Uncovering cell type-specific prefrontal neural mechanisms of visuospatial selective attention in freely behaving mice using a high-throughput touchscreen-based training system

项目摘要 选择性空间注意力，即在最重要的空间位置选择和优先处理信息的能力。 位置对于适应性行为至关重要。尽管对灵长类动物的广泛研究已经建立了 前额叶皮层（特别是额叶眼区，FEF）控制选择性视觉的必要性 注意，潜在的细胞类型和投射特异性神经回路机制仍然难以捉摸。我们最近 开发了严格的基于触摸屏的任务，用于灵长类动物在自由行为中的视觉空间选择性注意， 小鼠，以研究电路机制的问题，在遗传学上易于处理的模型，并在自然 （不受约束的）设置。然而，研究小鼠注意的细胞类型和投射特异性电路逻辑， （使用这些任务）是一项大规模的工作，迫切需要一个负担得起的高通量系统， 大量小鼠的平行训练。具体来说，对于广泛使用的触摸屏行为， 在行为神经科学界，这样的系统既不存在商业上的，也不作为开放源代码存在。 在这里，在目标1中，我们建议开发和建立低成本、高吞吐量、基于触摸屏的硬件 和软件平台，用于一次并行训练20只小鼠复杂的视觉引导行为 （包括我们的注意力任务）。我们假设这个开源系统的价格将低于1/10， 占据当前商业系统的不到1/3的空间，并提供灵活、易于使用的软件， 实验控制。初步数据-硬件和软件原型，建立这一目标的可行性。 接下来，在目标2中，我们将使用这种高通量系统在自由行为的小鼠中研究， 小鼠前额叶扣带回内生长抑素阳性抑制神经元 （被认为是FEF的类似物），在视觉空间选择性注意的控制中。我们会用细胞- SOM+ Cg中间神经元的类型特异性化学遗传学沉默在我们的小鼠侧翼任务上训练的小鼠中， 注意力，其将注意力位点与行为报告位点分离（总共35只SOM-cre小鼠）。 我们将把联合收割机行为测试与三维头部跟踪（和眼球跟踪）结合起来。我们假设Cg/SOM+ 神经元控制刺激竞争和跨越空间的目标选择，它们的破坏将损害目标， 选择准确性，而不产生纯粹的感觉或运动缺陷。这项工作的结果将有三个 重大影响。(a)他们将为Cg/SOM+中间神经元在注意力控制中的功能作用提供新的线索。 (b)他们将为我们计划的R 01奠定基础，该R 01旨在详细的细胞类型和投影特异性解剖。 扣带回子回路（使用光遗传学）和扣带回神经元表征（使用内窥镜Ca++ 成像）用于自由行为小鼠的视觉空间选择性注意。(c)同样重要的是， 这里开发的触摸屏培训平台将是一个强大的开源工具，用于更广泛的行为 神经科学界研究其他视觉引导认知功能的神经回路基础， 功能障碍，如认知控制，决策和成瘾，在自由移动的小鼠。

PROJECT SUMMARY Selective spatial attention, the ability to select and preferentially process information at the most important spatial location, is essential for adaptive behavior. Although extensive research in primates has established the necessity of the prefrontal cortex (and specifically, the frontal eye field, FEF) for the control of selective visual attention, the underlying cell-type and projection-specific neural circuit mechanisms remain elusive. We recently developed rigorous touchscreen-based tasks for primate-like visuospatial selective attention in freely behaving mice in order to investigate circuit mechanistic questions in a genetically tractable model and in a naturalistic (unrestrained) setting. However, investigating the cell-type and projection-specific circuit logic of attention in mice (using these tasks) is a large-scale effort that critically requires an affordable, high-throughput system for the parallelized training of large numbers of mice. Specifically, for touchscreen behaviors, which are used extensively in the behavioral neuroscience community, such a system does not exist either commercially or as open-source. Here, in Aim 1, we propose to develop and establish a low cost, high-throughput, touchscreen-based hardware and software platform for parallelized training of 20 mice at a time on complex visually guided behaviors (including our attention tasks). We hypothesize that this open-source system will cost <1/10th the price, and occupy <1/3rd the space, of current commercial systems, and offer flexible, easy-to-use software for stimulus and experimental control. Preliminary data - hardware and software prototypes, establish viability of this aim. Next, in Aim 2, we will use this high-throughput system to investigate in freely behaving mice, the causal role of somatostatin-positive (SOM+) inhibitory neurons in the cingulate subdivision (Cg) of the mouse prefrontal cortex (considered to be an analog of the FEF), in the control of visuospatial selective attention. We will do so with cell- type specific chemogenetic silencing of SOM+ Cg interneurons in mice trained on our mouse flanker task of attention, which dissociates the locus of attention from the locus of behavioral report (total of 35 SOM-cre mice). We will combine behavioral testing with 3-D head-tracking (and eye-tracking). We hypothesize that Cg/SOM+ neurons control stimulus competition and target selection across space, and that their disruption will impair target selection accuracy without producing purely sensory or motor deficits. Results from this work will have three major impacts. (a) They will shed new light on the functional role of Cg/SOM+ interneurons in attention control. (b) They will set the stage for our planned R01 aimed at detailed cell-type and projection-specific dissection of cingulate sub-circuits (using optogenetics) and cingulate neuronal representations (using endoscopic Ca++ imaging) for visuospatial selective attention in freely behaving mice. (c) Equally importantly, the high-throughput touchscreen training platform developed here will be a potent open-source tool for the broader behavioral neuroscience community investigating the neural circuit basis of other visually guided cognitive functions and dysfunctions as well, such as cognitive control, decision-making, and addiction, in freely moving mice.