Comparison of brain mechanisms underlying the processing of Chinese characters and pseudo-characters: An event-related potential study

Most Chinese characters are composed of a semantic radical on the left and a phonetic radical on the right. The semantic radical might provide the semantic information and the phonetic radical may provide information concerning the pronunciation of the whole character. The pseudo-characters consisted of different sub-lexical parts of real Chinese characters and consequently they also had the semantic radical and the phonetic radical. But they could not be readable and have no actual meaning. In order to investigate the spatiotemporal cortical activation patterns underlying the orthographic, phonological and semantic processing of Chinese characters, we used Event-related brain potentials (ERPs) to explore the processing of Chinese characters and pseudo-characters when 14 healthy Chinese college students viewed the characters passively. Results showed that both Chinese characters and pseudo-characters elicited an evident negative potential peaking around 120 ms (N120), which appeared to reflect initial orthographic distinction and evaluation. Then, Chinese pseudo-characters elicited a more positive ERP deflection (P220) than did Chinese characters between 200-250 ms after onset of the stimuli. It was similar to the RP (recognition potential) and might reflect the integration processes of phonological and semantic processing on the basis of early orthographic information. Dipole source analysis of the difference wave (pseudo-characters minus characters) indicated that a generator localized in the left temporal-occipital junction contributed to this effect, which was possibly related to phonological and perceptual-semantic information integration. Between 350-450 ms, a greater negativity (N360) in pseudo-characters as compared to characters was found over midline fronto-central scalp regions. Dipole analysis localized the generator of N360 in the right parahippocampal cortex. Therefore, the N360 might be an N400 component and reflect the higher-level semantic activation on the basis of orthographic, phonological and perceptual-semantic processing.