考量語言能力下之情感曲風音樂與學齡高功能自閉症類群障礙兒童心情臉孔辨識的關聯性

研究背景：心情臉孔辨識在社交溝通發展上扮演重要的角色，自閉症類群障礙症（Autism Spectrum Disorder，簡稱ASD）兒童指認臉孔表情能力落後一般發展兒童。先前研究發現ASD兒童辨認基本情緒的臉孔表情有困難，然而ASD成人聆聽情感音樂時的腦部活化位置及生理激發強度和同年齡參與者沒有顯著差異，而且在情感音樂背景下進行心情臉孔辨識的表現會提升，故音樂能否做為促進ASD兒童心情心情臉孔辨識的媒介？研究問題：本研究問題有三，（一）ASD兒童臉孔的辨識表現與一般發展兒童的是否有差異，是否隨年紀不同？（二）研究問題2：情感音樂對於ASD兒童在中性、快樂、生氣心情臉孔辨識表現的效果是否與一般組不同，是否隨年齡不同？（三）自閉特質與心情臉孔辨識能力與的關聯性？研究方法：本研究招募70位自閉症組（年紀小組：Mage (SD)=7.18 (0.42)y；年紀大組（Mage (SD)=9.18 (1.08)y），以及70位一般組6-12歲兒童（年紀小組：Mage (SD)=7.64 (0.21)y; 年紀大組：Mage (SD)=9.52 (1.42)y），家長填寫背景資料表及自閉症光譜量表兒童版（The Autism Spectrum Quotient： Children’s Version，簡稱AQC），分別蒐集人口學變項及兒童自閉特質嚴重程度（包含社會技巧、注意力轉換、對細節的注意力、溝通分數、想像力分量表），兒童以「畢保德圖畫詞彙測驗 ( PPVT-R )」量測語文智力，以「自編心情臉孔辨識電腦化作業」量測兒童心情臉孔辨認能力，該電腦作業於無音樂、快樂音樂、生氣音樂情境中分別評分30張臉孔表情照片（快樂、中性、生氣各10張），音樂情境及刺激順序採受試者間對抗平衡，採雙向度（快樂－生氣）7點量表，紀錄正確率、反應時間及評分強度。結果：（1）以組別（ASD、一般組）、年齡分類（年紀小、年紀大）、表情（中性、快樂、生氣）為自變項，以正確率、反應時間、評分強度為依變項進行三因子混合設計ANOVA，發現ASD組的心情臉孔辨識正確率顯著低於一般組（特別是年紀大ASD辨識中性表情），反應時間及評分強度則否。另加入語文智力作為控制變項進行ANCOVA，發現ASD組兒童辨識的正確率、反應時間、評分強度皆與一般組相當（2）除上述自變項外，再加入音樂（無、快樂、生氣音樂）作為自變項，以正確率、反應時間、評分強度為依變項，分別於中性、快樂、生氣表情下進行三因子混合設計ANOVA，發現情感音樂在中性表情的辨識中可以促進ASD及一般組辨認正確率（特別是在年紀大ASD組），其他表情及指標則否。另加入語文智力作為控制變項進行ANCOVA，發現音樂促進中性表情的正確率之效果仍在。（3）以AQC總分、各分量表分數及無音樂情境之心情臉孔辨識各指標作為變項進行相關分析，發現AQ-C總分、溝通困難、注意力轉換困難分數愈高，辨識正確率愈低，其他指標則沒有關聯。總結：ASD兒童臉孔的中性表情辨識表現落後一般發展兒童，且與自閉特質之嚴重程度有關聯，然而，情感音樂可以作為提升模糊表情的辨識正確率並縮短反應時間，但在評分強度則否。另控制語文能力後，ASD兒童及一般組的辨識表現差異不明顯，且音樂促進中性表情辨識正確率之效果仍在。上述結果意涵語文能力在這臉孔表情表現及音樂情緒建構歷程中皆扮演的重要角色，臨床工作者在使用情感音樂刺激作為情緒概念相關能力之介入媒介時可以將此納入擬定治療計畫之重要因素。

Background/Purpose: Facial expression recognition is crucial for social communication; however, children with Autism Spectrum Disorder (ASD) often struggle in this area compared with typically developing (TD) children. Prior research has indicated difficulties in identifying basic emotions from facial expressions in individuals with ASD. However, studies have shown that the neural activation patterns and physiological arousal levels in adults with ASD, when listening to affective music, resemble those in individuals with TD. Additionally, the presence of affective music enhances performance in facial emotion recognition, raising the question of whether music can facilitate facial emotion recognition in children with ASD. This study aimed to investigate: (1) differences in facial recognition performance between children with ASD and TD children, considering age variations; (2) effects of affective music on facial emotion recognition performance for neutral, happy, and angry expressions in children with ASD and TD children, considering age differences; (3) the association between autistic symptoms and facial emotion recognition ability.Method: The study participants consisted of 70 children with ASD (Young group: Mage (SD) = 7.18 (0.42) years; Old group: Mage (SD) = 9.18 (1.08) years) and 70 TD children (Young group: Mage (SD) = 7.64 (0.21) years; Old group: Mage (SD) = 9.52 (1.42) years). We used the Autism Spectrum Quotient - Children's Version (AQ-C) to measure the severity of autistic symptoms in children (including total AQ-C, social skills, attention switching, attention to detail, communication, and imagination scores). Additionally, participants’ language abilities were assessed using the Peabody Picture Vocabulary Test-Revised (PPVT-R) and facial expression recognition abilities were assessed using a Facial Expression Recognition Task. In this task, the children identified facial expressions in 30 photos (10 each of happy, neutral, and angry) under the conditions of no music, happy music, and angry music. A two-dimensional (happy–angry) 7-point scale was used to record the correct rate, reaction time, and intensity. The order of the music conditions and stimuli was counterbalanced across participants. Results: (1) The ASD group exhibited significantly lower accuracy rates in facial emotion recognition than the TD group, particularly in identifying neutral expressions among older ASD participants. However, after controlling for language ability, there were no significant differences in accuracy, reaction times, or rating intensity between children with ASD and TD children. (2) Affective music enhanced the accuracy of neutral expression recognition in both the ASD and TD groups, especially among older ASD participants. Even after controlling for language ability, the effect of the music remained significant. (3) Correlation analysis revealed that higher AQ-C total scores, communication, and attention switching were associated with lower facial emotion recognition accuracy.Conclusion: Children with ASD showed lower performance in recognizing facial expressions, which is associated with autistic symptom severity, compared to TD children. Affective music improved the accuracy and reaction times for recognizing ambiguous facial expressions, but not intensity ratings. After controlling for language ability, the differences in recognition performance between children with ASD and typically developing children were not significant, and the facilitative effect of music on the accuracy of recognizing neutral expressions remained. These results emphasize the role of language ability in facial expression recognition and emotional music processing.