Instructive Signals for Motor Learning

 描述(申请人提供)：神经元之间的突触是可塑性的--能够变得更强或更弱。当我们学习时，新的记忆被编码到我们大脑中修改过的突触中。这个项目的目的是了解决定哪些突触在学习过程中会发生变化的规则，以及这种变化如何导致适应的行为。特别是，我们将分析运动中的错误是如何触发突触变化的，而突触变化反过来又提高了后续运动的准确性。当做出不正确的动作时，大脑会得到关于错误的反馈，并使用这些信息来引导适当突触的可塑性诱导。大脑中负责运动学习的部分，即小脑，通过一种被称为爬行纤维的突触输入来获得关于错误的反馈。当运动中出现错误时，攀爬纤维中的活动是向小脑发送信息的“错误信号”，即控制运动的电路需要通过调整一些突触的强度来调整。被修饰的通常是最近活跃的突触。我们将分析攀爬纤维或其他小脑神经元中的哪些活动模式是诱导可塑性所必需的，也是充分的。控制电路中突触可塑性诱导的规则定义了电路用来学习的算法。更好地理解这些规则可以指导策略，更有效地挖掘神经回路的学习潜力，无论是在健康的个人中，那些患有神经障碍的人，还是在中风后重新学习如何控制自己的运动的患者中。这项提议不仅解决了运动学习领域中尚未回答的问题，而且与更广泛地理解神经回路中的突触可塑性机制如何针对其控制的行为的特定计算需求进行微调有关。

 DESCRIPTION (provided by applicant): Synapses between neurons are plastic - able to become stronger or weaker. When we learn, new memories are encoded in modified synapses across our brains. The aim of this project is to understand the rules that determine which synapses will change during learning, and how that change results in an adapted behavior. In particular, we will analyze how an error in a movement acts as a trigger for synaptic change that, in turn, improves the accuracy of subsequent movements. When an incorrect movement is made, the brain gets feedback about the error, and uses this information to guide the induction of plasticity at appropriate synapses. The part of the brain responsible for motor learning, the cerebellum, gets this feedback about errors through a synaptic input known as a climbing fiber. When an error in movement occurs, activity in the climbing fiber is an "error signal" that sends the message to the cerebellum that the circuit controlling the movement needs to be adjusted by adjusting the strength of some of the synapses. It is usually the synapses that were recently active that are modified. We will analyze which patterns of activity in the climbing fibers or othe cerebellar neurons are necessary and sufficient to cause plasticity to be induced. The rules governing the induction of plasticity at the synapses in a circuit define the algorithm that circui uses to learn. A better understanding those rules can guide strategies to more effectively tap the learning potential of neural circuits in both healthy individuals, those with neurological disorder, and in patients relearning how to control their movements after stroke This proposal addresses not just unanswered questions in the field of motor learning, but is relevant for a more general understanding of how the synaptic plasticity mechanisms in a neural circuit may be finely tuned for the specific computational demands of the behavior it controls.