孤独症谱系障碍儿童非典型面部表情特征及其在早期筛查中的应用  

Atypical facial expression characteristics in children with autism spectrum disorder and their application in early screening

作　　者：杨平 方润秋 翁旭初 YANG Ping;FANG Runqiu;WENG Xuchu(School of Psychology,Guizhou Normal University,Guiyang 550025,China;Institute of Brain Research and Rehabilitation,South China Normal University,Guangzhou 510898,China)

机构地区：[1]贵州师范大学心理学院,贵阳550025 [2]华南师范大学脑科学与康复医学研究院,广州510898

出　　处：《心理科学进展》2025年第4期588-597,共10页Advances in Psychological Science

基　　金：国家社会科学基金重大项目(20&ZD296);广东省重点领域研发计划(2019B030335001);国家自然科学基金项目(32260211)资助。

摘　　要：孤独症谱系障碍(Autism Spectrum Disorder,ASD)儿童表现出特有的非典型面部表情特征,包括中性表情居多、积极表情减少、社交微笑频率低以及自发面部表情模仿能力不足。这些特征从幼儿期到儿童期表现稳定,已成为ASD风险评估的重要标志。然而,传统研究方法(如人工评估和面部肌电图)在分析ASD儿童面部表情时存在主观性强、耗时长且难以推广等局限性。近年来,人工智能的迅速发展使基于计算机视觉和深度学习的自动化表情识别技术得以应用,不仅显著提高了分析效率,还降低了人为评估的主观误差,为基于非典型面部表情特征的大规模ASD早期筛查提供了强有力的支持。未来研究可进一步优化识别模型,通过设计更接近自然情境的诱发范式,深入探索ASD儿童多样化的面部表情特征,同时提升模型的准确性和灵敏度,以推动ASD早期筛查和干预的发展。Children with Autism Spectrum Disorder(ASD)exhibit atypical facial emotional expressions,such as a prevalence of neutral expressions,reduced positive expressions,lower frequency of social smiles,and limited spontaneous facial mimicry.These characteristics remain stable from infancy to childhood,making them important markers for ASD risk assessment.However,traditional assessment methods,such as manual observation and facial electromyography,have limitations in analyzing facial emotional expressions in ASD children due to high subjectivity,time consumption,and difficulties in large-scale application.In recent years,advancements in artificial intelligence have facilitated the application of automated facial expression recognition technology based on computer vision and deep learning,significantly enhancing efficiency and reducing subjective bias,thereby providing strong support for large-scale ASD early screening based on atypical facial expressions.Future research could further optimize recognition models by designing more naturalistic induction paradigms to explore the diverse facial emotional expressions of ASD children,thus improving the accuracy and sensitivity of automated models and advancing ASD early screening and intervention efforts.

关 键 词：孤独症谱系障碍 面部表情 计算机自动识别 

分 类 号：R395[哲学宗教—心理学]