The article by Emanuel Todorov discusses the principles of optimal control in sensorimotor systems, emphasizing the evolution, development, and adaptation processes that shape motor function. Traditional theories focus on optimizing average trajectories while ignoring sensory feedback, but recent work has shifted attention to feedback control laws and mechanisms that generate behavior online. This shift has led to a more unified theoretical framework for interpreting motor function, with a focus on the relationship between high-level goals and real-time sensorimotor control strategies. Optimal control models have explained a wide range of empirical phenomena, offering both theoretical and practical advantages. Theoretical advantages include the justification of these models a priori, as the sensorimotor system is the product of optimization processes. Practically, optimal control modeling provides autonomy and generality, as it only requires a performance criterion to automatically generate movement details. The article categorizes models into two main types: open-loop and closed-loop control. Open-loop models optimize simple costs subject to boundary constraints, while closed-loop models incorporate sensory and motor noise, predicting both average behavior and task-specific sensorimotor contingencies. Closed-loop models are more flexible and can address a wider range of phenomena, including the scaling of movement duration with amplitude and accuracy, and the response to perturbations. The article also discusses the role of internal models in sensorimotor control, the emergence of motor synergies, and the hierarchical structure of sensorimotor function. It highlights the importance of considering low-level feedback loops in understanding complex behaviors and the potential of hierarchical optimal control models for real-time control in robotics and prosthetics.The article by Emanuel Todorov discusses the principles of optimal control in sensorimotor systems, emphasizing the evolution, development, and adaptation processes that shape motor function. Traditional theories focus on optimizing average trajectories while ignoring sensory feedback, but recent work has shifted attention to feedback control laws and mechanisms that generate behavior online. This shift has led to a more unified theoretical framework for interpreting motor function, with a focus on the relationship between high-level goals and real-time sensorimotor control strategies. Optimal control models have explained a wide range of empirical phenomena, offering both theoretical and practical advantages. Theoretical advantages include the justification of these models a priori, as the sensorimotor system is the product of optimization processes. Practically, optimal control modeling provides autonomy and generality, as it only requires a performance criterion to automatically generate movement details. The article categorizes models into two main types: open-loop and closed-loop control. Open-loop models optimize simple costs subject to boundary constraints, while closed-loop models incorporate sensory and motor noise, predicting both average behavior and task-specific sensorimotor contingencies. Closed-loop models are more flexible and can address a wider range of phenomena, including the scaling of movement duration with amplitude and accuracy, and the response to perturbations. The article also discusses the role of internal models in sensorimotor control, the emergence of motor synergies, and the hierarchical structure of sensorimotor function. It highlights the importance of considering low-level feedback loops in understanding complex behaviors and the potential of hierarchical optimal control models for real-time control in robotics and prosthetics.