A neural network model of Borderline Personality Disorder

Abstract

The etiology of Borderline Personality Disorder (BPD) is unknown. This paper develops an etiological hypothesis by constructing a neural network with constraints from neuroanatomy, neurophysiology, and behavior. The neural network ascribes roles to the brainstem's periaqueductal gray, the amygdala, and the anterior cingulate/ventromedial prefrontal cortex (ACC/vmPFC). Neural network simulations show how these brain structures might interact during BPD behavior. The simulations suggest that long term depression (LTD) in ACC/vmPFC may explain several BPD symptoms. The network makes testable suggestions. The current work is the first-ever neural network simulation of BPD.