A core assumption of how humans understand and infer the intentions and beliefs of others is the existence of a functional self-other distinction. At least two neural systems have been proposed to manage such a critical distinction. One system, part of the classic motor system, is specialized for the preparation and execution of motor actions that are self realized and voluntary, while the other appears primarily involved in capturing and understanding the actions of non-self or others. The latter system, of which the mirror neuron system is part, is the canonical action 'resonance' system in the brain that has evolved to share many of the same circuits involved in motor control. Mirroring or 'shared circuit systems' are assumed to be involved in resonating, imitating, and/or simulating the actions of others. A number of researchers have proposed that shared representations of motor actions may form a foundational cornerstone for higher order social processes, such as motor learning, action understanding, imitation, perspective taking, understanding facial emotions, and empathy. However, mirroring systems that evolve from the classic motor system present at least three problems: a development, a correspondence, and a control problem. Developmentally, the question is how does a mirroring system arise? How do humans acquire the ability to simulate through mapping observed onto executed actions? Are mirror neurons innate and therefore genetically programmed? To what extent is learning necessary? In terms of the correspondence problem, the question is how does the observer agent know what the observed agent's resonance activation pattern is? How does the matching of motor activation patterns occur? Finally, in terms of the control problem, the issue is how to efficiently control a mirroring system when it is turned on automatically through observation? Or, as others have stated the problem more succinctly: "Why don't we imitate all the time?" In this review, we argue from an anatomical, physiological, modeling, and functional perspectives that a critical component of the human mirror neuron system is sensorimotor cortex. Not only are sensorimotor transformations necessary for computing the patterns of muscle activation and kinematics during action observation but they provide potential answers to the development, correspondence and control problems. © 2008 Pineda; licensee BioMed Central Ltd.