Since Sept. 11, 2001, more than 30,000 active duty members and veterans have taken their own lives – four times as many as those killed in post-911 military operations.
Current methods to detect early signs of behavioral and mental health risk factors rely on self-reporting and screening questionnaires, which can’t reliably predict suicidality. Effective behavioral health assessment is a mission-critical capability requiring novel tools to identify and help those at risk.
DARPA has announced the Neural Evidence Aggregation Tool (NEAT) program. NEAT aims to develop a new cognitive science tool that identifies people at risk of suicide by using preconscious brain signals rather than asking questions and waiting for consciously filtered responses.
“NEAT is a proof-of-concept effort attempting to develop a new tool for mental and behavioral health screening that moves us beyond historical and current methods of questions and consciously filtered responses,” said Greg Witkop, a former Army surgeon and current program manager in DARPA’s Defense Sciences Office. “Using the preconscious will hopefully enable us to detect signs of depression, anxiety, or suicidal ideation earlier and more reliably than ever before. If successful, NEAT will not only significantly augment behavioral health screening, but it could also serve as a new way to assess ultimate treatment efficacy, since patients will often tell their clinicians what they think the clinician wants to hear rather than how they are truly feeling.”
By way of analogy, NEAT is envisioned to be for mental health what an MRI is for the physical body: a way to assess injury. As an MRI can detect an early meniscal tear prior to a more serious injury developing that can impact a soldier’s readiness, NEAT would identify psychological and behavioral changes before they impact readiness.
NEAT is not focused on lie detection, truth detection, or assessing someone’s credibility but, rather, on aggregating preconscious brain signals to determine what someone believes to be true. The screening process envisioned could involve presenting various statement stimuli of behavioral health relevance, such as biographical information, actions, or intentions (e.g., I want to end my life/enjoy my life) to measure preconscious responses. NEAT would triangulate responses to aggregate evidence and determine if the person reading the stimuli statements believes they are true, false, or indeterminate.
The program includes two technical areas. The first area focuses on research and development and will include several multidisciplinary teams spanning cognitive science, bioengineering, and machine learning to address all of the key technical challenges to develop NEAT processes. The second technical area addresses independent validation and verification. Those selected for this second area will work with DARPA and technical area one teams to provide expertise and independently validate NEAT methods.
NEAT is planned as a 3.5 year program, with a 24-month proof of concept phase, followed by an 18-month operational setting phase. Throughout the program lifecycle, DARPA will leverage an independent Ethical, Legal, and Societal Issues (ELSI) group to advise NEAT program leadership and performers on ELSI concerns.
“Ultimately, NEAT intends to augment current behavioral health screening programs by providing clinicians with previously unavailable information to enable earlier interventions and more reliable measures of successful treatment,” Witkop said. “Just as the objective evidence of an X-ray or MRI is sometimes necessary to help a military member not feel like they are letting their unit down because of a visible injury, NEAT will attempt to provide objective evidence of invisible injuries for help to be provided in time.”
A virtual Proposers Day for potential proposers is scheduled for March 15, 2022. For more information and registration details, visit SAM.gov. A Broad Agency Announcement (BAA) solicitation is expected to be available on SAM.gov in the coming weeks.
Read the announcement at DARPA
