Research in the areas of suicide and self-aggression has traditionally relied on non-experimental methodologies (e.g., case studies and correlational findings). Though useful in identifying correlates of lethal and non-lethal acts of self-aggression, these methodologies are unable to specify causality. A similar problem with aggression research was addressed via the use of laboratory analogues of aggression. However, early attempts at creating a laboratory analogue of self-aggression were methodologically flawed, and no studies have been published using a laboratory measure of self-aggression for almost 20 years. The proposed investigation will assess the validity of a newly developed laboratory measure of self-aggression, using a largely clinical population. Participants will consist of 108 individuals with a DSM-IV Borderline Personality Disorder (BPD) and 27 psychopathology-free control participants with no past self-aggressive behavior. Participants with a BPD will be placed in one of four groups based on their history of suicidal behavior (past attempts vs. no past attempts) and non-suicidal self-injurious behavior (SIB vs. no SIB). All participants will complete self-report and interview measures. Some measures will assess risk factors for self-aggression (e.g., depression, hopelessness, impulsivity), while others will assess unrelated constructs (e.g., competitiveness, social desirability). Increased platelet serotonin (5-HT)-2a receptor binding has been found to be positively correlated with suicidal behavior. Therefore, blood will be drawn to assess platelet 5- HT-2a receptor density. Participants will then begin the self-aggression paradigm (SAP) in which they will compete against a fictitious opponent. Prior to each of the SAP trials, participants will select the level of shock they will receive if they lose that trial. Shock settings will range from no shock to a shock level the participant believes is twice what they previously stated was very unpleasant. Construct validity for the SAP would be evidenced by: (1) significantly higher mean and maximum self-selected shocks among groups with a history of self-aggressive behaviors, (2) significant correlations between self-selected shocks on the SAP and other known correlates of self-aggression, (3) non-significant differences for self-selected shock between the BPD and non-BPD groups with no history of self-aggressive behaviors, and (4) non-significant correlations between self-selected shock and constructs unrelated to self-aggression.

R03MH067193

2004-03-01

2007-02-28