Recurrent Circuit Model of Neural Response Dynamics in V1

V1 中神经反应动力学的循环电路模型

基本信息

批准号：
10710967
负责人：
DAVID J HEEGER
金额：
$ 47.41万
依托单位：
NEW YORK UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2023
资助国家：
美国
起止时间：
2023-09-01 至 2027-05-31
项目状态：
未结题

项目摘要

Project Summary/Abstract Primary visual cortex (V1) is one of the most studied areas of the cerebral cortex, but we lack a theoretical framework for a comprehensive understanding of V1 neurophysiology. Through the proposed research, we aim to provide one. A class of circuit models, called Oscillatory Recurrent Gated Neural Integrator Circuits (ORGaNICs), simulates many key neurophysiological phenomena. Our goal is to develop a theory for the full range of neurophysiological phenomena in V1, i.e., a single predictive model with biophysically-realistic parameters, and to test that theory with previously published datasets acquired with a wide range of methodologies. Preliminary results demonstrate predictions of the theory commensurate with experimental observations of V1 response dynamics (including onset transients and the stimulus-dependence of gamma oscillations), the dynamics of attentional modulation, experimental evidence for recurrent amplification and inhibitory stabilization, experimental observations about adaptation including tuning changes and decorrelation, noise quenching, the dependence of noise correlations on similarity in orientation preference and attention, and psychophysical contrast discrimination. Aim 1 key contributions: 1) an analytical theory (i.e., closed-form expressions) that makes experimentally- testable predictions about a wide range of phenomena related to the dynamics of V1 activity; 2) closed-form expressions derived from the theory for LFP power spectra; 3) a novel explanation for oscillatory activity in visual cortex. Aim 2 key contributions: 1) an analytical theory of adaptation in V1 that makes experimentally-testable predictions about a wide range of neurophysiological phenomena related to adaptation; 2) the demonstration that adaptation maintains an efficient neural code, subject to finite resources (overall activity in the circuit), despite dynamically changing stimulus statistics. Aim 3 key contributions: 1) an analytical theory that makes experimentally-testable predictions about the variability and covariability of neural responses; 2) experimentally-testable predictions about psychophysical discrimination. The proposed research has the potential to be transformative. We will provide a new set of analytical results and computational tools for characterizing a broad range of neural circuit models, which will have a significant impact on the analysis of experimental data and experimental design, and will make new experimentally- testable predictions for both ORGaNICs and alternative models. We will provide a roadmap for understanding the underlying circuit mechanisms (the cell types, their interconnections and biophysics), and how manipulating those mechanisms may change circuit function to correct disorders of visual perception and attention.

项目概要/摘要初级视觉皮层（V1）是大脑皮层研究最多的区域之一，但我们缺乏理论依据全面了解 V1 神经生理学的框架。通过拟议的研究，我们的目标是提供一个。一类电路模型，称为振荡循环门控神经积分电路（ORGaNICs），模拟许多关键的神经生理现象。我们的目标是发展一种理论 V1 中的一系列神经生理学现象，即具有生物物理现实性的单一预测模型参数，并用先前发布的数据集来测试该理论方法论。初步结果证明理论预测与 V1 的实验观察相一致响应动力学（包括瞬态开始和伽玛振荡的刺激依赖性），注意力调节的动力学，循环放大和抑制的实验证据稳定性、关于适应的实验观察，包括调谐变化和去相关、噪声淬灭，噪声相关性对方向偏好和注意力相似性的依赖性，以及心理物理对比歧视。目标 1 的主要贡献：1）一种分析理论（即封闭式表达式），可以通过实验- 对与 V1 活动动态相关的各种现象的可测试预测； 2) 封闭式由 LFP 功率谱理论导出的表达式； 3）对振荡活动的新颖解释视觉皮层。目标 2 主要贡献：1）V1 中的适应分析理论，使得实验可测试对与适应相关的各种神经生理现象的预测； 2）演示适应在有限资源（电路中的整体活动）的情况下维持有效的神经代码，尽管刺激统计数据不断变化。目标 3 个关键贡献：1）一种分析理论，可以对神经反应的变异性和协变性； 2）关于心理物理学的可实验检验的预测歧视。拟议的研究具有变革性的潜力。我们将提供一组新的分析结果和计算工具来表征广泛的神经回路模型，这将具有重大意义影响实验数据的分析和实验设计，并将使新的实验- 对有机集成电路和替代模型的可测试预测。我们将提供理解路线图潜在的电路机制（细胞类型、它们的互连和生物物理学），以及如何操纵这些机制可能会改变回路功能，以纠正视觉感知和注意力障碍。