A great deal is known about the neural basis of associative fear learning. However, many animal species are able to use social cues to recognize threats, a defence mechanism that may be less costly than learning from self-experience. We have previously shown that rats perceive the cessation of movement-evoked sound as a signal of danger and its resumption as a signal of safety. To study transmission of fear between rats we assessed the behavior of an observer while witnessing a demonstrator rat display fear responses. With this paradigm we will take advantage of the accumulated knowledge on learned fear to investigate the neural mechanisms by which the social environment regulates defense behaviors. We will unravel the neural circuits involved in detecting the transition from movement-evoked sound to silence. Moreover, since observer rats previously exposed to shock display observational freezing, but naive observer rats do not, we will determine the mechanism by which prior experience contribute to observational freezing. To this end, we will focus on the amygdala, crucial for fear learning and expression, and its auditory inputs, combining immunohistochemistry, pharmacology and optogenetics. Finally, as the detection of and responses to threat are often inherently social, we will study these behaviors in the context of large groups of individuals. To circumvent the serious limitations in using large populations of rats, we will resort to a different model system. The fruit fly is the ideal model system, as it is both amenable to the search for the neural mechanism of behavior, while at the same time allowing the study of the behavior of large groups of individuals. We will develop behavioral tasks, where conditioned demonstrator flies signal danger to other naïve ones. These experiments unravel how the brain uses defense behaviors as signals of danger and how it contributes to defense mechanisms at the population level.