Rival Networks: Dissecting the Canonical Circuit of Bi-stable Visual Perception

首席调查员(最后、第一、中间)：甘纳帕拉吉纳 摘要/摘要 竞争对手网络：剖析双稳态视觉感知的典型回路 用几种相互排斥的解释观看视觉刺激会导致主观知觉动摇 在两种解释之间。这一过程被称为多稳定感知，并提供了一个极好的 研究知觉如何在大脑中形成和维持的受控模型。多个假设是 提出了多稳态知觉的电路机制：神经元同步性的变化，适应 群体放电率、相互竞争的神经元群体之间的相互抑制、神经噪声和分级推理 横跨皮质区域和皮质下结构的网络。 支持参与这些过程的证据来自计算模型、心理物理 人类和灵长类单位电生理学的研究、功能磁共振成像和经颅磁共振仪。这些方法确立了 这种多稳定感知是一个分布式的过程，涉及低水平和高水平的合作网络。 平整皮质区域。他们还明确指出，大脑中单个单位的活动不能被用作明确的 这是相互竞争的知觉的指示器，人们必须观察神经元回路的水平才能理解这一过程。 然而，直到最近，研究不同大脑皮层区域的群体反应和回路之间的相互作用 在单细胞分辨率下，可能性有限。目前还没有对规范的机械论理解。 在大脑皮层柱和局部子网络水平上构成知觉转换的皮层计算。 此外，没有关于跨皮质区域的电路之间的相互作用的电路级单细胞数据 在知觉竞争中的边界。最近的进展使我们第一次能够在单个细胞分辨率下绘制 双稳态知觉过程中柱状子网络的动力学。我们将在初级感觉区(V1)进行这项工作 是感知变化所必需的。研究电路元件的行为贡献并检验因果关系 我们将使用SLM对特定细胞群体进行光遗传控制(目标1)。然后我们会全程跟进- 大脑半球成像识别属于多稳态知觉功能网络一部分的高阶区域。 我们将对V1、V2和VRL/A/AL中的2/3层子电路进行多区域成像，以掌握远程 初级感觉V1区回路成分的相互作用与知觉编码和翻转编码的进化 在双稳态知觉过程中，大脑皮层上的更高阶区(目标2)。 期望：1)确定小鼠大脑的“双稳态知觉网络”，假定由多个区域组成 参与感知稳定性和反转2)获得电路动力学的第一个全面图像 双稳态时皮质柱内锥体细胞子网络与中间神经元的相互作用 感知，跨V1、LGN输入到V1、V2和视觉运动区VRL/A/AL 3)识别特定的 知觉逆转的过程：放电频率适应、相互抑制相竞争的子网络、同步性和 射击的可变性。

Principal Investigator (Last, First, Middle): Palagina, Ganna Abstract / Summary Rival Networks: Dissecting the Canonical Circuit of Bi-stable Visual Perception Viewing visual stimuli with several mutually exclusive interpretations causes subjective perception to vacillate between the interpretations. This process is known as multi-stable perception and provides an excellent well- controlled model for studying how the percepts are formed and maintained in the brain. Multiple hypotheses are proposed for the circuit mechanisms of multi-stable perception: changes in neuronal synchrony, adaptation of population firing rates, mutual inhibition between rival neuronal populations, neural noise and hierarchical inference across the network of cortical areas and subcortical structures. Supporting evidence for involvement of these processes comes from computational models, psychophysical studies, fMRI and TMS studies in humans and single-unit primate electrophysiology. These approaches established that multi-stable perception is a distributed process involving the cooperative network of both low-level and high- level cortical areas. They also made it clear that the activity of single units in the brain cannot be used as a clear indicator of the rivaling percepts, and that one must look at the level of neuronal circuit to understand the process. However, until recently, studying population responses and interplay between the circuits in different cortical areas at the single-cell resolution was a limited possibility. Currently, there is no mechanistic understanding of canonical cortical computations that underlie perceptual transitions at the level of cortical column and local sub-networks. Moreover, there is no circuit-level single-cell data on the interactions between the circuits across the cortical area borders during perceptual rivalry. Recent advances enable us for the first time to map at single cell resolution the dynamics of columnar sub-networks during bi-stable perception. We will do this in in primary sensory area (V1) that is required for percept alternations. To study the behavioral contribution of circuit components and test for causality we will use optogenetic control of specific cell populations with SLM (aim #1). We will then follow up by whole- hemisphere imaging to identify higher-order areas that are the part of multi-stable perception functional network. We will do multiple-area imaging of layer 2/3 sub-circuits in V1, V2 and VRL/A/ALto get a handle on long-range interactions between circuit components and evolution of percept and reversal encoding in primary sensory area V1 and higher-order areas up in the cortical hierarchy during bi-stable perception (aim #2). Expectations: 1) Identify the “bi-stable perception network” of the mouse brain, composed of areas putatively involved in percept stability and reversal 2) Obtain the first comprehensive picture of the dynamics of circuit interactions between sub-networks of pyramidal cells and interneurons in cortical columns during bi-stable perception, across V1, LGN inputs to V1, V2 and visuomotor area VRL/A/AL 3) Identify the contribution of specific processes to perceptual reversals: firing rates adaptation, mutual inhibition of rival sub-networks, synchrony and variability of firing.