Abstract: Social communication is fundamental in animal life, offering a powerful framework for studying neural functions in the context of natural behaviors. Our lab aims to understand the operation and evolution of neural circuits underlying social communication. Specifically, we investigate courtship behaviors in Drosophila by developing novel genetic, physiological, and connectomic approaches to analyze circuits across model and non-model species. In this talk, I will share our recent finding on the neural circuits for courtship song production in the model species. By developing a technique that enables the recording of neural activity in singing flies, we determined how the nervous system drives the same muscles to generate different songs. Additionally, I will discuss how evolutionary changes in these neural circuits contribute to the emergence of distinct singing behaviors among closely related species.