Neonatal imitation has rich implications for neuroscience, developmental psychology, and social cognition, but there is little consensus about this phenomenon. The primary empirical question, whether or not neonatal imitation exists, is not settled. Is it possible to give a balanced evaluation of the theories and methodologies at stake so as to facilitate real progress with respect to the primary empirical question? In this paper, I address this question. I present the operational definition of neonatal imitation as differential imitation and discuss why it is important to keep it in mind. I also examine the principal explanations for the extant findings. In particular, I focus on the contrast between Meltzoff and Moore’s Active Intermodal Matching (AIM) model and the Association by Similarity Theory (AST), which interprets neonatal imitation as differential induction of behaviors that already tend to occur spontaneously. I argue that AST is preferable to AIM for empirical and theoretical reasons. With respect to methodology, I investigate what experimental design can best provide evidence for imitation, focusing on how differential induction may be maximized and detected. Finally, I discuss the significance of neonatal imitation for the field of social cognition. Specifically, I propose a link with theories of direct social perception. Overall, my goals are to help clarify the complex theoretical issues at stake and suggest fruitful guidelines for empirical research.