Experimentally Assessed Reactive Aggression in Borderline Personality Disorder

Abstract

Approximately 73% of patients suffering from Borderline personality disorder (BPD) exhibit aggressive behaviour, which severely hinders therapeutic work and clinical improvement. Because the underlying mechanisms of aggression in BPD are not yet completely understood, additional research in this domain has a high clinical and scientific relevance. We employed a modified version of the Taylor Aggression Paradigm (mTAP), in order to examine for the first time whether this task can be used to differentiate between BPD patients and healthy controls with regard to reactive aggression. In the mTAP, the amount of money subtracted by a virtual opponent was categorized into 'low' (10-20 cents) and 'high' (80-100 cents) provocations, enabling us to compare how much money BPD patients and healthy controls subtracted (i.e., how aggressively participants responded) following high and low provocation trials. Our results showed that, compared to healthy controls, BPD patients showed higher overall aggression, higher aggression after high provocation trials, as well as a larger difference between high and low provocation trials. This finding was corroborated by a neuropsychological assessment, demonstrating higher levels of aggression and impulsivity in BPD patients. Interestingly, reactive aggression in the mTAP was positively correlated with symptom severity and impulsivity in BPD patients. We suggest that the mTAP provides a valuable tool allowing psychiatrists to quantify reactive aggression in BPD. Therefore, clinicians and researchers might consider this task, as a short experimental measure of reactive aggression, either in future studies or to aid diagnostic assessment during clinical practice.

Fig 1. Differences in reactive aggression between BPD patients and healthy controls.

Visualized are data from the modified Taylor Aggression Paradigm (mTAP), comparing the results of patients with Borderline personality disorder (BPD) and healthy controls (HC). Error bars represent standard deviation (SD). (A) The BPD group showed significantly higher aggression overall, which is represented by the money subtracted from the opponents. (B) The patients also had a significantly larger difference of subtractions when comparing trials following high provocation vs. trials following low provocation. (C) A significantly higher aggression following high provocation trials for BPD patients was also observed. (D) BPD patients also showed a numerically, but not significantly, higher aggression following low provocation trials.

Fig 2. Emotional and effective changes following the modified Taylor Aggression Paradigm.

Here, data from the State Trait Anger Expression Inventory (STAXI), Positive and Negative Affect Schedule (PANAS) and the subscale anger of the Emotional Self Rating (ESR) is shown. Pre and Post represent the time points immediately before and after the mTAP. Error bars represent standard error of the mean (SEM).

Fig 3. Relationship between reactive aggression and Borderline symptom severity and impulsivity.

For the group of BPD patients, the figure visualizes the correlation (Pearson’s r) between reactive aggression (i.e., mean amount of money subtracted from the opponent during the mTAP), and the total scores of the Borderline personality inventory (BPI) and the Barratt Impulsiveness Scale (BIS-11). We found a significant positive correlation between reactive aggression and the BPI total score, which represents an overall indicator of BPD symptom severity, and between reactive aggression and the BIS-11 total score, which provides a reliable measure of impulsivity.