Computational psychiatry investigation of the role of unrealistic optimism in opioid use disorder and relapse

项目摘要/摘要 目前的阿片类药物流行是一种重大的健康危机，导致美国人的预期寿命减少。 导致病态和死亡的一个主要原因是阿片类药物在慢性病例中重复使用和复发。了解 因此，神经认知机制和潜在的重复使用脆弱性因素是迫切需要的。利用 神经认知工具和多疗程纵向设计的新组合，我们在阿片类药物使用方面的最新工作 无序(OUD)已经开始描绘阿片类药物再利用的精确决策机制，因为它表明 接受治疗的患者在表现出对未知事物的耐受性增加时，重复使用的风险更高 财务选择任务中的概率结果(模糊容忍度)(Konova等人，2019年《美国医学会精神病学》)。 但为什么患者在重新使用之前会变得更能容忍模糊的不确定性，这一点目前尚不清楚。 与决策理论一致的一个潜在解释是，在这些时期，他们变得过于乐观。 关于模棱两可的结果，这导致他们高估了好结果的可能性(或 低估了糟糕的结果)，当面临重复使用的决定时，因此更有可能这样做。在这里，我们 通过使用定义良好的量化方法对该框架进行多层次、收敛的测试 假定的“乐观主义偏见”，以及我们建议收集的不确定性容忍度的量化衡量标准 同时进行高分辨率功能磁共振成像记录，并与纵向临床评估挂钩。在目标1中，我们 目的建立不确定容忍度与乐观偏向之间的关系。 通过在一组选择和估计任务中研究这些行为来匹配对照(后者设计 为了抓住对简单财务和更复杂结果的乐观情绪，利用与药物选择相关的信息 例如健康结果等领域)。我们还通过不同的精神病理学检查潜在的缓和。 大量未选择的在线(MTurk)受试者的维度。在目标2中，我们收集了fMRI数据 同样的选择和估计任务来描绘乐观的神经表征的机制 不确定性可能会推动人们对这种不确定性的行为容忍，并在OUD中重复使用。在目标3中，我们使用多个- 会话纵向设计理解乐观主义偏向和不确定性容忍度之间的相互作用 它们涉及阿片类药物的再利用、会期到会期，使我们能够阐明具体的时间尺度和性质。 互动。通过这个项目，我们的目标是提供一个答案，为什么患者在 在重用之前的几个时期，并希望通过这样做来揭示一种上游机制(以乐观偏见为中心) 这个漏洞，包括它的神经实现。除了这一概念上的进步，这项工作还将 提供一套新的认知工具，以准确和客观地测量这些过程，并有可能 以一种易于在临床环境中实施的方式预测糟糕的结果，如重复使用。最后， 这项工作的发现将通过提供精确的神经认知靶点，为新的治疗干预提供信息，如 以及它们的理想时机，目标是降低重复使用风险和改善患者的长期结果。

Project Summary/Abstract The current opioid epidemic is a major health crisis that has contributed to decreased life expectancy in the U.S. A main cause of morbidly and mortality is opioid reuse and relapse in chronic cases. Understanding the neurocognitive mechanisms and factors underlying reuse vulnerability is thus a pressing need. Leveraging a novel combination of neurocognitive tools and a multi-session longitudinal design, our recent work in opioid use disorder (OUD) has begun to delineate a precise decision making mechanism for opioid reuse by showing that treatment-engaged patients are at higher risk for reuse when they exhibit increased tolerance of unknown probabilistic outcomes (ambiguity tolerance) in a financial choice task (Konova et al., 2019 JAMA Psychiatry). But why patients become more tolerant of ambiguous uncertainty in periods preceding reuse remains unknown. One potential explanation consistent with decision theory is that, in these periods, they become overoptimistic about ambiguous outcomes, which leads them to overestimate the probability of good outcomes (or underestimate bad outcomes) when faced with a decision to reuse, and therefore more likely to do so. Here, we propose a multi-level, convergent test of this framework by using well-defined, quantitative measures of this presumed “optimism bias”, alongside quantitative measures of uncertainty tolerance, which we propose to collect with concurrent high-resolution fMRI recordings, and yoked to longitudinal clinical assessments. In Aim 1, we aim to establish the relationship between uncertainty tolerance and optimism bias in patients with OUD and matched controls by studying these behaviors across a set of choice and estimation tasks (the latter designed to capture optimism about simple financial and more complex outcomes, tapping into drug-choice-relevant domains such as health outcomes). We also examine for potential moderation by various psychopathological dimensions in a large, unselected population of online (MTurk) subjects. In Aim 2, we collect fMRI data during the same choice and estimation tasks to delineate the mechanism by which optimistic neural representations of uncertainty might drive behavioral tolerance of this uncertainty, and reuse, in OUD. In Aim 3, we use a multi- session longitudinal design to understand the interaction between optimism bias and uncertainty tolerance as they relate to opioid reuse, session-to-session, allowing us to elucidate the specific timescale and nature of this interaction. With this project we aim to provide an answer to why patients become more uncertainty tolerant in periods preceding reuse and, in doing so, hope to uncover an upstream mechanism (centered on optimism bias) of this vulnerability, including its neural implementation. In addition to this conceptual advance, this work will provide a novel set of cognitive tools to precisely and objectively measure these processes with potential to predict poor outcomes such as reuse, in a way that can be easily implemented in clinical settings. Finally, the findings from this work will inform novel therapeutic interventions by providing precise neurocognitive targets, as well as their ideal timing, with the goal of mitigating reuse risk and improving long-term patient outcome.