Research Vision

Suicide risk can change rapidly, yet most systems assess and identify risk only intermittently. The ASSIST Lab seeks to close this gap by studying self-injurious thoughts and behaviors as they unfold in daily life and by developing scalable tools for real-time suicide prevention.

Our work combines intensive longitudinal assessment, passive sensing, screenomics, computational modeling, and adaptive intervention designs. Across studies, we aim to identify when risk is elevated, understand the emotional and social processes that contribute to risk, and develop technology-enabled supports that can extend suicide prevention beyond traditional clinical encounters.

Our long-term goal is to build real-time suicide prevention systems that are clinically useful, person-centered, scalable, and responsive to the needs of individuals experiencing suicidal thoughts and behaviors.

The Problem

---

Suicide risk often fluctuates over short periods of time, but traditional assessment and intervention approaches are not designed to capture these changes as they happen.

---

Our Approach

We use real-time and technology-enabled methods, including ecological momentary assessment, passive sensing, screenomics, computerized adaptive testing, and computational modeling.

---

Our Long-Term Goal

We aim to develop scalable tools that improve suicide risk assessment, identify moments when support may be most helpful, and deliver timely interventions in daily life.