VISUAL INFORMATION FOR REACH AND GRASP

便于获取和掌握的视觉信息

• 批准号: 2711184
• 负责人: DANIEL J KERSTEN
• 金额: $ 9.48万
• 依托单位: UNIVERSITY OF MINNESOTA
• 依托单位国家: 美国
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-07-01 至 2000-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2711184
• 关键词: arm; behavioral /social science research tag; body movement; cues; hand; human subject; neural information processing; size perception; visual depth perception; visual feedback; visual perception; visual stimulus

DESCRIPTION: Knowledge of how visual information determines perceived depth and shape has grown considerably in the past few decades. We know less, however, about how this knowledge relates to performance in everyday tasks. It is widely appreciated that perceived depth is important for determining large scale spatial layout for a variety of actions, and that shape perception is important for object recognition. However, the visual information for depth and shape also provides critical image measurements for programming reach and grasp by the hand. Despite the importance of vision for prehension, we know surprisingly little about how multiple sources of visual information are used for hand movements. A central issue is the degree to which perceptual judgements of shape and depth predict measures of reaching and grasping. A common working hypothesis is that motor planning makes use of the full perceptual "reconstruction" of the scene to program its movements. The investigator will critically examine this assumption in a series of experiments. In particular, he plans to 1) compare visual processing for reaching with that for conscious perceptual judgements of depth under cue conflict situations; 2) understand how multiple cues for depth, size, and orientation combine to determine reach and grasp; 3) examine the role of dependence on learned views of an object for grasping. These studies will make use of recently acquired laboratory instrumentation which allows computer-controlled presentation of both real and synthetic visual stimuli together with real-time monitoring of movements.

描述：关于视觉信息如何决定感知深度的知识 在过去的几十年里，形状已经有了相当大的增长。我们知道的更少， 然而，关于这种知识如何与日常任务中的表现相关。 人们普遍认识到，感知深度对于确定 用于各种动作大规模空间布局，以及形状 感知对于物体识别是很重要的。然而，视觉上 深度和形状信息还提供关键的图像测量 用于编程手部伸展和抓取。尽管重要的是 对于恐惧的视觉，我们令人惊讶的是知之甚少 视觉信息源用于手部运动。一个中心问题 是对形状和深度的知觉判断预测的程度 伸手和抓住的措施。一种常见的工作假说是 运动规划充分利用了人类对运动的感知 场景来规划它的动作。调查员将严格审查 这一假设在一系列实验中。特别是，他计划1) 比较触觉视觉加工与意识知觉视觉加工 在线索冲突情况下对深度的判断；2)理解 深度、大小和方向的多个提示组合在一起以确定覆盖范围 和把握；3)考察依赖于对一个对象的习得观点的作用 用来抓人。这些研究将利用最近获得的实验室 允许计算机控制的实数和实数表示的仪器 和合成视觉刺激，以及实时监控 动静。