Distinct sets of locomotor modules control the speed and modes of human locomotion.

Hikaru Yokoyama,Noritaka Kawashima,Masahiro Shinya,Kimitaka Nakazawa,Tetsuya Ogawa

doi:10.1038/srep36275

Hikaru Yokoyama, Noritaka Kawashima + Show 3 more

Open Access

https://doi.org/10.1038/srep36275

Copy DOI

Abstract

Although recent vertebrate studies have revealed that different spinal networks are recruited in locomotor mode- and speed-dependent manners, it is unknown whether humans share similar neural mechanisms. Here, we tested whether speed- and mode-dependence in the recruitment of human locomotor networks exists or not by statistically extracting locomotor networks. From electromyographic activity during walking and running over a wide speed range, locomotor modules generating basic patterns of muscle activities were extracted using non-negative matrix factorization. The results showed that the number of modules changed depending on the modes and speeds. Different combinations of modules were extracted during walking and running, and at different speeds even during the same locomotor mode. These results strongly suggest that, in humans, different spinal locomotor networks are recruited while walking and running, and even in the same locomotor mode different networks are probably recruited at different speeds.

Highlights

12 345 muscles time study, we applied the electromyographic (EMG) signal decomposition technique for extracting motor modules[20,21], i.e., spatially fixed locomotor muscle synergies (Fig. 1)
Changes in the EMG patterns depending on locomotor speed and mode were roughly divided into three types based on visual inspection, with the exception of a few others (TA: tibialis anterior, RF: rectus femoris RF, AM: adductor magnus)
The peak activation level gradually increased with increasing speed regardless of locomotor mode, while the activation timing was nearly constant throughout the majority of proximal leg muscles (GM: gluteus maximus, VL: vastus lateralis, VM: vastus medialis, BF: biceps femoris, ST: semitendinosus) and the trunk muscles (RA: rectus abdominis, erector spinae (ES): elector spinae)

Summary

Introduction

12 345 muscles time study, we applied the electromyographic (EMG) signal decomposition technique (nonnegative matrix factorization; NMF) for extracting motor modules[20,21], i.e., spatially fixed locomotor muscle synergies (Fig. 1). Regarding speed-dependency in running, it has been shown that similar temporal activity patterns of modules were utilized in a slow speed range (5–12 km/h [1.39– 3.33 m/s], considered as jogging30,32,33), the differences in the organization of modules (i.e., muscle weightings) are not well established. It remains unknown whether walking modules are truly unchanged regardless of the speed, and whether or not there are faster running modules in humans. The acceptance of this working hypothesis would provide indirect evidence of mode- and/or speed-dependency of neural networks in human locomotion

Objectives

Methods

Results

Conclusion