Decision Processes of Late-Life Suicide

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): This is a competitive renewal of a four-year study of cognitive markers of late-life suicide. Our overarching goal is to find factors predictive of high-lethality suicidal behavior in late life. Extending the stress-diathesis model, our research program seeks to understand how trait-like diatheses - impaired cognitive control, deficits in social processing, and impulsivity - are expressed in poor decisions, both in experimental paradigms and in the context of real-world decision making. Our current cross-sectional study focuses on cognitive correlates of late-life suicidal behavior, linking it to poor cognitive contro. Adding to a growing body of evidence, we have preliminary evidence that older suicide attempters are impaired in their decision-making. Further, we have important clues about dimensions of heterogeneity in suicidal behavior. We see that high-lethality suicide attempters struggle in complex environments and make poor social decisions. By contrast, we see a tendency toward short-sighted decisions falling under the broad category of impulsivity in people with low-lethality and unplanned suicide attempts. The proposed study examines the cognitive and neural underpinnings of disadvantageous decisions in individuals with high-lethality attempts. We propose to (i) relate cognitive control deficits to poor decision-making in complex, ambiguous environments in older suicide attempters, (ii) to describe neural system alterations that may underlie poor cognitive control and disadvantageous decisions, (iii) to extend our investigations to poor social decisions that may contribute to late-life suicidal behavior, and (iv to develop an algorithm for predicting suicide risk that incorporates these features in addition to established risk factors. These goals will be accomplished by two interlinked studies. Study I is a cross-sectional, case- control comparison of 70 high-lethality suicide attempters, 70 low-lethality suicide attempters, 70 non-suicidal depressed, and 50 non-psychiatric controls. We focus on older suicide attempters, as suicide rates are high in old age and because late-life suicide attempts, due to their serious and determined nature, offer a window into death by suicide. Study I will use a framework that combines behavioral economics and cognitive neuroscience, with multilevel clinical, behavioral, cognitive, and imaging assessments. Our sampling and assessment strategy was designed with particular attention to possible confounders. Study II will follow 250 older suicide attempters (110 enrolled to date + 140 newly recruited) for up to 4 years, capturing recurrent attempts. It will serve as a prospective validatin of our predictive model. The extensive clinical and behavioral/cognitive characterization combined with a machine learning approach will allow us to develop an algorithm for suicide risk prediction which can inform clinical practice. The investigative team includes experts in reward learning and model-based imaging (Dombrovski), geriatric neuroimaging (Aizeinstein), behavioral economics (De Bruine), analysis of neural and longitudinal data (Weissfield), decision neuroscience (Clark), neuropsychology of suicide (Keilp) and of geriatric depression (Butters).