The alarm response is an antipredator behavior displayed by many fish species and was first described 70 years ago. It is triggered through the olfactory system by substances released from injured skin and is characterized by dramatic, measurable changes in locomotion as well as physiology. We propose that this is an ideal time to revisit this response and to utilize it as an assay for understanding how neural circuits mediate innate fear. A suitable organism for these studies is the zebrafish, a genetic model with a rapidly expanding toolkit for molecular manipulation of the nervous system. Individual neurons mediating the response, ranging from receptor neurons to those in higher brain centers, should first be identified. New tools, specifically transgenic lines that allow spatial and temporal control of neural activity, provide a way to define and test the role of specific neurons, while genetic screens provide a route to identifying individual molecules essential for a normal response. Optical recording, which has proven successful in studies of information processing in the bulb, will provide valuable insights into neural circuitry function during the alarm response. When carried out on mutants, physiological analysis can provide insight into aspects of signal processing that are essential for normal behavior. The alarm response thus provides a paradigm to examine innate fear in a vertebrate system, enabling analysis at multiple levels from genes to the entire neural circuit. Additionally, the context dependency of the response can be utilized to investigate attention and decision making.