Game theory has served as a very useful tool to study complex social behaviors. It provides the mathematical frameworks to formulate various social scenarios to explore behavioral patterns during the interaction. Studies with human subjects have shown that various factors including cognition and social traits influence performance and strategies, and neuroimaging results reveal that multiple brain regions, such as the social brain network and the reward circuits, play crucial roles in strategic interactions. On the other hand, various animal species also show the ability to participate in complex games, with different levels of similarity to human subjects. Animal models allow high-resolution access to neural information at spatial and temporal scales that are currently not possible with human subjects and are thus very suitable for studying the neural basis of strategic social interaction. With advancements in technology and data analysis methods, the field is witnessing a surge of exciting new findings that reveal the neural mechanisms under more complex experimental settings. We therefore propose “comparative game theory” as a future direction for the field. We anticipate that by bringing ethology back to social neuroscience studies, we will take full advantage of animal models and understand brain function during strategic social interactions in their most naturalistic form.