Variability and Stability in Skill Acquisition

描述（由申请人提供）：感知运动技能的获得及其对不断变化的任务需求的适应是日常生活的基础。技能和适应性的丧失对功能有害，并且存在于感觉运动系统的许多神经疾病中。因此，进一步深入了解这些进程及其恢复至关重要。为了阐明运动的获取、适应和控制的过程，拟议的研究测试了这样一种假设，即中枢神经系统对自身的变化非常敏感，不仅减少了不必要的内在噪音，而且还制定了适应甚至利用这种噪音的策略。这一假说基于两个假设：第一，感觉运动系统具有来自复杂层次过程的内在神经运动噪声;第二，行为任务通常是冗余的，并提供许多不同的方式来实现等效的任务结果。因此，大脑寻求具有稳定性的解决方案，这些解决方案相对于其噪声是鲁棒的。上一个资助周期的工作确定了实现任务稳定性的三个概念上不同的途径：宽容：在实践中，人类探索和遍历解决方案的空间，以找到那些容忍错误和噪音的策略。协变：任务冗余提供的解决方案，相关变量之间的协变实现相同的结果，在任务性能，同时允许在个别变量的变化。噪声：必要时，可以降低随机分量的幅度。这三管齐下的TCN区别提出了定量框架来评估假设，即在获得熟练的行为中，中枢神经系统开发了“智能”解决方案，减少，适应和利用不可避免的神经运动噪声。12个新的实验验证了这一假设，并以研究结果为平台，设计新的干预技术。该研究分为三个目标：目标1下的实验专注于宽容和测试系统是否寻求最好的解决方案，以适应个人的可变性。相反，我们还测试是否操纵个人的可变性可以加速适应宽容的解决方案。目标2下的实验研究如何实现变量的协变，使得固有噪声对任务结果的影响最小。管理增强信息以调查是否可以促进这种轨迹的获取。目标3下的实验检查是否可以通过添加外部噪声和操纵错误信息来减少内在神经运动噪声。拟议的研究将并行进行两个任务：彩虹糖，主要是前馈控制下的目标导向的离散投掷动作，和球弹跳，连续的感知引导技能的节奏击球。通过对这两项任务进行等效的实验操作，我们测试了我们假设的一般性，即神经系统在熟练的行为中适应和利用内在的神经运动噪声。这种定量TCN方法的结果将揭示运动技能的控制和获得的方式，还没有在任何其他现存的研究。重要的是，我们还从新的基本见解直接过渡到干预技术。我们提出了三种类型的干预措施，专门针对优化任务的容忍度，最大限度地提高协变，并减少噪音。因此，我们建立了必要的桥梁，从理论概念到实用技术，将适用于各种神经功能缺损。虽然这里提出的研究重点是健康的人类，但目前正在与斯坦福大学医学中心的Terence桑格博士密切合作，在患有运动障碍性脑瘫的儿童中测试这些概念。拟议的健康人类工作是与德国吉森大学的Hermann M <$ller博士和荷兰奈梅亨大学的Tjeerd Dijkstra博士的国际合作，并将涉及学生交流。 公共卫生相关性：该建议旨在阐明获得，适应和控制熟练动作的过程。我们的基本实验与直接利用我们基本见解的干预措施的测试研究相结合。这种双重方法提供了一个坚实的基础，将导致更深入地了解技能的获取和理论接地干预，以恢复各种神经系统疾病的感觉运动功能。与斯坦福大学医学院的桑格博士正在进行的合作旨在证明这里开发的干预技术可以改善运动障碍性脑瘫儿童的运动功能和学习。

DESCRIPTION (provided by applicant): The acquisition of perceptual-motor skills and their adaptation to changing task demands is fundamental to everyday life. Loss of skill and adaptability is detrimental to functioning and is present in many neurological diseases of the sensorimotor system. Hence, further insights into these processes and their rehabilitation is of utmost significance. To elucidate the processes underlying acquisition, adaptation, and control of movements the proposed research tests the hypothesis that the central nervous system is exquisitely sensitive to its own variability and not only reduces unwanted intrinsic noise but has also developed strategies that accommodate and even utilize this noise. This hypothesis rests on two assumptions: first, the sensorimotor system has intrinsic neuromotor noise arising from complex hierarchical processes; second, behavioral tasks are typically redundant and afford many different ways to achieve equivalent task outcomes. Hence, the brain seeks solutions with stability that are robust with respect to its noise. Work of the previous funding cycle established that there are three conceptually distinct routes to achieve task stability: Tolerance: During practice humans explore and traverse the space of solutions in order to find those strategies that are tolerant to error and noise. Covariation: Task redundancy offers solutions where covariation among relevant variables achieves the same result in task performance while allowing variation in individual variables. Noise: When necessary, the amplitude of the random components can be reduced. This three-pronged TCN-distinction presents the quantitative framework to evaluate the hypothesis that in acquiring skilled behavior the central nervous system develops "smart" solutions that reduce, accommodate, and utilize the inevitable neuromotor noise. Twelve new experiments test this hypothesis and take findings as the platform to design novel intervention techniques. The research is organized into three aims: Experiments under Aim 1 focus on Tolerance and test whether the system seeks solutions that best accommodate for the individual's variability. Conversely, we also test whether manipulating the individual's variability can accelerate adaptation to tolerant solutions. Experiments under Aim 2 examine how Covariation of variables is achieved such that intrinsic noise has minimal effect on the task result. Augmented information is administered to investigate whether the acquisition of such trajectories can be facilitated. Experiments under Aim 3 examine whether intrinsic neuromotor Noise can be reduced by adding extrinsic noise and manipulating error information. The proposed research will be conducted on two tasks in parallel: Skittles, a target-oriented discrete throwing action predominantly under feedforward control, and Ball Bouncing, a continuous perceptually-guided skill of rhythmically hitting a ball. By performing equivalent experimental manipulations to both tasks, we test the generality of our hypothesis that the nervous system accommodates and utilizes intrinsic neuromotor noise in skilled behavior. Results from this quantitative TCN-approach will shed light on the control and acquisition of movement skills in ways that have not been addressed in any other extant research. Importantly, we also make the much-desired transition from new basic insights directly to intervention techniques. We propose three types of interventions that specifically aim to optimize task tolerance, maximize covariation, and reduce noise. We thereby establish the necessary bridge from theoretical concepts to practical techniques that will be applicable to a variety of neurological deficits. While the research proposed here is focused on healthy humans, complementary work is currently under way in close collaboration with Dr. Terence Sanger at Stanford University Medical Center that tests these concepts in children with dyskinetic cerebral palsy. The proposed work on healthy humans is an international collaboration with Dr. Hermann M¿ller at the University of Giessen, Germany, and Dr. Tjeerd Dijkstra at the University of Nijmegen, Netherlands, and will involve student exchanges. PUBLIC HEALTH RELEVANCE: This proposal is directed at elucidating the processes underlying the acquisition, adaptation, and control of skilled movements. Our basic experiments are paired with studies that test interventions that directly capitalize on our basic insights. This dual approach provides a solid basis that will lead to a deeper understanding of skill acquisition and to theoretically grounded interventions to restore sensorimotor function in a variety of neurological disorders. The ongoing collaboration with Dr. Sanger at the Stanford Medical School aims to demonstrate that the intervention techniques developed here can achieve improvements in motor function and learning in children with dyskinetic cerebral palsy.