The cognitive mechanisms of complex planning

复杂规划的认知机制

• 批准号: 10704613
• 负责人: Wei Ji Ma
• 金额: $ 27.24万
• 依托单位: NEW YORK UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-14 至 2024-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10704613
• 关键词: Algorithms; Behavior; Behavioral; Behavioral Paradigm; Cognitive; Collaborations; Complex; Computer Models; Computing Methodologies; Data; Decision Making; Decision Trees; Effectiveness of Interventions; Electrophysiology (science); Eye; Eye Movements; Functional Magnetic Resonance Imaging; Funding; Future; Goals; Human; Impairment; Individual; Length; Lesion; Life; Location; Measurement; Measures; Mental disorders; Methods; Modeling; Mydriasis; Narration; Obsessive-Compulsive Disorder; Patients; Performance; Play; Positioning Attribute; Process; Psyche structure; Publishing; Research; Sports; Testing; Text; Thinking; Time; Trees; United States National Institutes of Health; Work; Writing; autism spectrum disorder; behavior measurement; clinically relevant; combinatorial; design; experience; experimental study; heuristics; improved; individuals with autism spectrum disorder; laboratory experiment; nervous system disorder; neural; neuroimaging; neuromechanism; nonhuman primate; novel; power analysis; programs; simulation; timeline; verbal

PROJECT SUMMARY/ABSTRACT The goal of this project is to lay the groundwork for understanding the neural basis of complex planning. Planning, defined as sequential decision-making that involves mental simulation of potential futures, is crucial for the organization of our behavior in everyday life — from navigation to playing sports or writing a long text. Real-world planning is often “complex”, in the sense that there is a explosively large number of possible futures and the decision-maker has to think multiple steps ahead. By contrast, studies of human planning typically use simple tasks, in which the number of possible states is low and thinking ahead is barely necessary. To serve as a suitable behavioral paradigm to study complex planning, a task should meet multiple criteria: it should require thinking ahead, it should be novel to subjects, it should have simple rules, and it should allow for computational modeling in order to disentangle component processes. We previously developed a behavioral paradigm that satisfies these requirements, as well as a computational process model of choices in this task based on a heuristic value function and partial tree search. This model can be used to estimate depth of planning (EDOP). The goals of the present proposal are two-fold: to prepare the model for use in future neural studies by establishing the construct validity of EDOP (Aim 1), and to go beyond choice data to probe the dynamics of complex planning using eye movements made while a choice is being prepared (Aim 2). Although this work does not have direct clinical relevance, it could in the future serve to improve the behavioral and neural characterization of deficits in planning, as well as the effectiveness of interventions. Planning is disrupted in many neurological and psychiatric disorders. For example, performance on planning tasks is impaired in individuals with Obsessive Compulsive Disorder, Autism Spectrum Disorder, and prefrontal lesions.

项目摘要/摘要 该项目的目的是为理解复杂的神经基础奠定基础 规划。计划，定义为顺序决策，涉及对 潜在的未来，对于我们日常生活中我们行为的组织至关重要 - 导航进行运动或写长文字。现实世界的计划通常是“复杂的” 感觉到有大量可能的未来，决策者必须 考虑多个步骤。相比之下，对人类计划的研究通常使用简单的任务 可能的状态数量很少，并且几乎没有必要。作为 一个合适的行为范式研究复杂的计划，任务应符合多个标准：它 应该需要提前思考，它应该是新颖的主题，它应该有简单的规则，并且 应该允许计算建模以解开组件进程。我们 以前开发了一个令人满意的行为范式，以及 根据启发式值函数和 部分树搜索。该模型可用于估计计划深度（EDOP）。目标 目前的提案是两个方面的：准备模型，以便在未来的神经研究中使用 建立EDOP的结构有效性（AIM 1），并超越选择数据以探测 准备选择时要进行选择的复杂计划的动态 （AIM 2）。尽管这项工作没有直接的临床相关性，但将来可能有 改善计划中缺陷的行为和神经表征以及 干预的有效性。在许多神经和精神病学中，计划中断了 疾病。例如，在计划任务方面的表现受到 强迫症，自闭症谱系障碍和前额叶病变。