Despite long-standing interest in the relation of early speech production and phonological development, little is known about how speech motor skills develop in tandem with emerging phonological representations in childhood. Recent experimental work in our laboratory has shown that better speech motor control at young ages is associated with better outcomes at later ages in phonological sensitivity. In this talk we will advance a new, theoretical model to account for these findings. It is termed MIPA (Motor Involvement in Phonological Acquisition) and proposes that highly developed sensorimotor maps for speech support the acquisition of detailed phonological representations. We will begin by providing an overview of the model, its core predictions, and compare it to existing frameworks. We will then describe experimental studies designed to explore the conceptual merits of the model by testing children longitudinally on a set of measures assessing the coordination of vocal-tract gestures, phonological sensitivity, and the relationship between those two skills. Application of the model to the study and treatment of children with developmental disabilities that affect speech production, such as stuttering, as well as language-related disorders arising from poor sensitivity to phonological structure, such as dyslexia, will also be briefly described.