Background/Objectives: Suicidal ideation and behavior in adolescence remain difficult to assess due to their multifactorial and fluctuating nature. Performance-based measures may provide additional information on psychological correlates associated with suicidality beyond self-report and interview data. Methods: This study examined multivariate associations between RorschachPerformance Assessment System (R-PAS) variables and suicidality in a clinical sample of 153 help-seeking adolescents. Elastic net penalized regression models were estimated to evaluate joint patterns of R-PAS variables derived from the Suicide Concern Composite (SC-Comp) and selected developmentally relevant indices, controlling for age, sex at birth, sociodemographic and clinical factors. Suicidal ideation severity and lifetime suicidal behavior were assessed using the Columbia–Suicide Severity Rating Scale (C-SSRS). Results: For both outcomes, morbid content (MOR) and Location, Space, and Object Qualities-Complexity (LSO-Cmplx) emerged as consistent multivariate correlates, with vigilance–avoidance (VFD) contributing only in the model for suicidal ideation severity. Model performance was modest (R2 = 0.09), indicating limited explanatory power. The findings do not support clinical or predictive use of R-PAS variables for suicidal ideation and behaviors, but indicate a small set of reproducible multivariate associations within this sample. Conclusions: The results suggest that certain R-PAS variables show weak but consistent associations with suicidality in a help-seeking adolescent sample. Given the cross-sectional design and modest explanatory power, the findings should be interpreted as exploratory and hypothesis-generating, and further studies are needed to clarify the robustness and meaning of these associations.
Girolamo et al. (Wed,) studied this question.