The Effect Of Relative Frequency Of Phonetic Radical And Sinogram On Sub-lexical Phonological Processing

The research of sub-lexical phonology processing is limited and most of research focuses on the pre-lexical phonology processing. As an ideographic character, Chinese characters are composed of radicals that are freestanding sonograms and have their own pronunciations and meanings. 87% modern Chinese characters have a semantic radical and a phonological radical. The phonetic radical provides voice information related to the Chinese characters and the semantic radical offers semantic information. The sub lexical phonological processing occurs processing the phonetic radical before the extraction of whole word speech information in reading Chinese characters, and most of research focus on the essence of sub-lexical phonological processing and its relative time course to the whole word phonological processing. Most studies have found that the processing of sub-lexical processing only occurs in low frequency words, while high frequency words do not have any sub-lexical processing. We further analyze the materials in previous studies and found there is confusion between the experimental variables and the relative frequency of phonetic radical and the whole word: for low frequency Chinese characters, the frequency of phonetic radical was significantly higher than that of whole word frequency. This means that the sub-lexical phonological processing of low-frequency words may be due to differences in the relative frequency of phonetic radical and the whole word. We conjecture that, if the relative frequency of the phonetic radical and the whole word is control, the sub-lexical phonological processing may disappear. In addition, the relative time course of sub-lexical phonological processing and whole word speech processing is also the key point of this article. The time course of each processing stage in lexical processing has been very clear, but the time course of sub-lexical phonological processing and the whole word characterization is need more attention. Accordingly, we design the Experiment 1 to explore the sub-lexical phonological processing and its processing time course when the material is low frequency irregular characters and the relative frequency of radical and the whole word is balance. On this basis, we further assume that the high-frequency word processing is also affected by the relative frequency of radical and the whole word. Based on this point, we designed the Experiment 2 to explore the influence of the relative frequency of radical and the whole word on the characterization of the sub-lexical processing and its time course.In Experiment 1, the priming paradigm and a semantic judgment task were adopted. The targets were low frequency irregular characters and the relative frequency of radical and the whole word was balanced. Each target matching four primed condition, respectively, the whole word related primes(pronunciation is same as the whole word) and the matched control condition, phonetic related primes(pronunciation is same as the radical of the word) and the matched control conditions. The task is to determine whether the priming words and target are semantically related words. We found the correct rate of whole word related primes was significantly lower than that of the control condition and had a long reacting time in the behavior result, and evoked a smaller P200; phonetic related primes and control conditions is no significant difference in the correct rate and reaction time, nor showed differences in ERP components. This verifies our hypothesis, for low frequency words, the relative frequency is an important factor influencing the occurrence of sub lexical processing.In Experiment 2, using the priming paradigm and phonological judgment task, using high-frequency irregular word as targets which divided into two classes. The frequency of the whole word is higher than that of phonetic radical in one class, and the frequency of the whole word is lower than the frequency of phonetic radical in the other class. Each target had three different types of primes: The whole word phonological primes, the radical phonological primes, and the unrelated control primes. The participants were asked to judge whether the target and the prime is the same. The results showed that there is no effect of relative frequency on sub-lexical phonological processing for high frequency word. The correct rate and reaction time of the radical phonological primes were significantly lower than those of the control primes, and elicited more positive P600 component; similarly, the correct rate of whole word primes was significantly lower than that of control conditions, and evoked a smaller N400 component. This suggests that, different from low frequency irregular characters, sub-lexical phonological processing of high-frequency irregular word of was not affected by relative frequency.The results in Experiment 1 and Experiment 2 show that, for low-frequency words, the sub-lexical processing is affected by the relative frequency of phonetic radicals and the whole word. when the phonetic frequency and whole word frequency is balance, the sub lexical phonological processing disappeared; for high-frequency words, the characterization of the sub-lexical phonological processing occurred stability whatever the relative frequency is and the sub-lexical phonological processing is early before the whole word phonological processing.