The Influence Of Twirling And Frequency Of Chinese Characters On The N400Semantic Priming Effect

Kutas and Hillyard firstly reported the N400event-related potential, which occurred when unexpected or surprising words occurred at the end of a sentence, in1980. Then, many psychologists have done a lot of studies on this brain electrical component, and put forword different cognitive processing theories, such as the post-lexical theory, the pre-lexical theory, interactive activation theory et al. Two classes of theories-the post-lexical and the pre-lexical differed on whether the activation of the N400is attributable to the semantic processing or the lexical processing. According to the post-lexical, N400was proposed to reflect the semantic processing, and the sensory cognition features, such as word frequency, or the semantic features do not affect the amplitude of N400. Correspondingly, the pre-lexical deemed that the N400reflected the lexical processing, such as the Theory of automatic spreading activation, which believes that the N400in the repetition and semantic priming reflected the unconscious or automatic processing of words. Word frequency and semantic features also have influence on the amplitude of N400.The Chinese language is unique in the characteristics of the language, sharp contrast to English and other alphabetic writing systems. Alphabetic systems are based on the association of phonemes with graphitic symbols, whereas the Chinese system, composed of ideographic scripts, is based on the association of meaningful morphemes with graphic units, i.e., characters. Studies have suggested that the meaning of Chinese single-character words is accessed directly from orthography with little phonological mediation, and phonology plays a stronger role in visual lexical processing in English.In experiment1of high frequency, the variables are twirling (twirled stimuli, intact stimuli) and semantic (primed stimuli, unprimed stimuli), using semantic activation paradigm and lexical decision (the same word or different words).The three-way repeated-measure ANOVA conducted on the mean amplitudes of300-500ms time window demonstrate that more negative-going ERP elicited by intact targets as compared with the twirled targets, and the unprimed targets also elicited a more negative-going N400in all the4columns. Similar analysis in the time window of140-250ms demonstrates the main twirling effect both on the peak and latency. Furthermore, the twirled targets elicited more positive and longer waveforms than the intact targets. In the time window of150-250ms, the four-way repeated-measure ANOVA conducted on the peak and latency indicates a main effect of twirling and a reliable interaction between twirling and occipital lobe.Low frequency experiment uses the same paradigm, decision and procedure with the high frequency experiment. The analysis demonstrates the N400twirling effect and no semantic effect in300-500ms time window. And in the time window of P2and N170, the twirled targets also elicit more negative-going and longer latency than intact targets.According to the two experiments, the conclusion is drawn as followed:First, the twirled targets elicit the more negative-going component, and the twirling interacted with the frequency. These results are consistent with the prediction of the interactive activation model. Twirling the words lead to the changes of orthography, and then the orthographic neighborhood changes. In Chinese language, the families of the orthographic often have the inhibition influence. Therefore, the reaction speed of small orthographic neighborhood is faster than the big orthographic neighborhood, and the mean waveforms are more positive.Second, the twirling effect is also reflected on the P2and N170. And the bigger and later negative component in the ccipito-tem-poral demonstrates that the words can elicit "expert processing".