自閉症兒童象徵遊戲的產出能力：三重表徵的觀點

Production of Symbolic Play in Children with Autism: A Case of Triune Representation

本研究主要以認知（控制參與者心理年齡）角度，探究自閉症兒童象徵遊戲的產出能力，並以社會角度探究該兒童在遊戲過程中社會注視行為。依循三重表徵概念（Tomasello, Striano, & Rochat, 1999），參考Striano、Tomasello及Rochat（2001）測量象徵遊戲產出能力的研究方法，比較自閉症（平均生理年齡69個月，平均心理年齡56個月）與心理年齡相當的發展遲緩兒童、一般發展兒童之象徵遊戲能力。本研究分為兩個實驗：實驗一為結構化遊戲研究，分為基準期及示範期，來觀察三組兒童在不同時期的象徵遊戲產出能力；實驗二為社會注視研究，設計功能物組及象徵物組玩具，來瞭解三組兒童在玩不同玩具時的社會注視型態。實驗一結果發現：在基準期的自發性象徵遊戲的產出表現，三組兒童的表現已出現明顯差異，自閉症兒童的表現落後於兩組對照組兒童。經過實驗者示範後，自閉症兒童在模仿及出現新奇象徵性遊戲行為的發生率，亦低於兩組對照組。實驗二結果發現：三組兒童進行功能遊戲動作期間及之後，眼光注視大人的次數與兩組對照組兒童相當，但在完成象徵動作後，自閉症兒童眼光注視的次數顯著的少於兩組對照組兒童。總結來說，本研究發現自閉症兒童，他們不僅仍有明顯的象徵遊戲產出障礙，在社會互動中，即使能表現出象徵遊戲能力，也較少以眼光注視引發社交溝通。本文最後從理論的角度來回應這項發現，並提出後續研究的方向。

Background: There is considerable empirical evidence for deficits in pretend play in children with autism compared with children with developmental delay (DD) and typically development (TD). We referred Tomasello and his colleagues's suggestion (Tomasello, Striano, & Rochat, 1999) and used the "triune representation" point of view to measure the child's symbolic play behaviors. Objectives: The purpose of this study was to (1) examine the production of symbolic acts in structural play condtion; (2) explore the patterns of social looking of symbolic acts in functional and symbolic play conditions. Methods: Based on the perspective of "triune representation," a modified paradigm from Striano, Tomasello, and Rochat (2001) was tested to 17 children with autism (mean chronological age [CA] = 69 months, mean mental age [MA] = 56 months), 17 MA-matched children with DD and 19 children with TD. Two experiments were arranged. In experiment 1, examiner (E) introduced the doll house divided into four rooms with appropriate prop sets and used the three pairs of actors, including replica dolls (a toy man and toy woman), instrumental objects (a pencil and a scissors), and natural objects (a rock and a stick) to play. Experiment 1 consisted two phases: baseline and testing phases. In baseline phase, examiner introduced the doll house and encouraged the child to use each of the three actors set in turn. Each pair can play in the doll house for two minutes. In testing phase, examiner modeled with the action prompt or verbal direction first and then encouraged the child to play each pairs of actors for two minutes. We measured child's symbolic acts spontaneously in the Action and Language conditions. In experiment 2, children were given the "functional sets" (i.e., doll + bed and pegboard + hammer) and "symbolic sets" (i.e., doll + block and pegboard + brush) randomly to play in front of examiner. We counted the child's looking behaviors during/after he/she engage in the functional and symbolic acts with object. Results: (1) In experimental 1, there was a significant difference of symbolic acts among the three groups in the baseline phase and the children with autism displayed fewer than those children with two control groups. After examiner's modeling, children with autism imitated less symbolic acts and displayed lower novel symbolic acts comparing with the two controls. (2) In experimental 2, children in the three groups revealed no significant difference on the frequencies of looks to examiner during/after performing functional acts. However, the children with autism displayed less looks to examiner than children with DD and TD after they displayed symbolic acts. Conclusions: Using the perspective of "triune representation" to measure the spontaneous symbolic play acts, children with autism not only showed impairments on symbolic production but also displayed fewer social looking for social communication. The theoretical and clinical implications were discussed.