The main purpose of the present study was to investigate how proper and common nouns are represented in the brain independent of memory retrieval processes. Participants were instructed to perform a lexical decision task while dense-array EEG was continuously recorded. Both ERP components (namely N400 and P300) and swLORETA suggested that proper name processing engaged a more widespread neural network and required more cognitive resources than common noun processing. Overall, our results come down in favor of the hypothesis that specific effects of proper vs. common noun processing exist, and they suggest a possible neuro-functional segregation of proper vs. common noun processing. The difference in proper and common noun processing seems to emerge at the level of storage or representation of lexical knowledge, and it may crucially depend on their semantic characteristics.