C3: Behavior Automation

C3：行为自动化

• 批准号: 10705970
• 负责人: Carlos D Brody
• 金额: $ 32.13万
• 依托单位: PRINCETON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-08-08 至 2028-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10705970
• 关键词: Animals; Automation; Behavior; Behavioral; Behavioral Research; Cognitive; Collaborations; Communities; Data; Data Science; Databases; Decision Making; Documentation; Ensure; Face; Foundations; Goals; Grant; Health; Individual; Intervention; Laboratories; Left; Maintenance; Measurement; Methodology; Methods; Neurosciences; Optics; Pupil; Reproducibility; Research Personnel; Research Project Grants; Resources; Rewards; Rodent; Schedule; Services; Short-Term Memory; Standardization; System; Training; Water; Work; design; experimental study; human error; imaging system; improved; innovation; laboratory experience; laboratory experiment; neural circuit; neuromechanism; programs; repository; tool; virtual reality; web site

Project Summary/Abstract: Core 3, Behavior Automation The main goal of this U19 project is to elucidate the neural mechanisms underlying working memory and decision-making. As a consequence, each research project in this application involves rodents trained on cognitively demanding tasks that require the animals to accumulate evidence to make decisions. All these animals are trained in the Behavior Automation Core’s central facility, which provides shared virtual-reality (VR) and non-VR behavioral rigs, standardizing behavior across the research projects. The first aim of this core is to provide this standardized behavioral training facility with cutting-edge efficiency and automation. As demand for our services increases, we will continue to expand its capacity. We have developed a miniature VR rig that will allow us to double the number of rigs in our space. As the training facility expands, the number of animals that each technician handles will increase as well, requiring us to develop new automation tools to ensure the health of the animals and the quality of the data. This core will develop and integrate new tools into the existing high-throughput workflow to reduce human error and improve technician interventions. We will also improve our systems to support and maintain experimental rigs; manage technician hiring, training, and scheduling to meet demand; and oversee technicians’ workflow, animal management, and facility regulatory compliance. The core’s second aim is to design a modular behavioral rig that meets diverse experimental needs. Rather than redesigning the rigs to accommodate new experiments, which would impose a substantial barrier to innovation, we plan to develop a modular rig that will provide a shared framework for diverse experimental needs. We will design and develop new modules, fitting them into the existing workflows or creating new workflows to store and share the generated data across all the research projects and other resource cores. Finally, the third aim is to provide behavioral resources for the neuroscience community. To share our methods with the broader scientific community, we are developing a website that contains comprehensive and easy-to-navigate documentation. This website will provide resources such as a detailed explanation of how to build experimental rigs, maintenance routines, a comprehensive description of facility workflow, how to set up the behavioral database and integrate it with the rigs, and animal workflows. By providing this information, we hope to facilitate wider standardization of these methods. This resource will enable researchers around the world to introduce automated training for well-controlled, cognitively demanding tasks in their own laboratories. In summary, our shared behavioral training facility will support the experimental components of all the research projects within this U19 program. More generally, we expect that our pioneering methods for automating and managing animal training, when shared with the broader community, will improve efficiency, rigor, and reproducibility in behavioral research across the field of neuroscience.

项目概要/摘要：核心3，行为自动化 这个U19项目的主要目标是阐明工作记忆的神经机制， 决策的因此，本申请中的每个研究项目都涉及经过以下训练的啮齿动物： 需要动物积累证据来做出决定的认知任务。所有这些 动物在行为自动化核心的中央设施中接受训练，该设施提供共享的虚拟现实（VR） 和非虚拟现实行为平台，标准化研究项目中的行为。该核心的第一个目标是 为这种标准化的行为训练设施提供最先进的效率和自动化。由于需求 我们的服务增加，我们将继续扩大其能力。我们开发了一种微型虚拟现实设备， 让我们的空间里的钻机数量翻倍随着训练设施的扩大， 每个技术人员的处理量也将增加，这要求我们开发新的自动化工具，以确保 动物的健康和数据的质量。该核心将开发新的工具并将其整合到现有的 高通量工作流程，以减少人为错误并改善技术人员的干预。我们还将完善 我们的系统支持和维护实验钻机;管理技术人员的招聘、培训和调度， 满足需求;监督技术人员的工作流程、动物管理和设施法规合规性。 核心的第二个目标是设计一个模块化的行为钻机，以满足不同的实验需求。 而不是重新设计钻机，以适应新的实验，这将强加一个实质性的障碍， 为了创新，我们计划开发一种模块化装置，为各种实验提供一个共享的框架， 需求我们将设计和开发新的模块，将其融入现有的工作流程或创建新的 工作流来存储和共享所有研究项目和其他资源核心生成的数据。 最后，第三个目标是为神经科学界提供行为资源。分享我们 方法与更广泛的科学界，我们正在开发一个网站，其中包含全面的， 易于导航的文档。该网站将提供资源，如详细解释如何 搭建实验台，日常维护，全面描述设施工作流程，如何设置 行为数据库并将其与钻机和动物工作流程集成。通过提供这些信息，我们 希望促进这些方法的更广泛的标准化。该资源将使研究人员能够在 世界各国都希望在自己的实验室中为控制良好、认知要求高的任务引入自动化培训。 总之，我们的共享行为训练设施将支持所有研究的实验部分 在U19项目中。更一般地说，我们希望我们的自动化和 管理动物训练，当与更广泛的社区分享时，将提高效率，严谨性， 在整个神经科学领域的行为研究的可重复性。