Abstract
Motor impulsivity is implicated in the transition from suicidal ideation to suicidal behavior. A prior study used a Go/No-go (GNG) task to show increased motor impulsivity in those with recent suicide attempt; additionally, computational modeling to extract latent cognitive variables from GNG has shown decreased decisional efficiency in those with an upcoming attempt. The current study aimed to examine a simpler version of GNG in a small sample of Veterans with prior history of suicidal behavior. Participants completed multiple GNG sessions over a one-year period. Each session was coded according to whether the participant had (1) an actual suicide attempt (ASA) in the 90 days following that session; (2) another suicide-related event (OtherSE) but not an ASA; or (3) neither (noSE) in the next 90 days. Although miss rates were low across all groups, results showed that, relative to noSE, an upcoming ASA was associated with 1) reduced false alarms in the behavioral data; 2) increased drift efficiency for foils in the computational modeling. These results complement the prior results which showed increased miss rate and increased drift efficiency for foils were associated with upcoming ASA. Importantly, in both studies, these variables were not associated with upcoming OtherSE, suggesting different cognitive processes associated with upcoming ASA, compared to other suicide-related events that fall short of an actual attempt. These two studies suggest the neurocognitive markers may represent both a cognitive risk factor and a behavioral marker of short-term risk for suicide attempt.