本研究目的是觀察部件辨識對中文的形音轉換的影響，並且在三個實驗中操弄表音一致性、聲旁結合度。過去研究顯示，低頻低一致性的字需要較久的時間才能唸出讀音。本研究試圖從行為以及腦波反應上，觀察部件被當作而聲旁使用的字數(聲旁結合度)會不會影響表音一致性效應。實驗一使用唸名作業。實驗二則是請受試者進行一個語音辨識作業，在心中默唸並且同時記錄腦電波。實驗三使用字彙判斷作業並且同時記錄腦電波。結果在唸名作業獲得抑制的聲旁結合度效果，以及接近顯著的聲旁結合度和表音一致性交互作用。語音辨識作業中，低表音一致性的字會產生比較正向的P200，而且N170和N400都沒有任何效果。在CDT的前兩階段則先呈現低表音一致性時有促進的聲旁結合度效應，然後高表音一致性時有抑制的聲旁結合度效應。每個階段在N170都呈現在低表音一致性時低聲旁結合度比高聲旁結合度較負向。只有在第二階段中，P200呈現聲旁結合度和表音一致性交互作用。這些結果顯示部件表徵和語音表徵之間有直接連結。Three experiments are aimed to examine the contribution of radical-level information to Chinese orthography-to-phonology transformation by manipulating character consistency and combinability of the phonetic radical. Since previous studies have demonstrated that to read aloud low frequency characters with low consistency reveal longer reaction time, we wonder if the how many phonograms share with a same phonetic radical (combinability) would interact with the consistency effect by behavioral and event-related potentials studies. Experiment 1 was a naming task. In Experiment 2, participants were required to silently read characters in a phonological recognition task (PRT) with EEG recording. Experiment 3 was a character recognition task (CDT) with EEG recording. In the naming task, there was an inhibitory combinability effect, which was opposite to the findings of previous studies, and marginal significant consistency-by-combinability interaction. In the PRT, low consistency characters showed more positive P200 than high consistency characters. There was no any effect neither in N170 nor N400 in PRT. In the first two blocks of the CDT, there was a facilitative combinability effect in low consistency characters first and then an inhibitory combinability effect in high consistency characters. Low consistency/ low combinability characters showed more negative N170 than low consistency/ high combinability characters in all blocks of CDT. Besides, there was consistency-by-combinability interaction in P200 of the second block only. These findings suggested a direct link between orthography and phonology.