Author(s): BOUDOLOS, K., CHALKIADAKIS, E., EMMANOUIL, A., PADERI, F., ROUSANOGLOU, E., Institution: NATIONAL AND KAPODISTRIAN UNIVERSITY OF ATHENS, GREECE., Country: GREECE, Abstract-ID: 2318

INTRODUCTION:
Gait transitions behave like bifurcations between attractors with the relative phase of the leg segments as an order parameter and step velocity and length as control parameters [1]. The vast majority of studies examine the walk-to-run (W-to-R) transition on the treadmil rather than on ground surface with few studies testing the external load attractor [2]. This study tested the load effect on the W-to-R transitional step (step0) and on the inter-thigh relative phase angle (IT-RPA) during ground surface gait.
METHODS:
Twenty-three men (18-26 yrs) performed 5 W-to-R trials (preferred speed, 3.5m walking, 2.3m transition, and 4.5m running length corridor) in 3 load conditions (no load: NoL, Trunk load: TrL-vest of 5.1kg front, 5.1kg back, Extremities Load: ExtrL-2.5kg each wrist and ankle). Five inertial sensors (Xsens Master MTx, sampling at 50Hz) attached (velcro straps) at the dorsal trunk lumbar level, and at thighs and shanks (center of mass, lateral surface) were used to record the segmental angular displacement. Trials were videotaped (camera at 4.0m distance vertical to the middle of the transition length, at 0.71m height, sampling at 50Hz). The step0 velocity (m/s) and length (m), and IT-RPA (deg) of the 4 steps before (walking: Wsteps) and 3 steps after (running: Rsteps) step0 were inserted in statistics. One-way repeated measures ANOVA (pairwise comparisons) tested the load effect on step0 velocity and length, and the IT-RPA of Wsteps and Rsteps. Trend analysis was used on IT-RPA of the Wsteps and Rsteps (SPSS v28, p≤0.05).
RESULTS:
In both load conditions, step0 velocity and length were decreased (p≤0.05) (velocity, NoL: 2.56±0.34 m/s, TrL: 2.46±0.36 m/s and ExtrL: 2.49±0.39 m/s, length, NoL: 1.16±0.15m, TrL: 1.10±0.16m and ExtrL: 1,13±0.14m) with no significant load effect on IT-RPA at any of the examined steps (p>0.05). Across all load conditions, Wsteps yielded a 3rd order polynomial trend (p≤0.05) (descending limb, step-4: 49.0±0.5m, step-3: 46.9±1.0m, step-2: 45.8±0.9m, ascending limb step-2 to step-1: 52.5±1.2m), and Rsteps a linear decreasing trend (p≤0.05) (step+1: 42.8±0.8m, step+2: 38.9±1.4m, step+3: 31.5± 0.8m). Step0 (48.4±0.9 m) was smaller and greater, respectively, than the last Wstep and the first Rstep (p≤0.05).
CONCLUSION:
The decrease of step0 velocity and length during the W-to-R transition, irrespective of the load condition, indicates their significant role as control parameters of the load attractor. The significant decrease of step0 velocity agrees with Diedrich et al. [2], however, the step0 length was not altered in their study. The increase and decrease, respectively, of IT-RPA in the NoL Wsteps and Rsteps, agrees with the dynamic theory [1] indicating the IT-RPA as potential order parameter for the overall W-to-R behavior which may be possibly modulated by the individual hip range of motion [3].

1) Diedrich & Warren, J Exp Psychol Hum Percept Perform, 1995
2) Diedrich et al., J Mot Behav, 1998
3) Wade et al., J Biomech, 2022