EYES AS A WINDOW ON EFFERENT COPY IN MOTOR CONTROL

有证据表明，眼球运动系统有特权接触到驱动手运动的神经过程。虽然这种特权访问的性质本身就是一个有趣的研究对象，但这项研究将试图利用它来加深我们对运动学习过程的理解。也就是说，眼球运动将被用来表明受试者对他们的手运动结果的期望，从而允许运动表现和意图之间的分离。为了以这种方式使用眼球运动，有必要将视觉和本体感觉反馈的效果与那些反映控制信号的效果(即传出复制)分开。这可以通过应用计算模型，例如卡尔曼滤波，以及通过操纵视觉和本体感觉反馈来实现。我们的初步结果显示，在这种情况下，眼球运动似乎可测量地反映出手臂运动指令的传出副本。这些初步结果将得到充实和扩展，因为该项目的成功取决于开发一种范例，在这种范例中，手臂动作的学习效果可以可靠地证明眼睛的运动。一旦开发出这样的范式，研究将集中在运动学习的时间进程上。实验室以前的工作表明，运动学习包括发展正向模型(用于预测手的行为和执行反馈控制)和反向模型(用于开环控制)。然而，一个开放而重要的问题是，这两种模式是如何相互影响的，以及它们在多大程度上是同时或顺序学习的。为了解决这个问题，我们将使用眼球运动作为替代，以表明手臂运动的向前模型的适应性。

There is evidence suggesting that the oculomotor system has privileged access to the neural processes driving hand movements. While the nature of this privileged access is an interesting object of study in its own right, this research will seek to exploit it in order to deepen our understanding of the processes of motor learning. That is, eye movements will be used to indicate the subjects' expectations regarding the results of their hand movements, thus allowing a disassociation between motor performance and intention. In order to allow the eye movements to be used in this way, it will be necessary to separate effects of visual and proprioceptive feedback from those reflecting control signals (i.e., efferent copy). This can be done by application of computational models, for example Kalman filters, and through manipulation of visual and proprioceptive feedback. Our preliminary results show that in such conditions, eye movements appear to measurably reflect efferent copy of motor commands to the arm. These preliminary results will be fleshed out and expanded, as the success of the project depends on developing a paradigm in which effects of learning of arm movements can be reliably demonstrated movements of the eyes. Once such a paradigm is developed, research will focus on the time course of motor learning. Previous work in the laboratory shows that motor learning involves developing both a forward model (for predicting behavior of the hand and performing feedback control) and an inverse model (used for open-loop control). However, an open and important question is how these two models influence each other and to what degree they are learned simultaneously or sequentially. In order to resolve this question, we will use eye-movements as a proxy to indicate adaptation of the forward model of arm movements.