Neural mechanisms of color

描述（由申请人提供）：神经活动、感知和认知之间的联系尚不清楚。这一建议将颜色作为一种模型系统来填补这些知识空白。颜色是视觉体验的一个基本特征，关于眼睛发出的锥体信号是如何编码并传递到大脑皮层的，我们已经知道了很多。但令人惊讶的是，人们对解码这些信号、产生感知颜色并指导感知决策的机制知之甚少。人们提出了两种相互竞争的解码方案：一种间隔码，它需要一群具有鲜明色彩调谐的细胞，它们一起包含所有色彩空间，再加上赢家通吃的规则；还有一个种群代码，它至少需要两组色彩调节的神经元，再加上加权平均规则。目前还不清楚大脑皮层的哪一组神经元参与其中。一个线索来自恒河猴的下颞叶皮层（IT）损伤，它会导致严重的色盲，类似于人类某些脑中风所伴随的色盲。IT是涉及对象编码的许多方面的组织的广阔区域，IT的功能组织很少被理解。没有这些信息，几乎不可能知道哪些神经元最有可能参与颜色处理。可能的组织方案包括由独特的专门领域组成的模块化模式；分布式处理模型；或者是一个混合模型，由一系列分层阶段组成，每个阶段都包含功能子区域的完整补充。目标1要求在警觉的猴子身上进行一系列功能性磁共振成像（fMRI）实验，以确定IT中颜色编码区域的分布、它们的功能连通性以及与其他功能定义区域的关系

DESCRIPTION (provided by applicant): The links between neural activity, perception and cognition are poorly understood. This proposal advances color as a model system to fill these gaps in knowledge. Color is an essential feature of visual experience, and much is known about how cone signals from the eye are encoded and transmitted to the cortex. But surprisingly little is known about the mechanisms that decode these signals to bring about perceived colors and guide perceptual decisions. Two competing decoding schemes have been proposed: an interval code, which requires a population of cells with sharp chromatic tuning that together encompass all color space, coupled with a winner- take-all rule; and a population code, which needs at minimum two groups of color-tuned neurons, coupled with a weighted-average rule. It is unclear which groups of neurons within the cerebral cortex are involved. One hint comes from lesions of inferior temporal cortex (IT) in rhesus monkeys, which cause profound color blindness similar to the achromatopsia that accompanies certain cerebral strokes in humans. IT is an expansive region of tissue implicated in many aspects of object coding, and the functional organization of IT is poorly understood. Without this information, it is almost impossible to know which neurons are the most likely to be contributing to color processing. Possible organizational schemes include a modular model comprising uniquely specialized areas; a distributed-processing model; or a hybrid model, consisting of a series of hierarchical stages, each comprising a full complement of functional subregions. Aim 1 calls for a battery of functional magnetic resonance imaging (fMRI) experiments in alert monkey that will determine the distribution of color-coding regions in IT, their functional connectivity and relationship to other functionally defined regions to test which model accounts for the organization of IT. Aim 2 outlines fMRI-guided microelectrode recordings of IT color regions paired with microstimulation while monkeys perform color tasks, to test the causal link between neural activity and perceived color, and to determine which of the two decoding schemes, interval or population, is implemented in IT. The research will uncover principles by which perception and cognition emerge from the activity of neural circuits. This information is required to understand the etiology, diagnosis, and treatment of mental illnesses and strokes that impair cognition and perception. Moreover, the work will establish the relationship of higher-order areas between humans and monkeys, which is necessary in order to use monkeys as models of human vision and disease.