Fear is an emotion that exerts powerful effects on our behavior and physiology. A large body of research implicates the amygdala in fear of painful stimuli, but virtually nothing is known about the circuits that support fear of predators and social threats, despite their primal importance in human behavior and pathology. Unlike painful stimuli, predator and social threats activate the medial hypothalamus, a cluster of highly conserved brain nuclei that control motivated behavior. Intriguingly, predator and social threats recruit largely non-overlapping nuclei in the medial hypothalamus, and we have recently demonstrated that separate medial hypothalamic circuits are essential for predator and social fear. We aim to build a functional wiring diagram of predator and social fear in the mouse that will explain how these fears are triggered, coordinated, and remembered. Such a functional wiring diagram will reveal the network logic of innate fear and put us in a position to selectively intervene in fear processing. Electrical stimulation of the medial hypothalamus in humans elicits panic responses and pharmacological agents that block these circuits will offer unexplored therapeutic approaches to treat anxiety disorders such as panic, social phobia, and post-traumatic stress disorder. Moreover, the relatively simple architecture of the medial hypothalamic fear network and its robust and direct behavioral readout in the mouse will be a powerful platform to test the role of several fundamental circuit features that are common to a wide range of behavioral networks, but whose function remains unknown, including the role of feedback loops, sparse cellular encoding of behavior, and overlapping processing of distinct behavioral responses. In this way, the project will provide the first circuit-level understanding of predator and social fear and answer a series of fundamental questions about how neural networks control behavior.