組成字對學童雙字詞辨識歷程的影響

Abstract

本研究旨在探討國小學童詞彙辨識歷程是否牽涉詞以下單位詞素組成的運作，以及學童辨識中文詞時是否受到詞的鄰項個數影響？Liu, Chung, McBride-Chang和Tong（2010）提出國小四年級學童在詞的辨識歷程以整體式進行辨識，對於字，則是以分析式進行辨識。該研究透過顛倒／隨機兩個情境進行，詞的部分，顛倒型為真詞顛倒雙字的順序，隨機型為兩個詞素隨機組成；字的部分，顛倒型為真字顛倒部件的位置，隨機型為兩個部件符合組字規則的假字。以詞彙判斷作業進行研究，發現顛倒型辨識反應時間慢於隨機型；但顛倒部件位置的非字，受試者的拒絕反應時間比假字快。因為在詞、字裡得到不一樣的結果，該論文認為中文詞是以整體式，字是以分析式進行辨識。本研究針對詞的部份進行探討，由於Liu等人（2010）在詞的研究部份有若干可能的混淆變項，分別為呈現方式以及實驗程序，因此本研究實驗一、實驗二嘗試澄清這些可能的混淆變項。雖然研究結果支持Liu等人的結果，不過在瞭解詞彙辨識採用分析式歷程後，實驗三檢視學童是否受到組成字，鄰項個數效應的影響，雖然在真詞沒有得到達統計顯著效果，但在假詞獲得抑制的鄰項個數效應。顯示組成字影響學童詞彙的辨識，具有詞以下單位的運作。This study aimed to explore whether Chinese two-character words are processed in a holistic or analytic manner, and the role of constituent component on word recognition in school children. Liu et al. (2010) suggested that grade 4 children adopted a holistic way for two-character word recognition process, but an analytic way for single character recognition process. They employed two conditions ( reversed / random ). On the word level, the reversed condition, with the order of the two morphemes; and the random condition, with two morphemes randomly combined to create nonwords. On the character level, the reversed condition, with the order of the semantic and phonetic radicals; and the random condition, with a semantic radical and a phonetic radical randomly combined using correct orthographic rules to create noncharacter. The results revealed that children performed worse in the reversed condition than in the random condition on the word level, whereas on the character level the results were opposite. Accordingly, they suggested that a holistic approach to word processing and an analytic approach to character level processing. This study intended to examine solely on word recognition. Due to the possible existence of some confounding variables in Liu et al. (2010) research, for instance word positional pattern and experimental procedures, the current study tried to control those confounding variables in experience 1 and 2 to clarify Liu et al. (2010) findings. The study results were in line with the findings of Liu et al. (2010), but an analytic approach to word recognition processing. Experience 3 explored the influence of constituent component for recognizing Chinese two-character words through neighborhood size effect. The finding revealed that constituent component processing was involved in word recognition.