Collaborative Research: Sensorimotor control of hand-object interactions

合作研究：手与物体交互的感觉运动控制

• 批准号: 1455865
• 负责人: Andrew Gordon
• 金额: $ 37.17万
• 依托单位: Teachers College, Columbia University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1455865&HistoricalAwards=false
• 关键词: Collaborative; Research; Sensorimotor; control; hand

How people use their hands to interact with objects is one of the most complex and least understood sensorimotor skills. For example, when people pick up the same object multiple times they use different fingertip forces to compensate for differences in their fingertip positions. This compensation is important because it allows people to manipulate an object skillfully without having always to grasp it at the exact same points. Several important questions remain: What is the relative contribution of individual sensory modalities, such as touch and vision, in enabling the compensatory modulation of fingertip force and position? Are fingertip forces and positions represented separately by the central nervous system? What are the mechanisms underlying their generalization to different limbs or tasks? These questions represent a major gap in our understanding of how the central nervous system learns, plans, and executes complex motor behaviors. An understanding of how sensory modalities are seamlessly integrated to enable the development and implementation of high-level internal representations of hand-object interactions should inspire the design of more dexterous robotic manipulators endowed with sensory feedback, improve neuroprosthetics, and aid development of advanced bioengineering research tools to quantify biological control mechanisms.The overall goal of this collaborative research is to elucidate the mechanisms responsible for building high-level representations of hand-object interactions accounting for end-effector position and force. The aims are 1) to quantify the mechanisms underlying the weighting of sensorimotor integration during hand-object interactions and 2) to determine the principles underlying generalization of a learned hand-object interaction to a new context. The investigators will test three hypotheses: 1) the weighting of different sensory modalities is time dependent according to its role within a given task epoch; 2) co-variation between end-effector force and position is a general feature of hand-object interactions, independent of the effectors used (fingertip, whole hand, or two-hands); and 3) generalization of learned hand-object interactions is sensitive to the frame of reference in which they were learned.

人们如何用手与物体互动是最复杂和最不了解的感觉运动技能之一。例如，当人们多次拿起同一物体时，他们会使用不同的指尖力来补偿指尖位置的差异。这种补偿很重要，因为它允许人们熟练地操纵物体，而不必总是在完全相同的点上抓住它。几个重要的问题仍然存在：什么是相对贡献的个人感觉方式，如触摸和视觉，在使补偿调制的指尖的力量和位置？指尖的力和位置是由中枢神经系统分别代表的吗？它们泛化到不同的肢体或任务的机制是什么？这些问题代表了我们对中枢神经系统如何学习、计划和执行复杂运动行为的理解存在重大差距。了解感觉模态如何无缝集成，以实现手-物体交互的高级内部表示的开发和实现，应该会激发设计具有感觉反馈的更灵巧的机器人操纵器，改善神经修复术，并帮助开发先进的生物工程研究工具，以量化生物控制机制。这项合作研究的总体目标是阐明机制负责构建考虑末端执行器位置和力的手-对象交互的高级表示。其目的是1）量化的机制加权的感觉运动整合在手对象的相互作用和2），以确定原则的基础泛化的学习手对象的相互作用，以一个新的背景。研究人员将检验三个假设：1）根据其在给定任务时期内的作用，不同感觉模态的权重是时间依赖的; 2）末端效应器力和位置之间的共变是手-物体交互的一般特征，与所使用的效应器无关（指尖、整只手或双手）;以及3）所学习的手-对象交互的泛化对它们被学习的参考系敏感。