Postdoctoral Research Fellowships in Biology for FY 2009

这一行动为NSF 2009财年博士后研究奖学金提供了资金。该研究金支持一项名为“飞行发动机控制中的信息处理”的研究和培训计划。这项研究的主办机构是华盛顿大学，赞助科学家是托马斯·丹尼尔。要了解动物非凡的运动能力，一个核心挑战是确定神经反馈是如何通过肌肉和骨骼元素控制运动的。信息论为神经和机械信号与感觉刺激、运动指令和身体动力学之间的联系提供了一种通用的度量信息。然而，要理解这些信息的含义，需要通过神经机械实验、模拟和验证来使相关信号在其功能背景下扎根。这项研究为运动控制开发了基于信息的分析工具，同时解决了有翼昆虫Manduca sexta sexta的飞行控制的具体假设。首先，使用信息优化程序来揭示假想的权力和控制肌肉群的功能组织。其次，最大信息算法描述了翅膀运动学和身体动力学的组成部分，这些肌肉组通过神经反馈获得了最多的信息。现有的生物力学模拟和对行为动物神经和肌肉活动的直接操纵被用来验证所提出的肌群、翅膀参数和身体动力学之间的功能联系。这项研究的信息传递的一般原则为生物系统中感觉运动的整合提供了一种共同语言。在应用方面，揭示神经机械系统中的信息流原理对于神经假体、生物启发的机器人和脑机接口等新兴领域至关重要。也许最重要的是，这个项目提供了一种跨学科的培训方法。具体的培训目标包括：1)开发一门新的计算神经力学短期课程；2)在需要实验、工程和编程技能的项目中直接指导年轻的(高中)研究人员；3)通过与太平洋科学中心飞行博物馆的合作向公众宣传。

This action funds an NSF Postdoctoral Research Fellowship for FY 2009. The fellowship supports a research and training plan entitled "Information processing in the motor control of flight." The host institution for this research is the University of Washington, and the sponsoring scientist is Thomas Daniel.A central challenge in understanding the remarkable locomotion capabilities of animals is determining how neural feedback acts through muscles and skeletal elements to control motion. Information theory provides a common metric, information for relating neural and mechanical signals across sensory stimuli, motor commands, and body dynamics. However, understanding the meaning of this information requires grounding the relevant signals in their functional context through neuromechanical experiment, simulation, and validation. This research develops information-based analytical tools for motor control while addressing specific hypotheses of flight control in a winged insect, the hawkmoth, Manduca sexta. First, an information optimization procedure is being used to reveal the functional organization of groups of hypothesized power and control muscles. Second, maximal information algorithms describe the components of wing kinematics and body dynamics most informed by neural feedback through these groups of muscles. Existing biomechanical simulations and direct manipulation of neural and muscular activity in behaving animals are used to validate the proposed functional link between muscle groups, wing parameters, and body dynamics. The general principles of information transfer from this research offer a common language for sensorimotor integration in biological systems. On the applied side, revealing principles of information flow in neuromechanical systems is critical to the emerging fields of neural prosthetics, biologically-inspired robotics, and brain-machine interfaces. Perhaps most importantly, this project offers an interdisciplinary training approach. Specific training goals include 1) developing a new computational neuromechanics short course 2) direct mentoring of younger (high school) researchers in projects requiring experimental, engineering, and programming skills and 3) public outreach through an established collaboration with the Pacific Science Center Museum of Flight.