Although several methods have been used to estimate exercise-induced changes in human neuronal networks, there are growing doubts about the methodologies used. This review describes a single motor unit–based method that minimizes the errors inherent in classical methods. With this method, it is now possible to identify human neuronal networks' changes due to exercise.