Decision Processes of Late-Life Suicide: Competitive Renewal

摘要 这是对晚年自杀的纵向研究的竞争性更新。我们的最终目标是 了解老年自杀行为背后的决策过程。为了实现这一目标，我们有 积累了一组抑郁症老年人的高危队列(包括173例既往自杀未遂和82例严重自杀 自杀意念)，并预期有较高的自杀行为发生率。目标1.到目前为止，我们已经证明 一方面，决策过程改变和认知控制缺陷之间的强烈关联，以及 另一方面，自杀行为的历史与这些损伤破坏搜索的概念一致 寻找自杀的建设性替代方案。作为关键的下一步，我们将前瞻性地确认这些关联 通过对自杀未遂和自杀死亡事件的预测。我们预测决策指数将有 在解释了最近的自杀念头后的预测力。找出对事故最有用的预测因素 对于自杀行为，我们还将开发一种基于广泛潜在风险的节俭预测模型 各种因素。目标2.到目前为止，我们的研究主要集中在非社会背景下的决策。然而，许多人 自杀的诱因和威慑根植于社会关系。在自杀危机中，人们往往不能 想想他们的自杀对他们周围的人会有多大的破坏性。与年龄有关的规范性事件影响 社会地位，如残疾或退休，可能会在有自杀倾向的晚年引发自杀行为- 依赖的态度。探索灾难性决策背后的机制，以应对社会损失 优势，我们提出了一项涉及地位丧失操纵的社会决策研究。此外, 我们建议测试在高自杀风险个体中社会关系的钝化动机效应是否 与社会刺激引起的巴甫洛夫行为偏差和神经反应的改变有关。要实现 为了实现这些目标，我们建议延长现有队列的后续行动，并增加80名新征聘人员 试用者和60名理想者。我们对社会决策的研究将采用病例对照设计(包括 没有自杀未遂或意念和非精神控制的抑郁症患者)以检测效果 超越那些由抑郁和自杀念头所带来的，并尽量减少混杂因素。我们将使用 决策神经科学框架，将基于模型的成像与多层次临床、行为和 认知评估。我们已经建立了以安全、合乎道德的方式进行这项研究的程序。 调查小组在自杀行为和神经认知过程方面提供了互补的专业知识 (Szanto)，在巴甫洛夫和工具性奖励学习和基于模型的成像(东布罗夫斯基，哈尔奎斯特)中， 社会决策(克罗基特)，神经和纵向数据分析(加尔法维)，决策神经科学和 实验设计(克拉克)，老年抑郁症的神经心理学(巴特斯)。

ABSTRACT This is a competitive renewal of a longitudinal study of late-life suicide. Our ultimate aim is the greater understanding of decision processes that underlie suicidal behavior in old age. Toward this goal, we have accrued a high-risk cohort of depressed older adults (including 173 with past suicide attempts and 82 with serious suicidal ideation) with prospectively ascertained high rates of suicidal behavior. AIM 1. To date, we have shown robust associations between decision process alterations and cognitive control deficits on one hand, and the history of suicidal behavior, on the other, consistent with the notion that these impairments undermine the search for constructive alternatives to suicide. As a crucial next step, we will confirm these associations prospectively through prediction of incident suicide attempts and death by suicide. We predict that decision indices will have predictive power after accounting for recent suicidal ideation. To identify the most useful predictors of incident suicidal behavior, we will also develop a parsimonious prediction model based on a broad range of potential risk factors. AIM 2. Our studies to date have focused on decision-making in non-social contexts. However, many precipitants and deterrents to suicide are rooted in social relationships. In a suicidal crisis, people often fail to consider how devastating their suicide would be for those around them. Normative age-related events affecting social status, such as disability or retirement, may trigger suicidal behavior in late life in a predisposition- dependent manner. To probe the mechanisms underlying catastrophic decisions in response to loss of social dominance, we propose a study of social decision-making involving a loss-of-status manipulation. In addition, we propose to test whether blunted motivational effects of social relationships in high-suicide-risk individuals are related to alterations in Pavlovian behavioral biases and neural responses evoked by social stimuli. To achieve these aims, we propose to extend the follow-up of the current cohort and expand it with 80 newly recruited attempters and 60 ideators. Our study of social decision-making will employ a case-control design (including depressed individuals with no history of suicide attempt or ideation and non-psychiatric controls) to detect effects beyond those by conferred by depression and suicide ideation, and to minimize confounders. We will use a decision neuroscience framework, combining model-based imaging with multilevel clinical, behavioral, and cognitive assessments. We have established procedures to conduct this research in a safe, ethical manner. The investigative team contributes complementary expertise in suicidal behavior and neurocognitive processes (Szanto), in Pavlovian and instrumental reward learning and model-based imaging (Dombrovski, Hallquist), social decision-making (Crockett), analysis of neural and longitudinal data (Galfalvy), decision neuroscience and experimental design (Clark), neuropsychology of geriatric depression (Butters).