Hippocampal and prefrontal contributions to memory integration

海马和前额叶对记忆整合的贡献

• 批准号: 8480390
• 负责人: Alison R Preston
• 金额: $ 37.47万
• 依托单位: UNIVERSITY OF TEXAS AT AUSTIN
• 依托单位国家: 美国
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-04-17 至 2018-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8480390
• 关键词: Accounting; Address; Algorithms; Behavior; Behavioral; Biological Neural Networks; Brain; Code; Cognitive; Data; Development; Event; Foundations; Functional Magnetic Resonance Imaging; Hippocampus (Brain); Human; Individual; Influentials; Intervention; Judgment; Knowledge; Lead; Learning; Link; Maintenance; Measures; Medial; Mediating; Memory; Mental disorders; Neurobiology; Neurons; Nodal; Pattern; Play; Population; Prefrontal Cortex; Process; Recruitment Activity; Relative (related person); Research; Resolution; Retrieval; Role; Signal Transduction; Temporal Lobe; Testing; Therapeutic; Update; Vertebral column; Work; adjudicate; base; benefit sharing; effective intervention; effective therapy; experience; flexibility; improved; nervous system disorder; neuroimaging; novel; public health relevance; relating to nervous system; response; success; theories

DESCRIPTION (provided by applicant): Leading memory theories emphasize that new learning occurs on the background of existing knowledge. Retrieving prior knowledge during new experiences allows new information to be integrated into existing memories, resulting in the formation of rich, cohesive memory networks that relate discrete events. This integration process is proposed to facilitate new learning and enable memories to extend beyond direct experience to anticipate the relationships among events. However, memory for evidence neurobiological inability to directly measure the contents of reactivated memories during new experiences. To address this critical gap, the proposed studies employ a new experimental paradigm that uses highly sensitive pattern classifier algorithms applied to functional neuroimaging data to quantify incidental memory reactivation during new event encoding. Quantifying memory reactivation allows us to test mechanistic predictions about how past memories influence learning in the present. Aim 1 will use this paradigm to test the hypothesis that hippocampus and ventromedial prefrontal cortex (VMPFC) work in concert to support memory integration during new learning. We propose that by linking new information with well-established memories, this hippocampal-VMPFC mediated encoding process improves new learning and enables novel judgments about relationships among distinct events. Aim 2 will examine how temporal context and memory strength influence the formation of integrated memory traces for related events. We propose that learning overlapping events within the same temporal context facilitates memory integration by enhancing memory reactivation and recruiting hippocampal-VMPFC encoding processes. We will also adjudicate between opposing theoretical perspectives of learning that make competing predictions for whether strong or weak memories lead to enhanced memory integration. Aim 3 will use high- resolution fMRI focused on the medial temporal lobe to determine the precise hippocampal computations and coding strategies that underlie memory integration. We will determine the relationship between memory reactivation and hippocampal mismatch responses that signal differences between current events and existing memories to test the hypothesis that mismatch responses trigger memory integration. We will also use pattern- information analysis to test the hypothesis that the hippocampus creates integrated memories by forming overlapping neural codes for related events. Collectively, this work will determine how internally generated content influences new learning and will isolate the precise neural networks, computations, and coding strategies that underlie memory integration. Understanding how the brain uses prior experience to make sense of new information will lay the foundation for translational work onintegration and its functional significance is sorely lacking due primarily to an effective for interventions therapeutic psychiatric and neurological disorders that require acquisition and maintenance of new behaviors.

描述（由申请人提供）：领先的记忆理论强调新的学习发生在现有知识的背景下。在新的经历中检索先前的知识可以将新的信息整合到现有的记忆中，从而形成与离散事件相关的丰富的、有凝聚力的记忆网络。提出这种整合过程是为了促进新的学习，并使记忆能够超越直接经验来预测事件之间的关系。然而，记忆证据神经生物学无法直接测量新经历期间重新激活的记忆的内容。为了解决这一关键差距，拟议的研究采用了一种新的实验范式，该范式使用应用于功能神经影像数据的高度敏感的模式分类器算法来量化新事件编码期间的偶然记忆重新激活。量化记忆重新激活使我们能够测试有关过去记忆如何影响当前学习的机械预测。目标 1 将使用此范例来检验海马体和腹内侧前额叶皮层 (VMPFC) 协同工作以支持新学习过程中的记忆整合的假设。我们提出，通过将新信息与已建立的记忆联系起来，这种海马-VMPFC 介导的编码过程可以改善新的学习，并能够对不同事件之间的关系做出新的判断。目标 2 将研究时间背景和记忆强度如何影响相关事件的整合记忆痕迹的形成。我们提出，在同一时间背景下学习重叠事件可以通过增强记忆重新激活和招募海马-VMPFC 编码过程来促进记忆整合。我们还将在对立的学习理论观点之间进行裁决，这些理论观点对强记忆或弱记忆是否会增强记忆整合做出相互竞争的预测。目标 3 将使用专注于内侧颞叶的高分辨率功能磁共振成像来确定作为记忆整合基础的精确海马计算和编码策略。我们将确定记忆重新激活和海马失配反应之间的关系，该关系表明当前事件和现有记忆之间的差异，以检验失配反应触发记忆整合的假设。我们还将使用模式信息分析来检验海马体通过形成相关事件的重叠神经代码来创建整合记忆的假设。总的来说，这项工作将确定内部生成的内容如何影响新的学习，并将隔离作为记忆整合基础的精确神经网络、计算和编码策略。了解大脑如何利用先前的经验来理解新信息将为整合转化工作奠定基础，其功能意义严重缺乏，主要是因为对需要获得和维持新行为的精神和神经疾病的干预治疗有效。