Suicide is a lethal problem for Veterans. A major challenge in predicting and preventing suicide is that ideation and suicide behaviors fluctuate over time, as do the cognitive, emotional, and physical states that precede them. Recent advances in mobile health technology hold promise for intensively measuring risk factors for suicide that are proximal, time-varying, and occur naturalistically in daily life using smartphones. Cognitive factors - including cognitive control and social cognition - play a role in suicide risk; to date, however, mobile health tools have not been adopted to objectively measure cognitive performance fluctuations in Veterans. To address this gap, we propose to use our previously developed mobile health tool for measuring cognitive control and social cognition in the daily lives of Veterans with elevated suicide risk. Drawing on our experience delivering mobile health assessments in individuals at-risk for suicide, we will examine the dynamic relationships between cognitive performance and suicide, as well as other contextual variables associated with suicide risk. A transdiagnostic sample of Veterans with current suicidal ideation will complete concurrent mobile cognitive tests (MCTs) and ecological momentary assessments (EMA) within a two-week intensive assessment. Our aims are to identify associations between cognitive test performance and constructs closely linked to suicide risk (restricted coping ability, negative interpersonal beliefs, ideation). We also seek to examine how relationships established in the intensive mobile health longitudinal dataset relate to real-world suicide outcomes which we will derive from the electronic medical record and telehealth assessments. We will use the dataset to identify subtypes of Veterans based on their temporal pattern of suicidal ideation and examine associations between subtypes and real-world suicide outcomes. Finally, we will explore the role of specific context variables (e.g., sleep) in suicide risk. Deliverables from this project will include an intensive longitudinal dataset that overcomes limitations of traditional static laboratory-based assessments. This dataset can be leveraged to assess additional real-time proximal risk factors associated with suicide in longitudinal studies and interventional research on the cognitive underpinnings of suicide.