Automatic and Attentional Components in Visual Space Perception

视觉空间感知中的自动和注意成分

• 批准号: 8706858
• 负责人: William Epstein
• 金额: $ 12.7万
• 依托单位: University of Wisconsin-Madison
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1987
• 资助国家: 美国
• 起止时间: 1987-09-01 至 1990-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8706858&HistoricalAwards=false
• 关键词: Automatic; Attentional; Components; Visual; Space

This grant for two years is on visual perception. The function of the visual system is to generate representations of the environment that will support adaptive behavior. It is important for certain "visual constancies" to be developed, so that for example a small object nearby is not mistaken for a larger object farther away. For an observer to compute a visual representation of the size of an object, both the projected size of the object and the viewing distance must be considered. Similarly, other "visual constancies" of objective shape, depth and velocity must involve considering features other than the projected view of the object. One explanation for considering these other features is that mental "attention" is required for adequate processing, in addition to some "automatic" processing components that work in the absence of attention (known as "pre-attentive" processing). This two-stage model suggests that two-dimensional properties (size, shape and position) do not require attention, but the 3- dimensional construct does depend on attention, operating on the 2-D representation. This project will assess the computational steps for producing such representations, to see whether a two- stage model of processing is required for 3-D analysis. Human observers will be presented with visual stimuli along with different kinds of attention-distractors. Then they will be given a test stimulus and asked if it is the same or different from the original. If a two-stage process occurs, distraction of attention is expected to cause false positive responses. Tests will concentrate on shape-at-a-slant, and size-at-a-distance. A second set of experiments will utilize search time necessary to distinguish such targets in a multiple array. Here longer search times are expected to reflect the longer processing required for attentive steps, compared to the pre-attentive steps. This research uses a novel approach, and has important implications for understanding visual space perception, and the field of computational vision. It is likely to have impact on concepts of the role of attention in perception in general, and also on design of artificial vision systems.

这两年的补助金用于视觉感知。 的功能 视觉系统将生成环境的表示 来支持适应性行为。 重要的是， “视觉恒定性”的发展，例如，一个小的 附近的物体不会被误认为是更远的物体。 为 观察者计算物体大小的视觉表示， 物体的投影尺寸和观看距离必须 考虑了 同样，客观形状的其他“视觉恒定性”， 深度和速度必须考虑的功能以外的 对象的投影视图。 考虑这些其他特征的一种解释是， 精神上的“注意力”需要足够的处理， 除了一些“自动”处理组件， 注意力的缺失（被称为“前注意”处理）。 这个两阶段模型表明， (size，形状和位置）不需要注意，但3- 维度结构确实依赖于注意力， 二维表示 该项目将评估计算 制作这些陈述的步骤，以确定两个- 3-D分析需要加工的阶段模型。 人类 观察者将被呈现视觉刺激沿着 不同类型的注意力分散物。 那么他们将会被 给一个测试刺激，并询问它是相同还是不同 从原来的。 如果发生两阶段过程， 预期注意力会引起假阳性响应。 测试 将专注于倾斜时的形状和远距离时的大小。 一 第二组实验将利用必要的搜索时间， 在多个阵列中区分这些目标。 这里更长的搜索 预计时间将反映以下方面所需的更长处理时间 注意的步骤，与预先注意的步骤相比。 这项研究采用了一种新的方法，具有重要的意义。 对理解视觉空间感知的影响，以及 计算机视觉领域。 它可能会影响到 注意力在一般感知中的作用的概念，以及 也是关于人工视觉系统的设计。