Could Different Directions of Infant Stepping Be Controlled by the Same Locomotor Central Pattern Generator?

Abstract

This study examined the idea of whether the same central pattern generator (CPG) for locomotion can control different directions of walking in humans. Fifty-two infants, aged 2–11 mo, were tested. Infants were supported to walk on a treadmill at a variety of speeds. If forward stepping was elicited, stepping in the other directions (primarily sideways and backward) was attempted. The orientation of the infant on the treadmill belt determined the direction of stepping. In some infants, we also attempted to obtain a smooth transition from one direction to another by gradually changing the orientation of the infant during a stepping sequence. Limb segment motion and surface electromyography from the muscles of the lower limb were recorded. Most infants who showed sustained forward walking also could walk in all other directions. Thirty-three of 34 infants tested could step sideways. The success of eliciting backward stepping was 69%. Most of the infants who did not meet our backward stepping criteria did, however, make stepping movements. The different directions of stepping had similar responses to changes in treadmill speed. The relationship between stance and swing phase durations and cycle duration were the same regardless of the direction of stepping across a range of speeds. Some differences were noted in the muscle activation patterns during different directions of walking. For example, the hamstrings were much more active during the swing phase of backward walking compared with forward walking. The quadriceps was more active in the trailing leg during sideways walking. In some infants, we were able to elicit stepping along a continuum of directions. We found no discrete differences in either the electromyographic patterns or the temporal parameters of stepping as the direction of stepping was gradually changed. The results support the idea that the same locomotor CPG controls different directions of stepping in human infants. The fact that most infants were able to step in all directions, the similarity in the response to speed changes, and the absence of any discrete changes as the direction of stepping was changed gradually are all consistent with this hypothesis.