Neural Mechanism of Attentional Guidance During Search Through Complex Scenes

描述(由申请人提供)： 人类和猴子都依靠视觉来寻找食物，寻找有吸引力的配偶，并发现危险。此外，我们经常在多变和复杂的环境中实现这些目标。前面提到的每一项任务--以及我们日常生活中的其他几十项任务--都是一项视觉搜索任务。在这些任务中，我们通过扫描杂乱的视觉场景来定位和识别感兴趣的对象。虽然认知心理学的研究已经逐渐明确了场景表征对搜索的影响程度，但灵长类动物场景感知背后的神经生理机制仍然知之甚少。具体地说，目前还不清楚场景是如何在大脑中编码的，以及这些表征如何影响搜索过程中的注意力引导。这项资助旨在通过结合使用行为测试和生理记录来揭示场景搜索背后的神经生理机制，这些测试和生理记录是由注意力计算模型的预测提供信息的。生理学实验将集中在与物体处理(下颞叶皮质，IT)和注意控制(外侧顶内区，LIP)有关的高级视觉区域。因此，本项目的具体目标是：1)了解颞叶下部皮质在场景表征中的作用；2)检验信息技术中与场景相关的活动作为一种生理机制的假设，通过该机制，场景语境线索可以偏向嘴唇上的活动，并引导视觉搜索。这些研究将为复杂场景环境的神经表征以及两个相互作用以产生高级视觉的大脑区域之间的交流提供重要的新见解。这项研究的长期目标是了解大脑如何有效地处理现实世界中遇到的信息，以便潜在地恢复或增强视觉系统受损的个人的知觉处理。

DESCRIPTION (provided by applicant): Both humans and monkeys rely on visual perception to seek food, find attractive mates, and spot danger. Moreover, we often accomplish these goals in a varied and complex environment. Each of the aforementioned tasks - and dozens of others in our everyday lives - is a visual search task. In these tasks, we scan through a cluttered visual scene to locate and identify objects of interest. While the degree to which scene representation affects search has progressively become clearer from research in cognitive psychology, the neurophysiological mechanisms that underlie scene perception in primates are still poorly understood. Specifically, it is still unknown how scenes are encoded in the brain and how these representations affect attentional guidance during search. This grant aims to reveal the neurophysiological mechanisms that underlie scene search by using a combination of behavioral testing and physiological recording informed by predictions from computational models of attention. Physiological experiments will focus on high level visual areas involved in both object processing (inferior temporal cortex, IT) and in attentional control (the lateral intraparietal area, LIP). Therefore, the specific aims of this project are to: 1) Understand the role of the inferior temporal cortex in scene representation; and 2) Test the hypothesis that scene-related activity in IT serves as a physiological mechanism through which scene contextual cues can bias activity in LIP and guide visual search. These studies will provide important new insights into both the neural representation of complex scene environments and the communication between two brain regions that interact to produce high level vision. The long term goal of this research is to understand how the brain efficiently processes information encountered in the real world so as to potentially restore or augment perceptual processing in individuals with compromised visual systems.