The origins and evolution of human languages remains today one of the most fascinating unresolved puzzles of science and many disciplines are currently working on the subject. Computer scientists have also become involved, trying to program artificial agents in such a way that they are able to self-organize a language-like system from scratch. There are already some successful simulations. They endow agents with language strategies in the form of networks of embodied cognitive functions for producing, comprehending, learning and aligning some aspect of language, for example to express participant roles in events using a case grammar. The present project goes one step further and inquires about the origins of these language strategies and how they can become shared in a population, based on the hypothesis that language strategies form in each individual by the epigenetic recruitment and exaptation of distributed networks, which support the strategies culturally emerging in a population.The project will develop the computational foundations for this ‘recruitment theory of language origins’, engage in simulations and robotic experiments, and validate them from a linguistic and biological point of view. It will look at the genetic bases of language by conducting Genome-wide Association Studies correlating phenotypic variation in linguistic aptitude, particularly for language learning, with genetic markers. At the end of this interdisciplinary project a more encompassing picture of language evolution should begin to emerge, illuminating the cultural, cognitive and genetic factors that have played a role in the origins and evolution of language.