Neurobiology and dynamics of Active Sensing

主动感知的神经生物学和动力学

基本信息

批准号：
10175032
负责人：
CHARLES E SCHROEDER
金额：
$ 177.35万
依托单位：
COLUMBIA UNIVERSITY HEALTH SCIENCES
依托单位国家：
美国
项目类别：
财政年份：
2017
资助国家：
美国
起止时间：
2017-04-15 至 2023-03-31
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/10175032
关键词：
Address Area Attention Attention Deficit Disorder Auditory Behavioral Brain Cells Code Cognitive Communication Computer Models Data Electric Stimulation Electroencephalography Emotional Environment Epilepsy Evolution Eye Eye Movements Frequencies Functional Magnetic Resonance Imaging Functional disorder Goals Human Image Individual Link Measures Modeling Monkeys Motor Neurobiology Neurons Neurosciences Operative Surgical Procedures Parietal Pathway Analysis Periodicity Physiological Processes Population Process Property Research Research Personnel Resources Role Route Sample Size Sampling Scanning Schizophrenia Sensory Source Statistical Models System Thalamic structure Update Work autism spectrum disorder biophysical properties cell assembly cell type cognitive process computational network modeling covert attention data format density expectation experience experimental study falls gaze indexing information gathering insight instrument neuropsychiatric disorder neuropsychiatry nonhuman primate novel operation outreach programs relating to nervous system sensory input theories

项目摘要

ABSTRACT: Two key principles define the core conceptual framework of our Conte Center. First, most sensory input is actively acquired by a motor and/or attentional sampling routine; e.g., rather than staring blankly and hoping that something will “fall” into our gaze, we Actively Scan the visible environment with eye movements. Even when fixating, we can actively (albeit covertly) scan the environment by shifting attention. Corresponding “scanning” of the auditory environment uses the more covert attentional sampling strategy, but is no less active. As a result, Active Sensing (i.e., strategic, goal-driven sampling of inputs) is “predictive” in that, it is guided by the subject's expectations (theories, models), accumulated through species' evolution, and refined by individuals' experience. Its central tenet is that sensing and perceiving can be fully understood only in the context of subjects' ongoing, goal-directed information-gathering activities. Second, neuronal oscillatory dynamics are critical mechanistic components of normal brain operation. Neuronal oscillations reflect rhythmic fluctuations of neuron ensembles between high and low excitability states. Mounting evidence indicates that such rhythmic activity is essential to normal brain operations, and that its disruption contributes to neuropsychiatric disorders. The idea that Active Sensing incorporates neuronal rhythms as fundamental instruments of operation represents an ongoing paradigm shift in systems neuroscience. Our Center is unified by support Cores and a set of mechanistic (linking) hypotheses concerning the “instrumental” functions of neuronal rhythms at local and network scales. The Center integrates electrocorticographic (ECoG) studies in humans with intracortical recordings in monkeys and computational modeling. Our Specific Aims are: AIM 1 – Exploit ECoG's strengths of distributed sampling and direct human brain recording to define dynamical circuits of top-down control and coordination across cortical areas in Active Sensing. To gain a sample size appropriate for our purposes, we will pool subjects across 5 surgical epilepsy centers using a common set of Active Sensing tasks, and a common data format. AIM 2 – Use recordings in nonhuman primates to elucidate and extend ECoG findings in humans. Laminar field potential (FP), current source density (CSD) and multiunit activity (MUA) profiles, along with single unit recordings will be obtained from monkeys performing tasks identical to those studied in humans. AIM 3 – Develop iterative interactions between computational and empirical studies of circuit dynamics at local (cell assembly) and global (brain network) levels. Tracking specific neuronal dynamics from the global-network level in humans down to the cellular and cell ensemble levels in monkeys will yield novel and unique insights into mechanisms of active brain operation. Statistical and Computational modeling will allow rapid exploration of possibilities suggested by ECoG and related multielectrode studies in monkeys, and will help in building accurately representing and integrating our findings across local and global scales.

摘要：两个关键原则定义了我们孔蒂中心的核心概念框架。感觉输入是通过运动和/或注意力采样程序主动获取的，例如，而不是凝视；我们茫然地希望有什么东西“落入”我们的视线，我们主动地用眼睛扫描可见的环境即使在注视时，我们也可以通过转移注意力来主动（尽管是秘密地）观察环境。相应的听觉环境“扫描”使用了更隐蔽的注意力采样策略，但是因此，主动感知（即战略性的、目标驱动的输入采样）在以下方面具有“预测性”。它以主体的期望（理论、模型）为指导，通过物种的进化积累，并且它的核心原则是只有感知和感知才能被充分理解。在受试者正在进行的、以目标为导向的信息收集活动的背景下，第二，神经振荡。动力学是正常大脑运作的关键机制组成部分。越来越多的证据表明神经系统在高兴奋性状态和低兴奋性状态之间存在节律性波动。表明这种有节奏的活动对于正常的大脑运作至关重要，并且它的破坏有助于主动感知将神经节律作为基础的想法。操作工具代表了系统神经科学正在进行的范式转变，我们的中心是统一的。通过支持核心和一组有关“工具”功能的机械（链接）假设该中心整合了局部和网络尺度的神经元节律研究。我们的具体目标是：AIM 1 – 利用 ECoG 的分布式采样和直接人脑记录的优势来定义动态电路主动感知中皮层区域自上而下的控制和协调以获得样本量。适合我们的目的，我们将使用一组通用的方法汇集 5 个癫痫外科中心的受试者主动传感任务和通用数据格式 AIM 2 – 使用非人类灵长类动物的记录来阐明。并扩展人类层流场电位 (FP)、电流源密度 (CSD) 和多单元的 ECoG 研究结果。活动（MUA）概况以及单个单元记录将从执行任务的猴子中获得与人类研究的相同。 AIM 3 – 开发计算和计算之间的迭代交互。局部（细胞组装）和全局（大脑网络）水平的电路动力学实证研究跟踪特定。从人类的全局网络水平到细胞和细胞群体水平的神经动力学猴子将对活跃的大脑运作机制产生新颖而独特的见解。计算模型将允许快速探索 ECoG 和相关的建议的可能性在猴子身上进行多电极研究，将有助于准确地表达和整合我们的发现跨越本地和全球范围。