2001-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/707639摘要：經由多年多次的重複檢驗，申請人的實驗室累積了大量繁體漢字唸名作業及字彙判斷作業數據，描繪出中文單字促發實驗的大致特性。在目標字特性上，出現頻率是一個最重要影響變項，其他任何操弄變項的影響力都遠遠不如。在促發項目特性上，整字與目標字間的關係會造成重要的影響，其中，語義相關或相似的促發字在目標字為低頻情況下會出現極強的正向促進效果，語音相同的促發字僅在唸名作業目標字為低頻的情況下偶爾出現微弱不穩定的正向促進效果，字形相似的促發字則在唸名與字彙判斷作業下都會對高頻或低頻目標字產生極穩定的「抑制效果」，雖然這個效果的影響力不若目標字頻效果及低頻目標字下的語意關連促發效果為強，但卻穩定的可被觀察。SOA 的操弄大致不會改變這個面貌。這些結果反駁了一些利用漢字促發程序以強調漢字辨識歷程語音角色所得的過去實驗結論，也蘊涵漢字可能在目前實驗技術上仍無法掌握的極短時間內就完成了穩定的辨識確認，不會有像英文文獻裡技術上所操弄SOA 範圍內所顯示的「時間歷程」。最近的後續研究針對了形似抑制的現象進行了更深入的實驗，在目標字頻被限定在中等頻率情況下，操弄了促發字的相對出現頻率，包括促發字頻比目標字頻為高、為低<br> Abstract: It was argued from the previous studies that the experimental results supporting the role of pre-lexical or pre-semantic phonology , conducted by Perfetti & Zhang (1991), Tan & Perfetti (1997), Perfetti & Tan (1998), Cheng & Shih (1988), and Cheng (1992), are untrustworthy. During the works in attempt to replicating their results, it was also found that obvious inhibitory effect was obtained under graphemic prime condition. This extends our theoretical viewpoint and thus proposed a three-year research plan to explore the reliability of graphemic inhibition effect, the rationale of priming procedure, and whether it belongs to the domain of "perception" or "language". During the first year we had thoroughly replicated many related experiments and comfirmed the reliability of graphemic inhibition effect. Furthermore, it seemed that using pseudo-character as prime there was no graphemic inhibition effect obtained, compared to those character prime conditions. It is thus proposed that the graphemic inhibition phenomenon belongs to the domain of language rather than perception. We are comfirming this proposal during the second year research. In the meantime, because there still cumulated no thorough data in priming research using radical component itself as prime, the third year was planed to conduct such a line of experiments.促發程序形似促發念名作業字彙判斷作業priming proceduregraphemic primingnaming tasklexical decision task