Reducing Self-Dehumanization to Examine the Effects on Oxytocin and Suicide Risk

Purpose: Humans are inherently social beings, and the need for belonging and connection is fundamental. Severe perceptions of social exclusion, non-belonging, and isolation can result in significant psychological harm, including suicide, which claims nearly 50,000 lives annually in the U.S. and affects over 13 million individuals through suicidal ideation (SI). A key predictor of SI, thwarted belongingness, arises from feelings of exclusion and non-belonging, yet more specificity is needed to identify the types of non-belonging that contribute most to suicide risk. Self-dehumanization-a perception of oneself as less than human-emerges as a promising factor in understanding and mitigating SI, with empirical evidence linking it to anxiety, depression, suicidal ideation, and social withdrawal. Preliminary research suggests that self-dehumanization is influenced by neurobiological processes, particularly oxytocin, a hormone critical to social connectedness and self-perception. Low oxytocin levels have been linked to self-dehumanization, reduced social reengagement, and heightened suicide risk, highlighting its potential as a target for intervention. Building on these insights, this study will experimentally induce and reduce self-dehumanization to examine its effects on SI and oxytocin levels. Thus, this study will use a self-dehumanization induction to analyze its impact on SI and oxytocin. Findings will provide a novel framework for integrating psychotherapeutic and neurobiological strategies into suicide prevention efforts.

Research Design/Method: The present study will utilize an experimental design with two groups, an experimental re-humanized group and a comparison control group.

Procedure: All interested participants will be instructed to fill out the screening survey to determine fit. Eligible participants will be invited to complete the study visit in person. Following consent, participants will complete a pre-induction fasted blood draw of 5 milliliters (i.e., approximately a tablespoon of blood). Next, participants will complete the Self-Injurious Thoughts and Behaviors-Short Form interview with the experimenter and a battery of randomized self-report assessments. After completion, eligible participants will then be randomly categorized into the re-humanization intervention or the control intervention. Following this, they will undergo a post-intervention 5 milliliter fasted blood draw. They will then complete a post-induction battery of self-report measures (see Materials). Participants will be queried for any changes in suicide risk and any current suicidal ideation and intent and will receive appropriate interventions (e.g., means safety counseling, safety planning). Upon completion participants will be compensated and provided mental health resources (e.g., the National Suicide Prevention Lifeline). Participants will be contacted one month later to fill out a brief battery of self-report measures. The appointment is expected to take approximately 2 hours to complete.

Data Analyses: All results will be analyzed in R. First, descriptive statistics and zero-order bivariate correlation analyses will be assessed. For this study, repeated measures analysis of covariance (ANCOVAs) will be used to examine group differences (i.e., the active re-humanized condition and control condition) in changes in oxytocin levels, self-dehumanization scores, and suicidal ideation, controlling for social isolation (i.e., thwarted belongingness, a factor of the Interpersonal Needs Questionnaire). Post-hoc pairwise comparisons will be conducted if the omnibus test indicates statistically significant effects. Following this, exploratory analyses will be conducted to assess if oxytocin partially mediates the relationship between self-dehumanization pre-experimental manipulation and suicide risk post-experimental manipulation (controlling for suicide risk at pre-experimental manipulation) through bootstrapped mediation procedures (i.e., 10,000 samples using the mediation and lavaan packages in R).