Visual Memory Mechanisms in Transsaccadic Integration and Overt Search

视觉科学中的一个基本问题是，当视觉信息通过一系列短暂的扫视进入视觉系统时，人们如何感知他们面前的连续视觉环境，这些扫视被频繁的快速眼球运动打断。部分原因可能是视觉系统依赖于某种形式的视觉记忆，以允许扫视之间的连续性。在目前的工作中，研究小组将研究这种类型的视觉记忆是如何运作的。更好地理解视觉记忆在眼球运动中的基本操作可能会导致实际应用，例如放射科医生，行李检查员，卫星图像分析师和其他职业需要他们在视觉搜索过程中搜索关键视觉信息的人的优化程序。另一个潜在的应用可能是人工视觉系统的优化设计。 这项工作的目的是回答两个基本的问题，关于操作的视觉记忆之间的扫视。首先，本研究的目的是调查这种视觉记忆的容量：这种类型的记忆可以容纳多少视觉信息？其次，本研究还旨在探讨视觉记忆机制的容量限制如何限制人类在视觉搜索任务中的表现。例如，当超过视觉记忆的容量时，这会如何影响一个人搜索可能隐藏在许多其他视觉项目中的特定视觉信息的能力？一组严格的计算和实验技术将被用来表征视觉搜索过程中的视觉记忆中包含的信息，以及评估视觉记忆对视觉搜索性能的影响。这些技术将允许详细和新颖的定量预测视觉敏感性和视觉记忆能力如何相互作用，以确定人类在视觉搜索任务中的表现。模型预测将通过心理物理实验进行测试。

A fundamental question in vision science concerns how people perceive a continuous visual environment before them when visual information enters the visual system through a series of brief glances interrupted by frequent rapid eye movements. Part of the answer may be that the visual system relies on a form of visual memory to allow for continuity between glances. In the present work, the research team will investigate how this type of visual memory operates. Achieving a better understanding of the basic operation of visual memory across eye movements could potentially lead to practical applications, such as optimized procedures for radiologists, baggage screeners, satellite image analysts, and others whose occupations require them to search for critical pieces of visual information during a visual search process. Another potential application could be optimized design of artificial vision systems. The proposed work aims to answer two basic questions regarding the operation of visual memory between glances. First, the research aims to investigate the capacity of such visual memory: How much visual information can this type of memory hold? Second, the research also aims to investigate how the capacity limitations of the visual memory mechanism constrain human performance in visual search tasks. For example, when the capacity of the visual memory is exceeded, how does this impact a person's ability to search for a specific piece of visual information that may be hidden among lots of other visual items? A set of rigorous computational and experimental techniques will be used to characterize the information contained within visual memory during visual search as well as to assess the impact of visual memory on visual search performance. The techniques will allow for detailed and novel quantitative predictions about how visual sensitivity and visual memory capacity interact to determine human performance in visual search tasks. Model predictions will be tested through psychophysical experiments.