Mechanisms of Visual Motion Perception

描述（申请人提供）：视觉处理面临整合和分割两种相互冲突的需求。固有噪声的视觉信号需要进行整合，而从视觉输入的时空变化中提取重要信息则需要进行分割。了解这两种机制之间的相互作用将揭示视觉处理的基本原理，并使我们能够深入了解分割过程的功能角色。在运动感知方面，PI最近的工作表明，运动信号的空间整合不是固定的，而是严重依赖于对比度等基本视觉因素，随着刺激可见性的增加，空间整合让位于空间抑制。本研究的总体目标是研究运动信号自适应整合/分离的神经机制，并阐明其在运动背景中物体分割中的作用。空间抑制的关键特性是对大的、高对比度刺激的运动感知受损。在目的1中，我们将确定这种抑制机制是否以及如何影响视觉和动眼肌加工。回答这个问题将约束空间抑制的可能神经相关性，并与目标2一起，为MT区域的空间抑制与周围抑制联系起来的假设提供检验。证实这一联系将允许将空间抑制和运动分离（目标2和3）之间的联系归因于MT周围抑制的参与。在目标2中，我们将通过颅磁刺激损伤颞叶和早期视觉区域的加工来寻找空间抑制的神经相关性的直接证据。我们预计，在空间抑制中至关重要的神经机制的破坏将允许通常被抑制的运动信号到达感知。同时，我们还将确定损害空间抑制的相同刺激是否也会破坏运动分离。在Aim 3中，我们验证了空间抑制通过抑制背景运动信号直接实现运动物体快速分离的假设。在这里，空间抑制在运动分离中的作用被定义为一个粗糙但快速的，基于区域的分割过程。这一假设预测，空间抑制的变化（以及背景运动的可见性）应该预测运动分离的相应变化，反之亦然。利用不同的实验方法，我们将通过利用刺激操作、个体差异和感知学习来产生空间抑制或运动隔离的变化来测试这一预测。其中一个重点将是老年人，他们表现出较弱的空间抑制能力。我们将确定这种异常是否预测运动分离缺陷，以及是否可以通过运动分离的感知学习来逆转与年龄相关的空间抑制缺陷。

DESCRIPTION (provided by applicant): Visual processing faces two conflicting demands: integration and segmentation. Integration is required by inherently noisy visual signals, while segmentation is needed to extract vital information from spatiotemporal variations in visual input. An understanding of the interplay between these two mechanisms will reveal fundamentals of visual processing and also enable insights into functional roles of segmentation processes. In motion perception, the PI's recent work demonstrated that spatial integration of motion signals is not fixed, but critically depends on basic visual factors such as contrast, with spatial integration giving way to spatial suppression as stimulus visibility increases. The overarching goal of this proposal is to investigate the neural mechanisms involved in this adaptive integration/segregation of motion signals, and to elucidate their role in the segmentation of objects from moving backgrounds. The key property of spatial suppression is impaired motion perception of large, high-contrast stimuli. In Aim 1, we will determine if and how this suppressive mechanism affects visual and oculomotor processing. Answering this question will constrain possible neural correlates of spatial suppression and, along with Aim 2, provide a test for the hypothesis linking spatial suppression to surround suppression in area MT. Substantiating this link will allow the attribution of links between spatial suppression and motion segregation (Aims 2 & 3) to the involvement of MT surround suppression. In Aim 2, we will seek direct evidence about neural correlates of spatial suppression by impairing processing in MT and early visual areas with TMS. We expect that a disruption of neural mechanisms critically involved in spatial suppression will allow normally suppressed motion signals to reach perception. Concurrently, we will also determine whether the same stimulation that impairs spatial suppression also disrupts motion segregation. In Aim 3, we test the hypothesis that spatial suppression directly enables rapid segregation of moving objects by suppressing background motion signals. Here, the role of spatial suppression in motion segregation is conceptualized as a coarse, but rapid, region-based segmentation process. This hypothesis predicts that variations in spatial suppression (and, thus, in the visibility of background motion) should predict corresponding changes in motion segregation and vice versa. Exploiting different experimental approaches, we will test this prediction by utilizing stimulus manipulations, individual differences and perceptual learning to produce variations in either spatial suppression or motion segregation. One focus will be on older adults, who are known to exhibit weak spatial suppression. We will determine whether this abnormality predicts motion segregation deficits and whether age-related deficits in spatial suppression can be reversed by the perceptual learning of motion segregation. PUBLIC HEALTH RELEVANCE: The knowledge of mechanisms underlying spatial suppression and motion segregation will contribute to the understanding of age-related changes in these two basic visual processes. Moreover, understanding how these age-related deficits can be alleviated through perceptual learning might lead to the development of viable interventions.