Implementation of dynamic threat belief updating within cerebro-spinal systems and opioidergic pathways

背景：学习预测威胁建立在厌恶，伤害，结果的整合上，形成威胁信念。然而，当环境发生变化时，威胁学习也捕获了已建立的威胁信念的动态更新。为了描述形成和动态信念更新（DynBU），建立了一个统计学习的神经计算模型。然而，一个统计学习框架来描述动态威胁学习的神经生物学机制是缺失的。目的：通过解读脊髓的伤害性结果加工、脑干介导的觉醒以及前脑回路的威胁信念编码之间的相互作用，揭示威胁信念动态更新的神经基础。本项目旨在进一步描述动态威胁信念更新的神经药理学和阿片能机制。假设：我们假设(1)在伤害性、厌恶结果的编码（如脊髓）和脑干介导的觉醒（如蓝斑，LC）之间存在相互作用，后者通过已建立的大脑回路（如杏仁核和背前扣带皮层，dACC）控制威胁信念的更新。特别是，我们假设(2)，无论是近端还是远端防御威胁反应的协调，都需要脑干介导的觉醒（如LC）和编码威胁邻近的大脑区域（导水管周围灰质，PAG）之间的功能连接。我们期望(3)阿片受体阻断可以增强厌恶结果整合，这可以在学习模型中模拟。最后，我们假设(4)阿片受体阻断对厌恶结果整合的增强将在PAG、杏仁核和LC之间的相互作用中实现。计划方法：参与者将接受一个变化点任务，该任务要求对近端和远端捕食者进行更快和更慢的防御（即“捕食者任务”，与项目8相协调），同时获得脑脊髓功能磁共振成像。阿片受体对捕食者任务的影响将首先在第二个样本中进行模拟和测试，并进行大脑功能磁共振成像。预期影响：该项目将提供潜在威胁的DynBU的心理、神经和计算机制，从而对该RU的主要目标做出重大贡献。具体来说，我们将通过威胁学习的机制来提出统计学习的计算和神经生物学模型。该项目将进一步通过识别控制人类威胁信念更新的神经精神药理学机制提供独特的贡献。与项目8一起，我们将进一步能够将健康志愿者的结果转化为焦虑表型。此外，与其他项目的独特联系允许在威胁和压力影响的统计学习（项目3）或从奖励结果的统计学习（项目4）之间架起桥梁。

Background: Learning to predict threats builds on integration of aversive, nociceptive, outcomes to form a threat belief. Yet, threat learning also captures dynamic updating of established threat beliefs, when the environment changes. To characterize both, formation and dynamic belief updating (DynBU), a neurocomputational model for statistical learning has been established. Yet, a statistical learning framework to delineate the neurobiological mechanisms of dynamic threat learning is missing. Aims: The overarching aim is to delineate the neural basis of dynamic updating of threat beliefs by deciphering the interplay between nociceptive outcome processing in the spinal cord, brain-stem mediated arousal, together with encoding of threat beliefs in forebrain circuits. This project further aims to delineate a neuropharmacological, opioidergic mechanism of dynamic threat belief updating.Hypotheses: We hypothesize (1) an interaction between encoding of nociceptive, aversive outcomes (i.e., spinal cord) and brainstem-mediated arousal (e.g., locus coeruleus, LC), which gates updating of threat beliefs via established brain circuits for threat processing (e.g., amygdala and dorsal anterior cingulate cortex, dACC). In particular, we hypothesize (2) that orchestration of defensive threat responses that are either proximal or distal requires a functional connection between brainstem mediated arousal (e.g., LC) and brain regions that encode proximity of threats (periaqueductal gray, PAG). We expect (3) that opioid-receptor blockade enhances aversive outcome integration, which can be simulated in learning models. Finally, we hypothesize (4) that this enhancement of aversive outcome integration by opioid-receptors blockade will be implemented in interactions between the PAG, amygdala and LC. Planned methods: Participants will a undergo a change-point task that requires faster and slower defense against proximal vs. distal predators (i.e., “predator task”, harmonized with project 8), while acquiring cerebro-spinal fMRI. Opioid-receptor effects on the predator task will first be simulated and tested in a second sample that undergoes brain fMRI. Expected impact: This project will provide psychological, neural, and computational mechanisms underlying DynBU of threats and thereby make a significant contribution to the major aims of this RU. Specifically, we will advance the proposed computational and neurobiological model of statistical learning by mechanism of threat learning. This project will further provide a unique contribution by identification of a neuropsychopharmacological mechanism that controls updating of threat beliefs in humans. Together with project 8, we will further be able to translate the results in healthy volunteers to the anxiety phenotype. Furthermore, distinct links with other projects allow to bridge between statistical learning of threats and the effects of stress (project 3) or statistical learning from rewarding outcomes (project 4).