NEURAL BASIS OF MONOCULAR AND BINOCULAR TEXTURE ANALYSIS

单目和双目纹理分析的神经基础

• 批准号: 2741811
• 负责人: IZUMI OHZAWA
• 金额: $ 27.22万
• 依托单位: UNIVERSITY OF CALIFORNIA BERKELEY
• 依托单位国家: 美国
• 财政年份: 1999
• 资助国家: 美国
• 起止时间: 1999-01-01 至 2002-12-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2741811
• 关键词: binocular vision; brightness discrimination; cats; cerebral cortex; computational neuroscience; electrophysiology; form /pattern perception; microelectrodes; neural information processing; neuroanatomy; neurons; neurophysiology; visual cortex; visual depth perception; visual fields; visual perception; visual stimulus

The long-term objective of this research is to determine how the visual part of the cerebral cortex analyzes information that enter through the two eyes, and transforms it into a form useful for perception. The proposed project approaches the problem of visual information processing using electrophysiological recording methods combined with elaborate visual stimulation and analysis techniques. The methods are applied to the problem of how cortical neurons analyze visual texture information and integrate it with that from brightness cues. A hypothesis is proposed that texture-sensitive neurons represent third and fourth stages of hierarchical processing in the visual cortex. To test this hypothesis, a cascaded model of texture cells is constructed based on the notion of "texture receptive field." The following specific aims will be addressed. (1) A model of texture cells will be developed based on a cascaded organization in which the identical neural computation is performed twice serially. Computational studies will examine how well this model predicts known properties of texture perception and cell responses. (2) The cascade model will be tested neurophysiologically by measuring texture receptive fields of visual neurons and comparing them to predictions of the model. Specific neural connections predicted by the model will be sought by analyzing activities of simultaneously recorded groups of neurons. (3) Neural responses to mixed texture and brightness-defined stimuli will be studied to determine how texture neurons integrate these two cues. (4) By stimulating texture neurons simultaneously through the two eyes, the roles that texture information plays in binocular vision and stereopsis will be examined. (5) Developmental time course of neural texture responses will be studied in young animals. These studies will yield a functional schematic diagram of the texture- brightness processing pathway of the visual cortex. Such detailed description of the brain circuitry will be valuable in diagnosing and understanding visual disorders.

这项研究的长期目标是确定视觉效果如何 大脑皮层的一部分分析通过大脑进入的信息 两只眼睛，并将其转化为一种有助于感知的形式。 这 拟议的项目解决视觉信息处理问题 采用电生理记录方法结合精心设计的 视觉刺激和分析技术。 该方法应用于 皮层神经元如何分析视觉纹理信息的问题 并将其与亮度提示相结合。 一个假设是 提出纹理敏感神经元代表第三和第四 视觉皮层的分层处理阶段。 为了测试这个 假设，纹理细胞的级联模型是基于 “纹理感受野”的概念。 具体目标如下 将得到解决。 (1) 将基于级联开发纹理单元模型 执行相同神经计算的组织 连续两次。 计算研究将检验该模型的效果如何 预测纹理感知和细胞反应的已知特性。 (2) 级联模型将通过测量进行神经生理学测试 视觉神经元的纹理感受野并将其与 模型的预测。 由预测的特定神经连接 将通过分析同时记录的活动来寻求模型 神经元组。 (3) 对混合纹理和亮度定义刺激的神经反应 将进行研究以确定纹理神经元如何整合这两个 提示。 (4) 通过两只眼睛同时刺激纹理神经元， 纹理信息在双眼视觉中扮演的角色 将检查立体视觉。 (5) 神经纹理反应的发育时间过程为 在幼年动物身上进行了研究。 这些研究将产生纹理的功能示意图 视觉皮层的亮度处理通路。这么详细 大脑回路的描述对于诊断和治疗很有价值 了解视觉障碍。