Computational dynamics in neural populations of freely foraging vs. restrained monkeys

自由觅食与受限制猴子神经群体的计算动力学

基本信息

批准号：
10447347
负责人：
Dora Angelaki
金额：
$ 282.8万
依托单位：
NEW YORK UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2022
资助国家：
美国
起止时间：
2022-04-15 至 2025-03-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10447347
关键词：
3-Dimensional Address Algorithms Animals Area Artificial Intelligence BRAIN initiative Behavior Behavioral Behavioral Model Belief Brain Code Communication Complex Data Decision Making Dimensions Environment Eye Eye Movements Food Future Goals Head Hippocampal Formation Hippocampus (Brain)Joystick Laboratories Learning Macaca Maintenance Mediating Memory Modeling Monitor Monkeys Neurons Neurosciences Parietal Parietal Lobe Pattern Population Prefrontal Cortex Process Psychiatry Psychological reinforcement Recurrence Resources Rewards Role Sensory Solid Strategic Planning System Testing Time Training Uncertainty Visual Work base computerized tools cost entorhinal cortex experimental study eye center flexibility free behavior frontal lobe gaze innovation neurophysiology relating to nervous system response restraint sensory input spatial memory theories tool virtual reality virtual reality environment way finding wireless

项目摘要

Summary This proposal will investigate the neural dynamics underlying three-dimensional foraging behavior, with three overarching goals. The first is to evaluate the neural computations of foraging in dynamic environments in naturalistic settings. The second is to compare the algorithmic behavioral computations and their neural substrates under naturalistic versus traditional laboratory settings. Most neuroscience studies assume that laboratory behavior has real-world implications, so any differences we observe would have remarkable consequences for the future of neuroscience. The third goal is to test the role of hippocampal formation, frontal, and parietal areas in the different components of foraging. We will compare three behaviors in two experimental setups. The behaviors are free-foraging for unpredictable, random rewards, a goal-driven foraging task with navigation among three resource patches under visual uncertainty and stochastic reward presentation, and foraging among three resource patches without navigation. The latter two tasks involve the formation and maintenance of internal models, memories, time-dependent sensory and reward contingencies, the costs of an animal’s own actions, and dynamically changing beliefs about the state of the world. The former two tasks will be performed during head-free navigation in a foraging room, and during head-fixed navigation in virtual reality. In both environments, we will record simultaneously from several mutually interconnected areas involved in visual navigation, spatial memory, path integration, and decision-making. Target areas include posterior parietal cortex, prefrontal cortex, retrosplenial cortex, entorhinal cortex, and hippocampus. We will use advanced behavioral models and theory to infer internal states and to identify their neural representation and interactions across a broad network of interconnected brain areas. In addition, we will regress neuronal activity against task-relevant variables to identify neural subspaces that encode them. Collectively, we expect that these experiments will rigorously illuminate the neural dynamics of foraging and spatial navigation behaviors and critically interrogate whether and how constrained laboratory conditions with head-fixed, restrained animals differ from ecological and naturalistic environments—a critical but still untested assumption of contemporary systems neuroscience.

概括该建议将研究三维觅食行为的神经动力学，三个总体目标。首先是评估在动态环境中觅食的神经计算自然设置。第二个是比较算法的行为计算及其中性自然主义与传统实验室环境下的底物。大多数神经科学研究都认为实验室行为具有现实世界的含义，因此我们观察到的任何差异都将具有显着的神经科学未来的后果。第三个目标是测试海马形成，正面，和觅食不同组成部分的顶叶区域。我们将在两个实验中比较三种行为设置。这些行为是自由锻造的，用于不可预测的随机奖励，这是目标驱动的觅食任务在视觉不确定性和随机奖励表现下的三个资源贴片之间的导航，以及在没有导航的三个资源补丁中觅食。后两个任务涉及编队和维护内部模型，记忆，与时间有关的感官和奖励意外事件，成本动物自己的行为和动态变化相信世界状况。前两个任务将在觅食室和虚拟现实中的头部固定导航期间，可以在无头部导航期间进行。在这两个环境中，我们都将仅记录来自视觉涉及的几个相互联系的区域导航，空间内存，路径集成和决策。目标区域包括后顶皮层，前额叶皮层，肾上腺后皮质，内嗅皮层和海马。我们将使用高级行为推断内部状态并确定其神经表示和相互作用的模型和理论互连大脑区域的广泛网络。此外，我们将回归与任务相关的神经元活动变量以识别编码它们的神经子空间。总体而言，我们希望这些实验将严格阐明觅食和空间导航行为的神经动态，并进行严重询问与生态和生态的固定，受限制的动物的约束实验室条件是否与生态和自然主义环境 - 当代系统神经科学的关键但未测试的假设。