Proj 6: Formal Models for the Neurodynamics of Decision-Making (p. 284 - 302)

项目 6：决策神经动力学的正式模型（第 284 - 302 页）

• 批准号: 7689952
• 负责人: JONATHAN D COHEN
• 金额: $ 13.6万
• 依托单位: PRINCETON UNIVERSITY
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/7689952
• 关键词: Address; Anterior; Area; Attention; Behavior; Behavioral; Biological Neural Networks; Cognitive; Collaborations; Complement; Condition; Conflict (Psychology); Data; Decision Making; Decision Modeling; Decision Theory; Diffusion; Episodic memory; Frequencies; Functional Magnetic Resonance Imaging; Goals; Hodgkin Disease; Human; Lead; Learning; Lip structure; Maps; Mediating; Mediation; Memory; Modeling; Monitor; Neurons; Neurophysiology - biologic function; Norepinephrine; Performance; Phase; Process; Psychological reinforcement; Range; Rate; Rewards; Role; Short-Term Memory; Source; Stimulus; Study models; System; Systems Theory; Testing; Time; Uncertainty; Work; abstracting; analytical method; base; behavior prediction; cingulate cortex; cost; locus ceruleus structure; memory recall; neuromechanism; neurophysiology; relating to nervous system; research study; selective attention; simulation; success; theories

This project develops, analyses, and tests neural network leaky accumulator/drift diffusion models of decision-making, perceptual choice, memory recall, and attention. The Center hypotheses will be operationalized in these models, and then numerical simulations and analytical methods of dynamical systems theory will be used to study the models, predict their behavior, and delineate parameter ranges that provide optional behaviors, and that best correspond to empirical data. Such models formalize and sharpen the major Center hypotheses, complementing the experimental work of Projects 1-5. The project will also work to connect these 'high level' models to biophysically-based models of heterogeneous neural pools. Success in this effort will lead to a more precise understanding of the mechanisms that govern the execution of cognitive tasks and their neural implementation.

本项目开发、分析和测试神经网络泄漏累加器/漂移扩散模型。 决策、知觉选择、记忆回忆和注意力。中心假设将是 在这些模型中可操作，然后是动力学的数值模拟和分析方法 将使用系统论来研究模型，预测其行为，并划定参数范围 它们提供了可选行为，且最符合经验数据。这样的模型形式化和 强化主要的中心假设，补充项目1-5的实验工作。该项目 我还将致力于将这些高级模型与基于生物物理学的异类神经模型联系起来 泳池。这一努力的成功将使人们更准确地理解管理 认知任务的执行及其神经实现。