Collaborative Proposal: Visual Attention in an Invertebrate Predator

合作提案：无脊椎动物捕食者的视觉注意力

• 批准号: 1656385
• 负责人: Ronald Hoy
• 金额: $ 36万
• 依托单位: Cornell University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-04-15 至 2021-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1656385&HistoricalAwards=false
• 关键词: Collaborative; Proposal; Visual; Attention; Invertebrate

All visual animals face the problem of distinguishing relevant from distracting stimuli quickly and efficiently. One way that visual systems accomplish this task is through selective attention, where an animal attends to a portion of the available visual stimuli at a time. Tiny jumping spiders have microminiature eyes that are nearly as acute as a human's and, like humans, pays attention and discriminates among visual targets. This project investigates how jumping spiders pay attention to and identify moving visual objects. The simplicity of the jumping spider eye and brain makes it easier to learn how the brain's neural networks "compute" visual movement and object identification, than in humans. In humans AND jumping spiders, selective visual attention is measured by rapid movements of their eyes to focus on targets of interest. This study employs an innovative eyetracker to measure the direction of a spider's gaze as it views video images. In addition, neural techniques are used to measure the activity of the spider's brain while its gaze is being monitored: specific parts of the brain respond differentially to particular images and sounds. Thus, it is possible to see both what engages a spider's visual attention as well as how different stimuli are interpreted. Understanding this elegant, miniaturized and extremely effective visual system will be of interest to roboticists and engineers, for whom micro-miniaturization of biosensors is a premium in small, self-autonomous robots. In addition, the PI will train a graduate student, a postdoc, and undergraduates, and will create videos for a project called "Faces and Voices of Science" that highlight the personal stories of researchers from different backgrounds. The videos will be made available online for teachers and professors to incorporate in lectures.Visual animals face the problem of distinguishing relevant from distracting stimuli quickly and efficiently. One way to do this is through selective attention, where the animal attends to only part of the visual field at a time, through a process of saccadic eye movements and selective target scan. This is true for humans and at the behavioral level seems true for jumping spiders. These spiders are highly visual and are among the rare invertebrate animals that have moveable eyes. The PIs have developed two novel technologies that allow a comprehensive study of selective visual attention. These include an innovative spider eyetracker can monitor with precision a spider's eye movements in real time as they scan a stimulus image and the first electrophysiological recordings in the brain of a living spider as it observes visual stimuli. The eyetracker will be deployed to monitor eye movements while simultaneously recordings from single units in the brain are recorded. The PIs will thus test explicit hypotheses about visual attention, eye movements, and correlated brain activity. These include "bottom-up" stimulus-driven attentional processes by testing how different visual stimuli influence eye movements and neural processes in the brain, as well as "top-down" processes by presenting cross-modal cues and measuring eye movements and neural processes as the spider searches for relevant stimuli among distractors.

所有视觉动物都面临着快速有效地区分相关刺激和干扰刺激的问题。视觉系统完成这一任务的一种方式是通过选择性注意，即动物一次关注可用视觉刺激的一部分。微小跳跃的蜘蛛有着几乎和人类一样敏锐的微型眼睛，像人类一样，注意并辨别视觉目标。这个项目调查跳跃蜘蛛是如何注意和识别移动的视觉对象的。与人类相比，跳跃蜘蛛的眼睛和大脑的简单使我们更容易了解大脑的神经网络是如何“计算”视觉运动和物体识别的。在人类和跳跃的蜘蛛中，选择性视觉注意力是通过眼睛快速移动来聚焦于感兴趣的目标来衡量的。这项研究使用了一种创新的眼镜蛇来测量蜘蛛观看视频图像时凝视的方向。此外，神经技术被用来测量蜘蛛凝视时大脑的活动：大脑的特定部分对特定的图像和声音做出不同的反应。因此，既可以看到蜘蛛的视觉注意力是什么，也可以看到不同的刺激是如何解释的。了解这种优雅、微型化和极其有效的视觉系统将引起机器人专家和工程师的兴趣，对他们来说，生物传感器的微型化是小型自主机器人的溢价。此外，PI将培训一名研究生、一名博士后和本科生，并将为一个名为“科学的面孔和声音”的项目制作视频，突出来自不同背景的研究人员的个人故事。这些视频将在网上提供，供教师和教授在课堂上使用。视觉动物面临着快速有效地区分相关刺激和分散注意力的刺激的问题。做到这一点的一种方法是通过选择性注意，通过眼跳运动和选择性目标扫描的过程，动物一次只关注视野的一部分。这对人类是正确的，在行为层面上，跳跃的蜘蛛似乎也是如此。这些蜘蛛具有高度的视觉效果，是罕见的眼睛可移动的无脊椎动物之一。PI已经开发了两项新技术，可以对选择性视觉注意进行全面研究。其中包括一种创新的蜘蛛眼镜蛇，它可以在扫描刺激图像时实时精确地监测蜘蛛的眼球运动，以及在观察视觉刺激时在活蜘蛛的大脑中首次进行电生理记录。Eyetracker将被部署用来监控眼睛的运动，同时记录大脑中单个单位的记录。因此，PI将测试关于视觉注意力、眼球运动和相关大脑活动的明确假设。这包括通过测试不同的视觉刺激如何影响眼睛运动和大脑中的神经过程来实现“自下而上”的刺激驱动的注意过程，以及通过呈现跨模式线索并在蜘蛛在分心物中搜索相关刺激时测量眼睛运动和神经过程来实现“自上而下”的过程。