Function of high-level visual cortex in awake infants.

项目摘要 功能磁共振成像（fMRI）揭示了功能组织的两个主要原则 腹侧颞叶皮层（VTC）的研究。首先，职业训练局的某些地区选择性地回应特定的 一类刺激，如梭状面区（FFA），它对面部的反应比任何其他刺激都要强烈 类别.组织的第二个原则是，不同类别的刺激具有系统性的区别。 整个职业训练局的反应模式。职训局职能组织的这两个主要方面， 成年人在发展？我的研究计划的目标是发现这些组织原则是否 存在于婴儿大脑中（目标1和2），并设计计算模型来测试不同的皮层理论。 发展（目标3）。 本论文的研究工作将为完善大脑皮层的理论提供宝贵的数据。 发展除了提高清醒婴儿fMRI数据质量的创新外，Aim 1还提供了 第一个证据表明，像成年人一样，婴儿在FFA中有面部选择性反应，在FFA中有场景选择性反应。 海马旁位置区（PPA）和纹状体外体区（EBA）的身体选择性反应。 目标2直接跟进这一点，询问婴儿是否有系统的不同模式的反应，在较高的- 水平视觉区域与成人大脑中的视觉区域相似。本提案的F99阶段将提供 培训，以优化机器学习（ML）技术，可以承受婴儿功能磁共振成像的独特挑战 数据-特别是不平衡和缺失的数据。F99阶段将在麻省理工学院进行， 环境与ML，认知神经科学和计算神经科学领域的领导者接触。 最后，本提案的目标3是建立旨在测试当前皮层神经元理论的计算模型。 发展本建议书的K 00阶段将提供人工 神经网络（ANN）模型以及使用婴儿fMRI测试ANN模型的最佳方法的培训 数据K 00阶段的研究将提供一个新的导师经验， 计算神经科学，并将在一个拥有蓬勃发展的智力环境的机构中进行， 访问MRI扫描仪和建立各种ANN模型所需的计算资源。 总之，拟议研究的目标是确定是否功能性组织的原则， 成人的大脑存在于婴儿中。从这个提议中得到的见解将推进和完善大脑皮层的理论， 发展，并有潜力适用于其他领域，如听力和语言。此外，由 结合我在清醒婴儿功能磁共振成像方面的博士前培训和我在计算方面的博士后培训， 建模，拟议的研究将使我成为一个独立的调查员和领导者在该领域 计算发展神经科学。

Project Summary Functional magnetic resonance imagining (fMRI) has revealed two major principles of the functional organization of the ventral temporal cortex (VTC) in human adults. First, some regions of VTC respond selectively to a specific category of stimuli, such as the fusiform face area (FFA) which responds more to faces than to any other stimulus category. The second principle of organization is that different categories of stimuli have systematically distinct patterns of response across the entire VTC. How do these two key aspects of VTC functional organization in adults arise in development? The goal of my research program is to discover if these organizing principles are present in the infant brain (Aims 1 and 2) and design computational models to test different theories of cortical development (Aim 3). The dissertation work in this proposal will provide invaluable data toward the goal of refining theories of cortical development. In addition to innovations that enhance the quality of awake infant fMRI data, Aim 1 provides the first evidence that like adults, infants have face-selective responses in the FFA, scene-selective responses in the parahippocampal place area (PPA), and body-selective responses in the extrastriate body area (EBA). Aim 2 directly follows this up by asking if infants have systematically distinct patterns of response across higher- level visual areas that are similar to those found in the adult brain. The F99 phase of this proposal will provide training to optimize machine learning (ML) techniques that can withstand the unique challenges of infant fMRI data – specifically unbalanced and missing data. The F99 phase will be conducted at MIT, an intellectual environment with access to leaders in the fields of ML, cognitive neuroscience, and computational neuroscience. Finally, Aim 3 of this proposal is to build computational models designed to test current theories of cortical development. The K00 phase of this proposal will provide training on the design and implementation of artificial neural networks (ANNs) models as well as training on the best methods to test ANN models using infant fMRI data. Research for the K00 phase will provide a new mentorship experience with an established investigator in computational neuroscience and will take place at an institution with a thriving intellectual environment that has access to an MRI scanner and the computational resources necessary to build a variety of ANN models. In summary, the objective of the proposed research is to determine if the principles of functional organization in the adult brain are present in infants. Insights from this proposal will advance and refine theories of cortical development and have the potential to be applicable to other domains such audition and language. Further, by combining my predoctoral training in awake infant fMRI with my proposed postdoctoral training in computational modeling, the proposed research will enable me to become an independent investigator and leader in the field of computational developmental neuroscience.