Neuro-computational predictors of treatment responsiveness in trauma-exposed Veterans.

而创伤后应激障碍的循证治疗(EBTS)在减少创伤相关焦虑方面是有效的 症状，大约一半到三分之二的创伤暴露的退伍军人在治疗和治疗期间没有完全恢复 维持他们的创伤后应激障碍诊断。快感缺乏症，即减少对有回报的活动的兴趣和参与，是 流行于创伤暴露的退伍军人中，与创伤后应激障碍严重程度较高和较差有关 对精神治疗的反应。因此，奖赏敏感度受损很可能在 退伍军人的治疗反应性。然而，到目前为止，这种改变的奖励敏感性的程度 影响创伤后应激障碍的治疗反应性尚未经过测试。为了验证这一假设，拟议的研究将 结合计算建模和事件相关功能磁共振成像(FMRI)进行分析 认知加工疗法(CPT)结束时退伍军人的奖赏加工功能 这些标记物在预测治疗反应方面的有用性。计算建模，特别是在 与神经成像相结合，提供了对复杂认知过程的详细机械洞察，这可以 比标准的行为和神经影像分析更准确地预测临床结果。我们会 利用这一方法来描述创伤暴露退伍军人治疗反应的可靠预测因素。 共有186名创伤暴露退伍军人将在参加CPT后立即招募。他们会 完成完整的临床评估和两个多臂强盗(MAB)任务(在传统和社会条件下，以 在探索性分析中进行比较)，其中他们必须从一组选项中选择每项试验 具有未知的回报概率，目标是最大化总回报。同时进行的大脑活动将是 在93名退伍军人中进行测量，这些退伍军人将在接受功能磁共振检查的同时完成任务。贝叶斯学习 模型将应用于参与者的决策，以得出代表a)个人的个人级别的参数 环境中未知奖励率的感知稳定性，以及b)他们的模型- 基于对回报的预期会影响他们的选择。神经激活在参数上与试验到- 基于试验模型的奖励期望和相关预测误差(即期望和 观察到的奖励)将被提取。所有参与者将完成后续临床和行为评估 治疗后即刻及治疗后3个月。计算参数和基于模型的神经 激活将作为治疗前后创伤后应激障碍严重程度变化的预测因子进行测试，控制治疗前 治疗创伤后应激障碍的严重程度和相关的精神合并症。该项目旨在确定是否 奖赏加工的计算标记(目标1)和奖赏的相关神经关联 EBT开始时的预期(目标2)可用于预测创伤后应激障碍症状的减轻 在受过创伤的退伍军人中。目标3将评估这些计算标记是否 预测治疗后3个月的治疗结果。与治疗相关的变化 奖赏加工的计算标记及其与快感缺乏和创伤后应激障碍变化的关系 还将探讨严重性(目标4)。这项研究的结果将有助于确定未得到满足的治疗需求 并为有康复不良风险的退伍军人制定治疗计划和复发预防工具 来自创伤后应激障碍。识别这种预测机制也将提供关键的神经和心理目标 开发更有效、个性化的治疗方法以改善创伤后应激障碍的康复(例如，认知培训以 提高奖励敏感度，减少快感缺失)。

While evidence-based treatments (EBTs) for PTSD are effective at reducing trauma-related anxiety symptoms, about half to two thirds of trauma-exposed Veterans do not fully recover during treatment and maintain their PTSD diagnosis. Anhedonia, i.e., a reduced interest and engagement in rewarding activities, is prevalent in trauma-exposed Veterans and is associated with including higher PTSD severity and poorer response to psychiatric treatment. Impaired reward sensitivity is therefore likely to play a critical role in treatment responsiveness in Veterans. However, to date, the degree to which such altered reward sensitivity impacts PTSD treatment responsiveness has not been tested. To test this hypothesis, the proposed study will combine computational modeling and event-related functional magnetic resonance imaging (fMRI) to assay reward processing function in Veterans at the end of Cognitive Processing Therapy (CPT), and test the usefulness of such markers in predicting treatment responsiveness. Computational modeling, particularly in concert with neuroimaging, provides detailed mechanistic insights into complex cognitive processes, which can predict clinical outcomes more accurately than standard behavioral and neuroimaging analysis. We will capitalize on this approach to delineate robust predictors of treatment response in trauma-exposed Veterans. A total of 186 trauma-exposed Veterans will be recruited immediately upon enrolling in CPT. They will complete a full clinical assessment and two multi-arm bandit (MAB) tasks (in classic and social conditions, to be compared in exploratory analyses), in which they must choose on each trial from among a set of options with unknown reward probabilities, with the goal of maximizing total rewards. Concurrent brain activity will be measured in a subset of 93 Veterans who will complete the task while undergoing fMRI. A Bayesian learning model will be applied to participants’ decisions to derive individual-level parameters representing a) individuals’ perceived stability of the unknown reward rates in the environment and b) the degree to which their model- based expectations of reward influence their choices. Neural activation parametrically associated with trial-to- trial model-based reward expectations and associated prediction errors (i.e., difference between expected and observed reward) will be extracted. All participants will complete follow-up clinical and behavioral assessments immediately after treatment and 3 months after treatment. Computational parameters and model-based neural activations will be tested as predictors of pre- to post-treatment change in PTSD severity, controlling for pre- treatment PTSD severity and relevant psychiatric comorbidities. This project aims to determine whether computational markers of reward processing (Aim 1) and associated neural correlates of reward anticipation (Aim 2) at the onset of EBT can be useful in predicting reduction in PTSD symptoms among trauma-exposed Veterans. Aim 3 will assess whether such computational markers are predictive of post-treatment outcomes 3 months after treatment. Treatment-related change in computational markers of reward processing and their relationship to change in anhedonia and PTSD severity will also be explored (Aim 4). The outcomes of this study will help to identify unmet treatment needs in Veterans and develop treatment planning and relapse prevention tools for Veterans at risk for poor recovery from PTSD. Identifying such predictive mechanisms will also provide critical neural and psychological targets for developing more effective, personalized treatments to improve PTSD recovery (e.g., cognitive training to boost reward sensitivity and decrease anhedonia).