本研究目的在探討自閉症兒童詞彙發展趨勢及被最多兒童習得的早期詞彙類型及內容。研究者使用幼兒版的「華語嬰幼兒溝通發展量表」，請領有醫院診斷證明的學齡前自閉症兒童家長勾選兒童的表達性詞彙。本研究資料及分析來自兩個小的研究：研究一採縱貫性研究設計，探討30 位自閉症兒童前後六個月的詞彙發展；研究二採橫斷性設計，將51 位詞彙總量在200 以內的自閉症兒童，依詞彙總數分為50 以內、100 以內及200 以內三組，並分析各組的詞彙類型與內容。本研究有以下發現：一、在詞彙總量50 以內時，占比最高的四大類詞彙分別是具體名詞、動詞、人稱及遊戲與日常活動等，顯示此四類詞彙是最早習得的詞彙類型。二、詞彙總量在50 以內、100 以內或200 以內之各組，廣義名詞的總數皆比動詞多，且兩者有顯著差異，顯示名詞優先是此三個詞彙區間皆有的發展現象。三、整體而言，四大類詞彙發展速度呈現名詞發展速度最快，其次是動詞，然後是修飾詞，而功能詞是最慢發展的詞彙類別，此四大類別詞彙習得速度快慢情形與文獻中一般兒童類似。四、自閉症最早習得的詞彙與台灣一般兒童的最初詞彙有部分雷同。五、自閉症兒童的具體名詞和人稱詞彙在初期詞彙的習得比例上可能與一般兒童有差異。上述研究結果顯示，自閉症兒童詞彙學習的趨勢及內容與一般兒童有相似處，但也顯現出一些獨特的學習現象，值得後續探討，並做為實務介入參考。

Purpose: The aim of this study was to explore the expressive vocabulary development of children with autism. Four research questions were investigated: (1) What are the types of words first acquired by children with autism? (2) Is “noun bias” still observed in the vocabulary development of these children? (3) Is there a significant rate difference in the acquisition of nouns, verbs, modifiers, and function words? (4) What are the first words acquired by children with autism? Methods: The toddler version of the Taiwan Mandarin-Chinese adaptation of the Communicative Development Inventories (MCDI-T) was employed as the main assessment tool. The research analysis was based on data from two small studies. Study 1 involved 30 children with autism; the children were tested twice, with a 6-month interval, and the average age of the 30 children was 45.97 months at the initial test. In study 2, the expressive vocabulary of 51 children with autism was analyzed using the MCDI-T. The average age of the 51 children was 52.08 months. The vocabulary on the MCDI-T was regrouped to investigate the developmental trend for each type of word. On the basis of previous research, the vocabulary was regrouped in various ways, such as nominals (including concrete nouns, people’s names, natural phenomena, and outdoor places) and nonnominals (including verbs, modifiers, games and routines, time words, and function words). Of the 51 children, 20 had a total vocabulary of between 1 and 50 words; 9, between 51 and 100 words; and 22, between 101 and 200 words. Results/Findings: The following major findings were obtained. (1) When the total vocabulary was 50 words or fewer, the main four categories of words were concrete nouns, verbs, people’s names, and games and routines. (2) The number of general nouns was significantly higher than the number of verbs when the vocabulary size was between 1-200, indicating that noun bias also occurs in children with autism. (3) Nouns were discovered to be acquired more quickly than verbs, verbs more quickly than modifiers, and modifiers more quickly than function words. The order of acquisition of different categories of words were similar to those of normal children. (4) The actual words first acquired in each semantic class were extremely similar to those for children without autism, as demonstrated in other research conducted in Taiwan. (5) The proportions of concrete nouns and people’s names in the total vocabulary may differ from those of children without autism and are worthy of further investigation in the future. Conclusions/Implications: The vocabulary development pattern of children with autism had similarities and differences with that of normal children. The findings can be applied to assessment and intervention practice.