Abstract How does the brain generate and maintain a persistent high-aggression state? While pathological, persistent aggression is a common symptom in many diverse mental health disorders including schizophrenia, bipolar disorder, post-traumatic stress disorder, autism, Rett syndrome, and traumatic brain injury, we lack a fundamental understanding of the neural mechanisms underlying persistent social states. Many models of aggression posit that this dysregulation occurs through the failure of “top-down” inhibitory control of subcortical circuits for aggression, through the circuit and synaptic basis for these models remain unclear. Here, we propose that pathological aggression may hijack circuit mechanisms used to generate persistent aggressive states in adaptive contexts. In particular, the experience of aggression has long been known to facilitate the emergence of a persistent high-aggression state, enabling animals to defend territory and status across long periods of time. Examining how experience “updates” neural circuits in the healthy brain to facilitate future aggression provides a unique window on how these circuits become dysregulated under pathological conditions. What are the neural mechanisms underlying experience-dependent updating? To explore this, we will look longitudinally at the changing relationship between neural activity in the ventromedial hypothalamus, ventrolateral area (VMHvl), an aggression output area with a well-described role in aggression in both sexes, and its “upstream” inhibitory inputs. In this proposal, we will test the novel hypothesis that aggression experience stabilizes a persistent aggressive state through a circuit “rerouting” mechanism rather than changes in the activity of inhibitory control loci. Using a variety of methods for supervised and unsupervised behavioral analysis, virally mediated anatomical tracing, synaptic physiology, optogenetics and cellular resolution high-density recordings, we will look longitudinally at how experience alters the fundamental properties of this circuit to implement behavioral change. First, we will map the putative identity of circuit nodes with the architectural capacity to reroute inhibition and characterize the changes in synaptic strength of this circuit across experience. Next, we will specifically examine the relationship between the activity of the regulatory input and the circuit-level output across experience. Lastly, we will perform high-density population recordings to elucidate the changes in the underlying computations being performed by the circuit to stabilize a high aggression state. Together, these data will provide a comprehensive integrated framework for understanding how experience generates a persistent behavioral state, and will pave the way for novel activity-dependent tools that may be able to detect neural signatures of experience and behavioral persistence in patient populations at risk for aggression dysregulation.