(PS9-D99) Differences in Suicide-specific Attentional Bias Based on Stimuli Across the Suicide Stroop and Disengagement Tasks

Previous research has shown that suicide-specific attentional bias is related to suicide attempts through a maladaptive attentional fixation on suicide as a solution (Niu et al., 2021; Wenzel & Beck, 2008). Because of this, suicide-specific attentional bias may be used as a behavioral marker to predict future suicide attempts (Cha et al., 2010).The suicide Stroop taskand suicide disengagement task are objective measures that can be used to gain a clearer understanding of how attentional bias correlates with the severity of an individual’s suicidal thoughts and behaviors. The suicide-Stroop (Stroop, 1935) measures the difference in reaction time when categorizing suicide words (“dead”, “suicide”, and “funeral”) compared to neutral words (“engine,” “museum,” and “paper”), while the suicide-disengagement task (Fox et al., 2001) measures the reaction time to identify a target following a suicide word (“hang”, “”suicide,” and “overdose”) compared to a neutral word (“ruler,” “chalk,” and “notebook”). However, previous research using the suicide Stroop task has inconsistently shown group differences possibly due to different scoring approaches. For example, Chung and Jeglic (2016) found that interference specifically to the word “suicide” but not “death” or “funeral” predicted future suicidal behavior. The aim of the current poster was to examine interference specifically to the word "suicide" in comparison to other suicide-related words among a sample of 109 psychiatric inpatients from an urban US hospital (age: M = 42.5, SD = 12.8 ; % male = 78.9) with varying suicide attempt histories (n = 54 with a suicide attempt within 2 weeks; n = 28 with a suicide attempt > 1 year ago; n = 27 with suicidal ideation only and no prior attempts). 

Preliminary results indicated significant group differences in interference to the word “funeral” on the Stroop task, F(2,107) = 5.31, p=.006, such that participants with a recent suicide attempt exhibited greater attentional bias toward the word “funeral” compared to the other groups. Groups did not differ based on the other suicide-related words (or suicide category) on the Stroop task, nor any suicide-related words (or suicide category) on the disengagement task. Logistic regression predicting group differences by “funeral” interference on the Stroop task remained after controlling for current severity of suicidal ideation and depressive symptoms, X2 (6, 109) = 53.66, p = .000. Surprisingly, interference on the disengagement task to the word “overdose” (and the suicide category) was negatively correlated with self-report measures of suicide ideation, r(107) = -.19, p = .046, and attentional fixation on suicide, r(107) = -.19, p = .046, such that quicker disengagement away from the word “overdose” as compared to neutral words was related to higher attentional fixation and suicidal ideation. Stroop interference was not correlated with any study measure. These results further support the relationship between attentional bias and suicide-related outcomes (suicidal ideation and behavior) but suggest the degree of attentional bias may depend on the specific stimuli and may differ based on the task used (Stroop or disengagement tasks).