The neural correlates of social reward in borderline personality disorder
Abstract
Borderline Personality Disorder (BPD) is a severe psychiatric condition marked by emotional instability, life-threatening behaviors such as non-suicidal self-injury (NSSI), and profound social dysfunction which persists even after other symptoms remit. While social dysfunction is considered the most debilitating aspect of this disorder, research investigating social processing deficits in BPD at a neurophysiological level is limited, and the role of social media has been overlooked. This doctoral thesis addresses these gaps with three studies investigating alterations in the neural processing of social rewards in BPD within naturalistic social media settings. In Study 1, we developed a novel, ecologically valid paradigm that uses participants’ personal Instagram photos to simulate real-life social media interactions. In this study, we tested the novel task in a sample of university students (N=30) using electroencephalography (EEG), and we show that it reliably elicits the most relevant event-related potentials (ERPs) associated with reward processing, that is, the Reward Positivity (RewP) and the Feedback-P3 (FB-P3). To explore whether borderline patients show altered RewP and FB-P3 responses to social rewards, in Study 2 we applied the novel paradigm in a group of BPD patients (BPD; N=21) and age and gender-matched healthy controls (HC; N=23). At a behavioral level, results showed that the BPD group rated the Likes and positive comments as less pleasant than HC. At a neural level, results showed no differences between groups in the RewP and FB-P3 responses to social rewards, but they did show a robust relationship between borderline traits and RewP amplitudes elicited by the receipt vs omission of positive social feedback in the BPD group.
That is, patients with more pronounced borderline traits showed a reduced RewP than patients with lower traits, providing a neurophysiological explanation for their reduced ability to derive pleasure from social rewards. Finally, to localize these alterations in the brain and to determine whether they are specific to BPD or reflect an unspecific alteration more broadly associated with NSSI, in Study 3 we adapted the paradigm for fMRI and we used a novel three-group approach, whereby we recruited a clinical group of BPD patients with recent and recurrent NSSI (NSSI+BPD; N=31), a subclinical group of young adults with similar NSSI characteristics but no comorbid BPD or other disorders (NSSI; N=27), and a healthy control group (HC; N=33). At a behavioral level, the NSSI+BPD group rated positive feedback as less pleasant than the HC group and negative feedback as more unpleasant, whereas the NSSI group rated the positive feedback similarly to HC but the negative feedback similarly to NSSI+BPD. Coherently, at a neural level, the NSSI+BPD group (compared to the HC group) showed a blunted activation in key regions of the reward
network when receiving social rewards vs social punishments, whereas the NSSI group showed an 4 intermediate activation in most studied regions, bridging the pattern observed in the other two groups and reflecting a continuum of severity in the reward network. Together with the behavioral results, these neural results support prior proposals of NSSI and BPD as two expressions of a developmental continuum, starting with a specific sensitivity to negative feedback in NSSI which may later generalize to positive social feedback in BPD. Taken together, findings from the three studies show a negative bias in decoding social media rewards in BPD and provide novel insights into the neural mechanisms that may underlie this bias.
In addition, findings show that this negative bias may be a unique feature of BPD and may be independent from NSSI, which may help better characterize alterations in social reward processing in BPD. Conclusively, findings refine our understanding of social dysfunction in BPD and may help design more precise interventions that expand the focus on social functioning and target the more persistent components of borderline pathology.