NEURAL COMPUTATIONS IN VISUAL CORTEX

猕猴V1区视觉形式的皮层处理将是 研究了信息是如何在丘脑输出中表现出来的， 在皮层处理的初始阶段被重新组织。 等偶极纹理和相关的新的视觉刺激将被利用， 分离皮质内特征提取，并分离该过程 空间滤波。 神经反应将通过现场监测 潜力、多单位活动和单单位活动。 神经 活动将以一种不仅能检测变化 在平均放电率锁相的刺激，但也更微妙 与刺激相关的变化，如相关结构的改变。 第一个子项目是分析形式的分布 在VI中处理，特别强调其与 柱状组织、层状组织和细胞色素氧化酶 斑点。 第二个子项目是详细分析大脑皮层 与表单处理相关的计算。 作为一个起点， 调查将围绕一个工作模式进行， 人类诱发电位和心理物理学研究，其中输出 的局部非线性子单元以协作方式组合。 这 将在单细胞反应的基础上改进或修订模型。 在这一模式框架内，空间、时间、 并且将确定子单元的色彩特性。 视觉信息处理被认为是一个模型的感觉 一般的信息处理。 分析了形态的形成过程， 以及它与颜色和运动处理的关系，使我们能够 在上下文中研究特征提取和跨模态交互， 可以提出哪些特定假设并且可以测试哪些模型。

The cortical processing of visual form in area V1 of the macaque will be investigated to learn how information represented in the thalamic output is re-organized during the initial stages of cortical processing. Isodipole textures and related novel visual stimuli will be exploited to isolate intracortical feature extraction, and to separate this process from spatial filtering. Neural responses will be monitored via field potentials, multi-unit activity, and single-unit activity. Neural activlty will be analyzed in a manner which will detect not only changes in mean firing rate phase-locked to the stimulus, but also more subtle stimulus-linked changes, such as alterations in correlation structure. The first subproject is an analysis of the distribution of form processing in VI, with particular emphasis on its relationship to the columnar organization, laminar organization, and cytochrome oxidase blobs. The second subproject is a detailed analysis of cortical computations related to form processing. As a starting point, this investigation will be organized around a working model developed from human evoked-potential and psychophysical studies, in which the outputs of local nonlinear subunits are combined in a cooperative fashion. This model will be refined or revised on the basis of single-cell responses. Within this model framework, the relationship of the spatial, temporal, and chromatic properties of the subunits will be determined. Visual information processing is considered to be a model for sensory information processing in general. Analysis of the processlng of form, and its relationship to the processing of color and motion, allows us to study feature extraction and cross-modality interactions in a context in which specific hypotheses may be proposed and models may be tested.