Abstract
Aim
Identifying correlates of suicidality is an important goal for suicide researchers because these correlates may predict suicidal behaviors. Psychological tasks that assess sensitivity to the outcomes of actions (i.e., consequence-based learning) have been commonly used by researchers seeking to identify correlates of suicidality. This is likely due to the straightforward integration of the tasks within most theoretical frameworks for understanding suicidality. Contextual factors have been shown to have a substantial effect on responding in behavior-outcome tasks. However, the direct relevance of these factors as determinants of behavior in suicide research is not clear. Thus, the purpose of this review was to assess the role of context in tasks involving behavior-outcome relations in suicide research.
Methods
Four databases were searched using terms from general learning theory. Articles that featured evaluation of tasks with hypothetical or real outcomes to differentiate suicidality were included.
Results
Eighty-two studies met inclusion criteria. Across studies there were 27 different tasks. Most instances of tasks across studies involved rewards (76.9%), while others emphasized punishment (15.7%), social (5.6%), or virtual suicide (1.8%) outcomes. Differentiation of suicidality was detected by 43.4%, 64.7%, 83.3%, and 50% of tasks featuring reward, punishment, social contexts, and virtual suicide respectively. All but five studies were retrospective.
Conclusion
Tasks that more closely mimic contexts and outcomes related to suicide appear to produce more pronounced differentiation of people with suicidality from people without suicidality. The lack of prospective designs is an important limitation of the literature.
Tasks that involve punishment or social outcomes better discriminate suicidality.
Reward-based tasks are overused in suicide research.
The conditioning hypothesis of suicidality is closely aligned with the literature.
Only 5 of 82 studies incorporated prospective measures.
HIGHLIGHTS
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the author(s).
Additional information
Notes on contributors
Mark J. Rzeszutek
Mark J. Rzeszutek, Department of Family and Community Medicine, University of Kentucky, 2195 Harrodsburg Rd, Suite 125, Lexington, KY 40504, USA; Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA. Anthony DeFulio and Grace E. Sylvester, Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA.
Anthony DeFulio
Mark J. Rzeszutek, Department of Family and Community Medicine, University of Kentucky, 2195 Harrodsburg Rd, Suite 125, Lexington, KY 40504, USA; Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA. Anthony DeFulio and Grace E. Sylvester, Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA.
Grace E. Sylvester
Mark J. Rzeszutek, Department of Family and Community Medicine, University of Kentucky, 2195 Harrodsburg Rd, Suite 125, Lexington, KY 40504, USA; Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA. Anthony DeFulio and Grace E. Sylvester, Department of Psychology, Western Michigan University, 1903 West Michigan Ave., Kalamazoo, MI 49001, USA.