A Neuroethological Approach to Understanding Cerebellar Function

神经科学的许多突破都来自对“冠军”动物物种的研究。例如，我们对听觉的了解大部分来自对猫头鹰和蝙蝠的研究，这些动物的生存依赖于在黑暗中捕捉食物。 关于神经元相互交流的电信号的基本知识来自对鱿鱼的研究，其中电信号通过特别大的“电线”传播，使鱿鱼能够快速反应，从而避免成为一顿饭（并使它们易于神经科学家研究）。在这里，PI将应用这种简单的逻辑来理解大脑中的一个区域，这个区域长期以来一直吸引着神经科学家，但其功能仍然是个谜。虽然小脑的体积相对较小，但它包含了人类大脑中所有神经元的约3/4。此外，小脑疾病会导致严重的运动、情感和思维障碍，包括自闭症。开发治疗方法需要更好地了解小脑的正常功能。为了实现这一目标，该项目建议研究动物的小脑，其中小脑是最大和最高度发达的-一组来自非洲的鱼类被称为mormyrids。象鼻虫的小脑是如此之大，以至于它们的大脑甚至比人类的大脑还要大（就它们的身体尺寸而言）。这将有助于他们建立小脑如何工作的模型，直到其微观结构的细节。由于这种微观结构在动物中极其相似，我们的工作可能揭示直接适用于人类的基本原理。尽管小脑的晶体回路长期以来一直激发着将神经回路结构和功能联系起来的努力，但我们对其功能的理解仍然局限于少数情况，例如经典的反射条件反射。该项目通过将综合实验和计算方法应用于研究具有比例最大和最高度发达的小脑的脊椎动物群-弱电拟鱼来解决这一挑战。对一种常见的拟鳚科物种彼得象鼻鱼的初步研究发现，小脑中有一个被称为C1的区域，专门用于控制手指状的下巴附属物或雪纳泽器官，这些鱼因此而得名。高移动性的雪纳瑞器官上密布着电感受器，对这个物种的觅食行为至关重要。值得注意的是，C1输出神经元直接投射到脑干运动神经元，这些神经元支配雪纳瑞器官肌肉并控制其运动。这种简洁的电路与哺乳动物小脑对运动或伸手等常见行为的复杂和高度分布的路径形成了鲜明对比。由于该系统的固有优势和生态有效性，加上紧密集成的实验和计算建模方法，预计将取得快速进展。在实验方面，PI将利用新方法进行高分辨率行为分析和自由游动的鱼类在觅食期间的大规模神经记录。 机器学习工具、神经回路模型和生物力学模型将使我们能够识别雪纳泽器官感觉运动控制中涉及的计算问题，并开发和测试有关如何解决这些问题的假设。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估来支持。

Many breakthroughs in neuroscience have come from studies of “champion” animal species. For example, much of what we know about hearing has come from studies of owls and bats, animals whose survival depends on capturing their meals in darkness. Fundamental knowledge about the electrical signals through which neurons communicate with one another came from studies of squid, where electric signals travel through “wires” that are particularly large, enabling fast reactions that help the squid to avoid becoming a meal (and making them easy for neuroscientists to study). Here, the PI will apply this straightforward logic to understand a region of the brain that has long fascinated neuroscientists but whose function remains mysterious. Though relatively small in size, the cerebellum contains ~3/4 of all the neurons in the human brain. Moreover, diseases of the cerebellum cause profound disorders of movement as well as emotion and thought, including autism. Developing treatments requires a better understanding of the normal functioning of the cerebellum. To achieve this, the project proposes to study the cerebellum in the animal in which it the largest and most highly-developed—a group of fish from Africa known as the mormyrids. The cerebellum is so large in mormyrids that their brain is even larger (for their body size) than the human brain. This will help them to build models of how the cerebellum works, down to the details of its microscopic structure. Because this microscopic structure is extremely similar across animals, our work may reveal fundamental principles that apply directly to humans.Though the crystalline circuitry of the cerebellum has long inspired efforts to link neural circuit structure and function, our understanding of its function remains restricted to a few select cases, such as classical reflex conditioning. This projects addresses this challenge by applying integrated experimental and computational approaches to studies of the vertebrate group with the proportionately largest and most highly developed cerebellum--weakly electric mormyrid fish. Preliminary studies of a common mormyrid species, Peter’s elephant-nose fish, have identified a region of the cerebellum, known as C1, dedicated to controlling the finger-like chin appendage, or schnauzenorgan, for which these fish are named. The highly-mobile schnauzenorgan is densely covered with electroreceptors and is vital for the foraging behavior of this species. Remarkably, C1 output neurons project directly to the brainstem motor neurons that innervate schnauzenorgan muscles and control its movement. This concise circuitry contrasts with the complex and highly distributed paths via which the mammalian cerebellum contributes to commonly studied behaviors such as locomotion or reaching. Rapid progress is expected due to the inherent advantages and ecological validity of this system coupled with the tightly integrated experimental and computational modeling approaches. On the experimental side, the PI will leverage new methods for high-resolution behavioral analysis and large-scale neural recordings in freely swimming fish during foraging. Machine learning tools, neural circuit models, and biomechanical models will allow us to identify the computational problems involved in sensorimotor control of the schnauzenorgan and to develop and test hypotheses regarding how they may be solved.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.