孤独症谱系障碍者的视觉感知:基于贝叶斯和预测编码视角OA北大核心CHSSCDCSSCICSTPCD

Bayesian and predictive coding theories have been instrumental in elucidating sensory processing aberrations observed in individuals with Autism Spectrum Disorder(ASD).However,these theories have sparked considerable debate within the scientific community.In this paper,we focus on non-social visual information,presenting a comprehensive examination of theoretical nuances and consolidating empirical evidence across three key dimensions:Bayesian inference,predictive coding processes,and predictive coding precision.Generally,hypo-priors and sharper likelihood hypothesis based on Bayesian inference merely provide descriptive insights into visual processing abnormalities associated with ASD.While perspectives emphasizing the predictive coding process enhance the specificity of visual processing aberrations,they fall short of offering a fully explanatory framework.On the other hand,hypotheses centered on predictive coding precision provide theoretical foundations,yet require further refinement of the theoretical details and their validity necessitates testing through refined empirical studies.Future research should generalize the specificity of predictive processing in ASD,test the theoretical content from the perspective of subjective experience of ASD,and examine the changes of predictive functioning in the growth of ASD through a developmental perspective.