Exploring and Modeling Entrainment to Rhythm in the Motor System

人类大脑不断地识别其环境的时间动态中的规律和模式，并利用它们来预测接下来会发生什么以及何时发生。在许多情况下，随着时间的推移，预测的过程涉及用于运动控制的大脑系统的关键节点，但我们对这些运动区域及其在非运动过程中的作用的理解非常有限。了解运动系统在时间预测中的作用，对于理解和治疗运动系统疾病（包括帕金森氏症和亨廷顿氏症）的各种症状非常重要，对于建立对我们在世界上的身体体验与我们的认知之间关系的更深层次的基本理解也非常重要。运动系统的两个组成部分，即辅助运动区和基底神经节，已经被最终证明是由以循环模式节奏（即，“夹带”到节拍）。在我最近的工作中，我提出了一些假设，详细说明了这些“运动”结构在感知过程中的作用。这些假设阐明了感知节奏跟踪和运动的选择和控制之间的生理相似之处。我进一步提出了一个概念和计算框架，在这个框架中，这个过程可以被正式理解和建模为一个关于潜在周期的相位和克里思的最佳推理过程，朝着将其与越来越有影响力的基于推理的运动控制理论统一起来迈出了一步。我的下一步是实验验证这两个理论框架。为了验证我的神经生理学理论，我计划在参与者听相同的节奏时，在有或没有适当的背景或学习识别节拍的情况下记录功能性磁共振成像，并使用多变量模式分析和格兰杰因果关系测量来寻找基底神经节和辅助运动区之间预测相互作用的证据。为了测试我的计算理论，我计划设计多个并行的心理物理学和EEG节律跟踪任务，在多个感官模态，这将使我能够适合模型参数的个人参与者，并确定这些参数是否显示出跨任务和模态的一致性。这些理论将被细化，扩展，和/或大修，必要时考虑到实验数据。通过验证和进一步发展这些理论，该提案旨在进一步发展运动系统在运动和非运动过程中的动态功能的统一生理和计算理论的长期目标。

The human brain is constantly identifying regularity and patterns in the temporal dynamics of its environment and using them to predict what comes next and when. In many cases, the process of prediction over time involves key nodes of the brain system used for motor control, but our understanding of these motor regions and their roles in non-motor processes is very limited. Understanding the role of the motor system in temporal prediction is important to making sense of and treating the diverse symptoms of motor system disorders including Parkinson's and Huntington's disease, and also to building a deeper fundamental understanding of the relationship between our physical experience in the world and our cognition. Two components of the motor system, the supplementary motor area and the basal ganglia, have been conclusively shown to be activated by the task of following and anticipating auditory events in a cyclically-patterned rhythm (i.e., "entraining" to a beat). In my recent work, I have developed hypotheses detailing the roles of these "motor" structures in this perceptual process. These hypotheses illuminate physiological parallels between perceptual rhythm tracking and the selection and control of movement. I have further proposed a conceptual and computational framework in which this process can be formally understood and modeled as a process of optimal inference about the phase and tempo of an underlying cycle, taking a step toward unifying it with an increasingly influential inference-based theory of motor control. My next step is to experimentally validate these two theoretical frameworks. To test my neurophysiological theory, I plan to record fMRI as participants listen to identical rhythms with and without appropriate context or learning to recognize a beat, and use multivariate pattern analysis and Granger causality measures to look for evidence of predicted interactions between basal ganglia and supplementary motor area. To test my computational theory, I plan to design multiple parallel psychophysics and EEG rhythm-tracking tasks in multiple sensory modalities that will allow me to fit model parameters to individual participants and determine whether these parameters show consistency across tasks and modalities. These theories will be refined, extended, and/or overhauled as necessary to account for experimental data. By validating and further developing these theories, this proposal aims to further the long-term goal of developing a unified physiological and computational theory of the dynamic function of the motor system in both motor and non-motor processes.