SUMMARY STATEMENT Aggression is an innate social behavior across vertebrate species. However, excessive aggression—ranging from school bullying to terrorist attacks—imposes a devastating risk to our society. Individuals with certain psychiatric disorders, such as bipolar disorder and post- traumatic stress disorder, are more likely to act violently, jeopardizing their own lives and those around them. While classic models of aggression have long-suggested that improper aggression results from dysfunction of “top-down” executive control of aggression-relevant circuitry, there is little direct physiological or neural circuit evidence to support this. Here, we will take an alternative “bottom-up” approach to investigate the inhibitory control impinged onto a neural locus with a clear role in aggression. We hypothesize that these inhibitory controls are essential to ensure aggression is expressed at the right time and towards the right target. We have identified the ventrolateral part of the ventromedial hypothalamus (VMHvl) as an essential locus for both “reactive” and “proactive” aggression. Whereas optogenetic activation of the VMHvl can promote both attack and aggression-seeking behavior, VMHvl inhibition has the opposite effect. In this study, we will address how local and long-range inhibitory inputs modulate responses of VMHvl cells, and consequently aggressive behaviors. Although neurons in the VMHvl are primarily excitatory, the VMHvl is directly and strongly inhibited by VMHvl “shell” neurons that are situated lateral and ventral to the VMHvl core. Thus, we hypothesize that the VMHvl shell is well positioned to shape aggression-relevant activity in excitatory VMHvl neurons. In the first part of the study, we will employ in vivo recording, functional manipulation, channelrhodopsin-assisted circuit mapping, and viral tracing to define the relationship between this inhibitory shell and excitatory “core” of the VMHvl. In the second part of the study, we will zoom out and examine the function of long-range inhibitory inputs to the VMHvl. Specifically, we will test the hypothesis that the inhibitory inputs from the medial preoptic area, a hypothalamic region critical for reproduction and parental care, is essential for suppressing VMHvl responses towards improper aggression targets and preventing misdirected attack. In summary, this study will provide a much needed framework to understand how inhibitory mechanisms control aggressive behavior, and pave the way for new types of circuit-level therapeutics for aggression control.