Proprioceptive feedback contributes to the adaptation toward an economical gait pattern

Abstract

Humans generally prefer gait patterns with a low metabolic cost, but it is unclear how such patterns are chosen. We have previously proposed that humans may use proprioceptive feedback to identify economical movement patterns. The purpose of the present experiments was to investigate the role of plantarflexor proprioception in the adaptation toward an economical gait pattern. To disrupt proprioception in some trials, we applied noisy vibration (randomly varying between 40-120Hz) over the bilateral Achilles tendons while participants stood quietly or walked on a treadmill. For all 10min walking trials, the treadmill surface was initially level before slowly increasing to a 2.5% incline midway through the trial without participant knowledge. During standing posture, noisy vibration increased sway, indicating decreased proprioception accuracy. While walking on a level surface, vibration did not significantly influence stride period or metabolic rate. However, vibration had clear effects for the first 2-3min after the incline increase; vibration caused participants to walk with shorter stride periods, reduced medial gastrocnemius (MG) activity during mid-stance (30-65% stance), and increased MG activity during late-stance (65-100% stance). Over time, these metrics gradually converged toward the gait pattern without vibration. Likely as a result of this delayed adaptation to the new mechanical context, the metabolic rate when walking uphill was significantly higher in the presence of noisy vibration. These results may be explained by the disruption of proprioception preventing rapid identification of muscle activation patterns which allow the muscles to operate under favorable mechanical conditions with low metabolic demand.

Keywords: Gait adaptation; Metabolic cost; Muscle activity; Tendon vibration.

Fig. 1.

Typical average trace of MG muscle activity pattern during the stance phase, from heel-strike to toe-off. The mean EMG level was calculated during a storage phase (30–65% of stance) and a return phase (65–100% of stance).

Fig. 2.

Effects of Achilles tendon vibration on posture. (a) The group average CoP trace illustrates the vibration-driven shifts in CoP location over time. To account for the initial location on the force plate, all changes in CoP position over time are illustrated relative to the average position during the first 30 s of standing. The shaded area indicates standard error. (b) Time period significantly (p=0.0004) influenced average anteroposterior CoP speed, which was highest during vibration. Error bars indicate standard deviation and asterisks (*) indicate a significant difference between the indicated time periods (post-hoc tests; p <0.017).

Fig. 3.

Effects of Achilles tendon vibration on gait adaptation. (a) Treadmill incline changed from 0% to 2.5% midway through the 10 min trial. For all panels, the column to the left of the dashed line illustrates behavior while walking on level ground, and the right column illustrates behavior while walking uphill. (b) Noisy vibration had minimal effects on stride period while walking on level ground, but slowed the gradual increase in stride period after the incline transition. (c) While walking on level ground, vibration slightly increased storage phase MG activity. In contrast, vibration caused reductions in storage phase MG activity during the first few minutes of walking after the incline transition. (d) Vibration increased return phase MG activity while walking on level ground, and for the first few minutes of walking uphill. (e) On level ground, vibration did not influence metabolic rate. After the incline transition, metabolic rate was significantly higher during trials in which vibration was applied. For clarity, this is magnified for the final 3 min of uphill walking. For panels (b)–(e), shaded areas indicate standard errors, while asterisks (*) indicate a significant effect of vibration (post-hoc tests; p <0.01) during the indicated time period.

Fig. 4.

Schematic illustration of the proposed relationship between proprioceptive feedback and gait behavior. (a) During steady-state walking, a “correctly” chosen activation pattern will allow the muscle (M) to operate under favorable contractile conditions while the tendon (T) stores and returns mechanical energy, producing strong push-off power. (b) Following an incline transition, the previous plantarflexor activation pattern may no longer be appropriate. The resultant less favorable muscle contractile conditions and reduced tendon energy return would decrease push-off power. Proprioceptive feedback could contribute to the adjustment of plantarflexor activation, in order to return to mechanically advantageous musculotendon behavior.