Learning priors for Bayesian perception in normal and atypical development

为了在环境中导航并与他人互动，人们必须不断根据不确定的信息做出决策。人类已经被证明可以将当前的感官证据与基于先前经验的期望相结合，其方式与贝叶斯决策理论一致。本研究的主要目的是：（1）研究正常发育儿童和成人的贝叶斯先验学习的时间过程和性质;（2）验证自闭症谱系障碍（ASD）患者的非典型贝叶斯先验学习是常见的感知异常的基础这一假设

To navigate the environment and interact with others, people must constantly form decisions based on uncertain information. Humans have been shown to integrate current sensory evidence with expectations based on prior experience in a manner consistent with Bayesian decision theory. However, fundamental questions remain regarding how priors are learned, and how learning progresses in typical and atypical development.This project has two provisional aims:(1) To characterise the time-course and properties of Bayesian prior learning in normally developing children and adults(2) To test the hypothesis that atypical Bayesian prior learning underlies commonly reported perceptual anomalies in autism spectrum disorder (ASD)