Background The time courses of the joint elevation angles of the thigh, shank, and foot in one stride during walking can be well approximated by a “plane” in a triaxial space. This intersegmental coordination (IC) of the lower limb elevation angles is referred to as the planar covariation law. We examined the effects of exercise habituation and aging on the thickness of the IC plane of the lower limbs under sinusoidal speed changing conditions. Methods Seventeen sedentary young (SY), 16 active young (AY), and 16 active elderly (AE) adults walked on a treadmill in accordance with a sinusoidal speed changing protocol at 120, 60, and 30 s periods with an amplitude of ± 0.56 m·s−1. Motion of the lower limbs from the sagittal direction was recorded to calculate the elevation angles of the lower limbs. When the best-fit IC plane was determined, the smallest standard deviation of the IC plane was considered as the anteroposterior gait variability of the lower limbs. The coefficient of variance of the step width was also quantified to evaluate the lateral step variability (CVSW). Results The standard deviation of the IC plane was significantly greater in the order of SY, AY, and AE, regardless of the sinusoidal wave periods of the changing speed. The CVSW was not significantly different among the three groups. Conclusions Exercise habituation influences anteroposterior gait variability of the lower limbs, but not lateral step variability, even in young adults. Given these, gait adaptability for sinusoidal speed changes does not always decline with aging. Trial registration UMIN000031456 (R000035911; registered February 23, 2018).

Objective: A U-shaped relationship between energy cost of walking (Cw; J·kg-1·m-1) and walking speed indicates that there is a specific speed minimizing the Cw, called economical speed (ES). It is mostly slower in older adults than young adults; however, effects of leg length on the ES have been ignored. We investigated the effects of aging and exercise habituation on the normalized ES by the leg length (nor-ES). Moreover, the diversity in the stride length (SL) and step frequency (SF) against sinusoidal speed change (SSC) at 30-s and 180-s periods with ±0.56 m･s-1 among sedentary young (SY), active young (AY), and active elderly (AE) adults was also quantified. Results: The ES was significantly faster in the following sequence: AY, SY, and AE, whereas nor-ES was faster in the AY and SY than in the AE (no difference was found between AY and SY). Delayed SL and preceding SF were observed at the 180-s SSC in the young adults only. Collectively, greater delay and/or precedence of SL-SF combinations against SSC may reflect locomotive flexibility for passive speed change during walking. The nor-ES was slowed by aging; however, it was not influenced by exercise habituation, at least, in young populations.