Role of Statistically Induced Changes in Sensory Cortex in Perceptual Development

描述(由申请人提供)：人类发展的一个标志是能够对由于经验而发生的感知进行适应性改变。影响感知的内部表征必须由自然环境中存在的多个层次的统计信息构成。为了说明什么是统计 根据信息，在听到叫声后，看到狗的可能性比看到猫的可能性高(即p(VDog|Aog)&gt；p(VCAT|Aog))。虽然人们普遍认为，通过统计学习获得的信息是基于经验的知觉发展的重要组成部分，但关于将经验转化为持久知觉变化的神经生物学和行为机制，或者这些机制中的变化如何导致非典型或延迟的发展，人们知之甚少。这一建议解决了我们对基于经验的知觉发展机制的三个主要认识空白：1)统计信息对婴儿期感觉皮层神经活动的影响；2)这些神经反应的偏差是否可以预测高危人群的发育迟缓；3)统计诱导的神经活动变化与感知行为变化之间的关系。目的1表征统计信息对出生后发育早期神经活动的影响。通常 在发育中，6个月大的婴儿将接受近红外光谱(NIRS)记录，以揭示在体验跨模式统计信息期间和之后听觉和视觉感觉皮质中神经活动的变化。我们假设，根据统计经验，婴儿的感觉皮层将表现出显著的活动变化(即，在意外的视觉遗漏期间活动增加)。目的2检验统计诱导的神经活动的偏差 在高危人群中。如果统计诱导的神经反应支持许多领域的典型发育，那么对统计信息的异常神经反应将预测出生后较晚的不良行为结果，并建立有助于改善异常发育的生物标记物。使用来自Aim 1的数据作为对照组，我们将描述由于极早产(28周)而导致知觉和认知发育延迟的高危婴儿的神经反应。目的3确定统计诱发的神经活动与知觉变化之间的关系。视觉刺激后的视皮层活动产生视觉感知，但在刺激意外遗漏期间，视觉皮质活动的行为后果尚不清楚，但对于理解统计诱导的神经变化如何支持知觉发展是必不可少的。在特定假设的指导下，我们将把对意外遗漏的神经反应与知觉任务中的行为联系起来。

DESCRIPTION (provided by applicant): A hallmark of human development is the ability to make adaptive changes in perception that occurred as a result of experience. The internal representations that influence perception must be constructed from multiple levels of statistical information that are present in the natural environment. To illustrate what is meant by statistical information, after hearing a bark, there is a higher probability of seeing a dog than seeing a cat (i.e. p(Vdog | Adog) > p(Vcat | Adog). While it is widely believed that information acquired via statistical learning is an essential component of experience-based perceptual development, little is known about the neurobiological and behavioral mechanisms that translate experience into lasting perceptual change or how variations in these mechanisms can lead to atypical or delayed development. This proposal addresses three major gaps in our knowledge of the mechanisms of experience-based perceptual development: 1) the effect of statistical information on neural activity in sensory cortices in infancy; 2) whether deviations in these neural responses predict developmental delays in a high-risk population; 3) the relationship between statistically-induced changes in neural activity and behavioral changes in perception. Aim 1 Characterize the effects of statistical information on neural activity early in postnatal development. Typically developing, 6-month-olds will undergo near infrared spectroscopy (NIRS) recording to reveal changes in neural activity in auditory and visual sensory cortices during and after experience with cross-modal statistical information. We hypothesize that infant sensory cortex will exhibit significant changes in activity (i.e., increased activity during an unexpected visual omission) as result of statistical experience. Aim 2 Examine deviations in statistically-induced neural activity in at-risk populations. If statistically-induced neural responses support typical development in numerous domains, abnormal neural responses to statistical information will predict poor behavioral outcomes later in postnatal life and establish biomarkers that could assist in ameliorating abnormal development. Using data from Aim 1 as a comparison group, we will characterize neural responses in infants at high risk for delays in perceptual and cognitive develop- ment as a result of being born extremely preterm (< 28 weeks). Aim 3 Determine the relationship between statistically-induced neural activity and perceptual change. Visual cortex activity after a visual stimulus produces a visual percept, but the behavioral consequences of visual cortex activity during the unexpected omission of a stimulus are unknown yet are essential to understanding how statistically-induced neural changes support perceptual development. Guided by specific hypotheses, we will correlate neural responses to unexpected omissions to behavior on perceptual tasks.