An Analysis of Muscle Co-contraction Strategies Using Dynamic Optimization and Electromyography While Walking on Uneven Surfaces

Abstract

Walking requires coordinated muscle activation, including co-contraction of opposing muscle groups, to maintain joint stability. The purpose of this study was to compare conventional measures of co-contraction with a novel muscle moment co-contraction index. Eighteen participants walked at 0.8 m/s and 1.6 m/s in three uneven footwear conditions: 1) 1 cm medial-posterior ridge (MPR), 2) 1 cm lateral-posterior ridge (LPR), 3) a no-ridge condition (NR). We estimated muscle forces using electromyography (EMG) constrained dynamic optimization. Co-contraction of the peroneus longus with the tibialis anterior as well as the soleus and gastrocnemius medialis was calculated during early and mid-stance using two co-contraction methods: (1) an EMG-Based co-contraction index, using normalized EMG only, and (2) an EMG-Track method, a novel muscle-moment-based index derived from dynamically optimized muscle forces. Overall, the mean co-contraction index predicted by the EMG-Track method was significantly lower than that predicted by the EMG-Based method. Significant interaction effects were observed using EMG-Track, but not EMG-Based (e.g., PL/MG early stance: F_(2,34)=6.779, p=0.003, =0.020). Our findings support a muscle moment-based approach to interpreting co-contraction, with implications for populations for whom co-contraction reflects stability during locomotion and/or overall locomotor function.