運用心智圖法之漢字聯想研究：以英語母語漢語學習者為研究對象

Abstract

漢字負載著形、音、義的多重訊息，對母語為拼音文字的二語學習者而言，歷來是學習難點；除了漢字系統本身的困難度，學習者個別的知覺偏好（perceptual preference）亦影響習得。本研究藉字詞聯想（word association）與心智圖法（mind mapping）的方法，透過調查與訪談，探討中級與高級學習者在漢字習得過程中，認知傾向與語義網路的變化。 字詞聯想能顯示語言使用者腦中的字詞連結方式，該方式影響字詞的感知、學習和記憶；心智圖法則是將資訊圖像化的筆記工具，發軔於語義網路的研究，呈現方式亦為放射狀網路圖。本研究以心智圖法為報告方式，選擇難度為TOCFL測驗基礎級及HSK測驗甲級，字頻在120至170名間的象形、指事、會意、形聲字各一（月、本、見、字）為刺激字，針對36位（中級19位，高級17位）在台灣學習漢語的英語母語者，一對一實施字詞聯想測驗、知覺偏好問卷調查，以及訪談；此外，亦訪問了5位現任華語教師，請其就漢字教學的現況提出看法。 研究結果顯示：（一）各字的主要反應詞及聯想類型分布不同，會意字的部件相關聯想傾向較高，形聲字的音韻聯想傾向較高。顯示漢字結構影響聯想內容，表音性與音韻聯想比例相關。（二）四字均以意義聯想為最多，其後依序為形體、音韻及個別化聯想。對照過去研究，可見二語者與母語者的共通性。（三）兩組的主要反應詞及排序不同，中級組傾向產出形體聯想，高級組則有較多意義聯想。顯示兩組處理速度不同，高級組較快觸及心理詞典。（四）聯想類型分布隨漢語程度改變，由中級至高級，形體及音韻聯想的比例降低，意義聯想增加，在比例上逐漸接近母語者，顯示學習者語言認知的發展趨勢。（五）個別受試者的知覺偏好與聯想有關，高視覺偏好者採用圖示及色彩表達的比例高，高聽覺偏好者易產出高比例音韻聯想。 最後筆者根據研究結果提出建議：（一）漢字學習有漸進性及個別差異性，應注重各字的排序與特性，安排於各級程度課程與教材內。（二）教學時可運用聯想網路概念，呈現適當的關聯字詞，以利學習。（三）字詞聯想及心智圖法原理可應用於教學及教材設計，亦可用於測驗編製。Alphabetic background learners always have problems in learning Chinese characters. In addition to the difficulty, learners’ individual perceptual preferences manipulate their acquisition as well. Based on word association and mind-mapping theory, the study investigates what cognitive trend is and how semantic network changes among intermediate and advanced learners. Word association displays the way how one word connects to another in a language user’s cognition system. It affects how words are being sensed, learned and memorized. . Mind-mapping is a graphical note-taking tool. Deriving from researches of semantic network, each mind map is illustrated as a radial network. In the study, 36 participants (19 intermediate learners and 17 advanced learners) are all English-speaking learners of Chinese. They were required to take part in a series of individual tasks: 1) a word association test with selected four target characters (月, 本, 見, 字, which are at same level of TOCFL and HSK) using mind-mapping method, 2) a perceptual preference questionnaire, 3)an interview. Another 5 Chinese instructors were also been interviewed. The study shows that 1) each target character differs in primary responses and association types, 2) participants have the most semantic responses in all four target characters, followed by form-related, phonetic, and individual responses, 3) the intermediate group shows more form-related responses than the advanced group, which has more semantic responses, 4) the distribution of association types changes with Chinese proficiency level, the advanced group’s performance is more similar to native Chinese speakers’, 5) individual perceptual preference exerts influence on responses. In conclusion, we suggest: 1) Chinese character learning is progressive and counts on individual differences, 2) the use of association networks would be beneficial in classroom, 3) word association and mind-mapping method can be applied to textbook compilation, instructional design, and test development.