Spectrum Disorder (ASD).