DOI:10.6129/CJP.202203_64(1).0002
中華心理學刊 民 111，64 卷，1 期，33-56
Chinese Journal of Psychology 2022, Vol.64, No.1, 33-56
熊恩筠（中國醫藥大學生物醫學研究所；林口長庚紀念醫院藥劑部）；簡惠玲（中國醫藥大學生物醫學研究所；中國醫藥大學神經科學與認知科學研究所 ）；高千惠（澳洲雪梨大學沃爾考克醫學研究所）；彭馨蕾（中國醫藥大學生物醫學影像暨放射科學學系）
摘要
能夠正確的辨認出臉孔和物體是人們重要的生存能力。先前研究探討自閉症類群障礙症(autism spectrum disorders, ASD)族群之臉孔辨識與物體辨識的意見分歧，有學者認為 ASD 族群的臉孔辨識較差、物體辨識與常人 無異，支持整體處理能力較弱的觀點。另一派學者則認為 ASD 族群的臉孔辨識正常、物體辨識優於一般人，支持 ASD 族群特徵處理能力較強的觀點。然而，上述研究有方法上的局限性，如未能控制臉孔作業和物體作業本身的 困難度，使得 ASD 族群之表現反映的是整體式處理的缺損，或特徵式處理的增進仍有待釐清。有鑑於此，本研究 採用變形典派系統性操弄刺激，將臉孔與物體區辨作業困難度調整到一致，以電腦化的變形臉孔與變形物體區辨作 業探討 ASD 族群和一般成人的正確率表現;並輔以中文版成人自閉特質量表(AQ)問卷初探自閉症特質與區辨作 業表現的關聯性。本研究招募了 24 位 ASD 成人以及 24 位年齡配對的一般成人，結果顯示:在變形臉孔區辨作業 中，ASD 組的正確率略低但未達顯著，曲線擬合推估的區辨閾值顯著高於一般成人(ASD 組:33.12%;一般成人 組:25.87%)。在變形物體區辨作業中，ASD 組的正確率與一般成人相同，而擬合的區辨閾值也一致(ASD 組: 21.67%;一般成人組:23.12%)。最後，趨勢分析結果顯示無論是臉孔或是物體作業，ASD 組皆僅有一次項成分 達顯著水準，表示他們的作答正確率變形程度增加而呈線性成長;然而，一般成人除了一次項成分顯著外，當變形 程度夠大時正確率亦趨於飽和。綜合以上結果，本研究顯示 ASD 成人區辨變形臉孔需要更高的閾值，顯示對於臉 孔刺激的整體處理能力有缺損;此外，臉孔和物體作業的正確率隨著變形程度增加呈線性遞增趨勢，表示 ASD 成 人代償性地以局部特徵式處理來辨識物體和臉孔。
關鍵詞：自閉症、臉孔知覺、物體知覺、變形典派
Comparing Face and Object Discrimination in Neurotypical Adults and Adults with Autism: A Morphing Paradigm
En-Yun Hsiung（Graduate Institute of Biomedical Sciences, China Medical University；Department of Pharmacy, Chang Gung Memorial Hospital）；Sarina Hui-Lin Chien（Graduate Institute of Biomedical Sciences, China Medical University；Graduate Institute of Neural & Cognitive Sciences, China Medical University）；Chien-Hui Kao（Woolcock Institute of Medical Research, University of Sydney）；Shin-Lei Peng（Department of Biomedical Imaging and Radiological Science, China Medical University）
Abstract
Recognizing faces and objects are essential perceptual skills for human survival. Previous studies exploring face and object perception in individuals with autism spectrum disorder (ASD) reported inconclusive results. Some studies suggested that adults with ASD exhibited impaired face processing but spared object perception; in contrast, other studies showed unimpaired face processing and enhanced object perception, suggesting superior featural- based processing in ASD. However, most studies did not control the task difficulties of face and object discrimination. Hence, we adopted the morphing technique to control physical similarities among the stimuli with equated task difficulties. Using computerized morphing face/object discrimination tasks, we investigated whether adults with ASD are impaired in holistic processing or enhanced in featural processing. We also explored the correlations between individual Autism-spectrum Quotient (AQ) scores and their task performances. Twenty-four adults with ASD and 24 NT adults received the morphing face/object discrimination task, and their accuracies were measured. The results showed that: For the morphing face discrimination task, the mean accuracies were not significantly different between the two groups; however, the estimated discrimination threshold of the ASD group (33.12%) was significantly higher than the NT group (25.87%). For the morphing object discrimination task, both groups had the same mean accuracies and estimated discrimination thresholds (ASD: 21.67%; NT: 23.12%). Lastly, the trend analyses revealed a significant linear component in the ASD group for both face and object tasks, while the NT group showed linear as well as quadratic components. In sum, the present study demonstrated that, compared to neurotypical adults, adults with ASD exhibited a larger discrimination threshold for morphing face, indicating impaired holistic processing. Additionally, the linear trend suggests that adults with ASD compensatorily relied on featural-based processing when differentiating among morphing images of faces and objects.
Keywords: autism spectrum disorder, face perception, object perception, morphing paradigm
