The force driven harmonic oscillator model accurately predicts the preferred stride frequency for backward walking

Abstract It has been established that when gait speed is freely selected, there is a strong natural tendency for a stride rate and length combination to be utilized that results in minimized metabolic cost. A possible mechanism for this self-optimizing behavior is that a biomechanically optimal movement pattern is adopted, which should reduce or minimize metabolic cost. When the legs are modeled as pendular force driven harmonic oscillators (FDHO) the stride frequency during freely selected gait is predictable as the pendulum's resonant frequency — the state in which force inputs to maintain oscillations are minimal. Thus, resonance has been proposed as a mechanism responsible for gait optimization. To further examine the durability of this mechanism, the FDHO model was applied to predict stride rates during both free forward and backward walking for two separate test sessions. Stride measures were stable across sessions. Forward walking resulted in 25% greater stride lengths than backward, however, stride rates were statistically equal. Also, the FDHO model successfully predicted the preferred stride rate for both backward and forward walking. The resulting invariance of the stride frequency and adjustment of stride length were interpreted as offering additional support for the resonance mechanism.