A synergy perspective on gait – over-ground vs. treadmill walking

INTRODUCTION: Increased stride-to-stride time variability is reported among elderly fallers and various patient groups [1]. Variability is therefore often regarded as an indicator of gait deficits. However, movement variability is also a general and natural phenomenon. A synergy perspective on movements has proposed that elemental and performance variables may represent good and bad components of variability [2].
We suggest that the gait pattern can be regarded as a movement synergy in which medio-lateral deviation in one stride can be corrected during the next stride (the elemental variables). Such corrections ensure a straight gait path (the performance variable).
AIM: The aim of this study was to apply a synergy approach to gait analysis by comparing over-ground and treadmill walking. The treadmill was hypothesized to demand a less variable walking path resulting in a larger good/bad variability ratio.
METHODS: Eight young subjects participated in the study. They walked over-ground down a 200 meter hallway and on a treadmill at preferred gait speed. A tri-axial accelerometer (Xsens) was fixed at the lower back of the participant by a belt around the pelvis. The gyro-corrected medio-lateral acceleration signal was summed up for each stride. Each acceleration stride sum was plotted against the subsequent stride sum in a coordinate system. Variability was evaluated in diagonal directions in the plot. Good variance was evaluated with respect to a straight line with a positive slope going through the mean of the strides, and bad variance with respect to a similar line with a negative slope. The good/bad variance ratio was calculated and the difference between treadmill and over-ground walking was finally evaluated.
RESULTS: The good/bad variance ratio for over-ground walking was 1.7 (CI95%: 1.5-2.0). When walking on the treadmill the ratio increased significantly to 2.4 (CI95%: 2.3-2.5); (p<0.01). The normal variance did not change significantly during the treadmill walking (p=0.46).
CONCLUSION: The good/bad ratio for stride-to-stride variability was larger than 1.0. This indicates the pattern of a synergy. The pattern was emphasized during treadmill walking reflecting construct validity of the measure. The synergy approach to gait variability may provide a new way to assess gait variability.


References
[1] Hausdorff JM. Gait variability: methods, modelling and meaning. J Neuroengineering Rehabil 2005 Jul 20;2:19.
[2] Latash ML. Synergy. New York: Oxford University Press; 2008.