It is quite evident that human locomotion is variable. In fact, no two footfalls are exactly alike. Historically, these variations were perceived to be noise superimposed on top of the control system. However, recent investigations in my laboratory and elsewhere have indicated that these variations have a deterministic pattern that is related to how the nervous system is organized for controlling movement. Although these observations are quite intriguing, it is not quite evident what neuromechanical variables may be governing these nonlinear gait dynamics. To initially address this question, we have developed reliable computer models that resemble the deterministic nonlinear gait pattern found in humans. These models have provided meaningful predictions of how changes in the nonlinear patterns are related to neural control of gait. The predictions from these models have aided our interpretation of complementary experiments with humans.