INTERNAL SIGNALS IN THE PARIETAL VISUAL PATHWAY

描述(申请人提供)：视觉感知不仅仅依赖于视觉输入。大脑不断地产生内部信号，调节对外部视觉刺激的反应。动态的内部信号允许神经网络对相同的刺激做出不同的反应，使有机体能够适应不断变化的任务需求。内部网络动力学也支持与视觉工作记忆相关的持续神经活动。在这项建议中，我们检查内部信号的影响，继续我们对顶叶皮质的关注。一类重要的内部信号与自我运动有关。我们已经确定了微扫视对顶神经元放电的强大抑制--但仅在需要瞬时运动检测的任务中--这可能是微眼跳的视觉效果所模仿的。目标1检验了这一假设，即抑制可能被任务要求灵活地“把关”。目的1还研究了嘴上丘在抑制眼球跳动中的潜在作用。Aim 2测试了已报道的顶叶皮质中两种不同的神经元信号--“分类”和“知觉决定”--是否实际上是同一件事。大多数知觉决策实验没有测试这些信号是否可以独立于动物报告的形式；我们认为，如果知觉决策信号独立于报告，它们将无法与“分类”信号区分开来。我们将检验嘴唇神经元是否提供一般的视觉分类信号，我们将在同一实验中，在相同的神经元中头对头地比较“知觉决定”和“分类”信号。目的3研究顶叶皮质持续放电的机制。递归网络模型假设，持续记忆延迟期活动是递归电路的自然动态。最近的一个递归网络模型做出了几个强有力但令人惊讶的预测；例如，如果在不同的条件下可以诱发不同水平的持续活动，那么整个神经元群体中活动的幅度应该是彼此和自发活动的比例版本。此外，神经元的持续放电应该有一个共同的偏好顺序，这对皮层来说是一种非常不寻常的组织。目标3使用目标1和目标2产生的实验数据来检验和改进该模型。我们的实验将提供一个关于大脑内部状态如何促进和影响外部视觉世界感知的机械论视角。获取这些过程的基本信息是了解正常和异常大脑处理的关键一步。

DESCRIPTION (provided by applicant): Visual perception depends on more than visual input. The brain constantly generates internal signals that modulate responses to external visual stimuli. Dynamic internal signals allow neural networks to respond differently to the same stimulus, enabling organisms to adapt to changing task demands. Internal network dynamics also support persistent neural activity related to visual working memory. In this proposal, we examine the influence of internal signals, continuing our focus on parietal cortex. An important class of internal signals are related to self-movement. We have identified a powerful suppression of the firing of parietal neurons by microsaccades - but only during a task requiring a transient motion-detection that may be mimicked by a microsaccade's visual effect. Aim 1 examines the hypothesis that the suppression may be flexibly "gated" by task demands. Aim 1 also investigates the potential role of the rostral superior colliculus, an area involved in microsaccade generation, in the suppression. Aim 2 tests whether two different neuronal signals that have been reported in parietal cortex -- "categorization" and "perceptual decision" -- are actually the same thing. Most perceptual decision experiments did not test whether the signals could be independent of the form of the animal's report; we propose that if perceptual decision signals were independent of the report, they would be indistinguishable from "categorical" signals. We will examine whether LIP neurons provide generic visual categorical signals, and we will compare "perceptual decision" and "categorical" signals head-to- head, in the same experiment, in the same neurons. Aim 3 examines the mechanisms underlying persistent firing in parietal cortex. Recurrent network models posit that persistent memory-delay period activity is a natural dynamic of recurrent circuits. A recent recurrent network model makes several robust yet surprising predictions; for example, if different levels of persistent activity can be evoked under different conditions, the amplitude of the activities across the neuronal population should be scaled versions of each other and of the spontaneous activity. In addition, there should be a common order of preference among neurons in their persistent firing, a highly unusual organization for cortex. Aim 3 uses the experimental data generated from Aims 1&2 to test and refine this model. Our experiments will provide a mechanistic perspective on how internal states of the brain facilitate and affect the perception of the external visual world. Gaining basic information on these processes is an essential step for understanding normal and abnormal brain processing.