Introduction:​ Examination of balance control has often been accomplished via evocation of stepping responses through external perturbations. These external perturbations can take the form of sudden underfoot surface translations, which are often comprised of controllable parameters including acceleration, velocity, and displacement. Interestingly, the values of these parameters incorporated within surface translation perturbations vary substantially within the literature. While this variance is understandable based on researchers’ research questions and infrastructure capacity, the systematic effect of perturbation characteristics on balance control responses during backwards surface translations is relatively understudied. Accordingly, the goal of this thesis was to improve the understanding how sudden posterior surface translation parameters affect spatial metrics of stepping responses and stability (study one) and to explore the effects of trial specific pre-perturbation participant-specific conditions on the same measures of balance control (study two).

Methods:​ Twenty-four young healthy adults (mean (SD):​ age 24.0 (3.61) years;​ height 1.71 (0.08) m;​ mass 73.2 (12.5) kg) with no history of balance impairment, recent musculoskeletal injury, or neurological disorder participated in the studies. Surface translations were initiated randomly during quiet stance in one of four directions (backward, forward, left, right). Platform acceleration values were varied from 1.0-3.5 m/s² (increments of 0.5 m/s²) while two platform peak velocity values (low and high) were implemented at each acceleration level. Displacement (0.30 m) and deceleration (5.0 m/s²) values were held constant across all perturbations. Backward translations (forward losses of balance) as well as single step responses were the focus of this thesis and thus the only trials analyzed. Dependent variables of normalized step length, maximum anteroposterior (AP) extrapolated centre of mass displacement (xCOM), and minimum AP extrapolated margin of stability (xMOS) were extracted from the kinematic data. Trial specific pre-perturbation values of underfoot weight distribution, AP centre of pressure (COP) location, ankle co-contraction index (CCI), AP COM location, AP COM velocity, and AP COM acceleration were extracted. In study 1, analysis of variance was used to analyze the effects of platform acceleration and velocity on the three dependent variables. In study 2, repeated measures stepwise linear regression was used to analyze the effects of the pre-perturbation factors on the predictive capacity of models predicting normalized step length and minimum AP xMOS.

Results:​ Study one demonstrated that increasing platform acceleration resulted in increased normalized step length and increased minimum xMOS (up to 30.7% and 90.4%, respectively), but only during high peak velocity trials. Increased platform velocity was also found to increase normalized step length and minimum xMOS by up to 26.8% and 127.6%, respectively. In contrast, participants’ xCOM displacement demonstrated a max increase of only 9.2% across acceleration levels. Study two identified both AP COM and COP position prior to perturbation as being the most commonly statistically relevant factors across perturbations. In comparison to models that incorporated variables accounting for the repeated measures within participants and external platform perturbation characteristics, participant factors at the moment of perturbation onset only increased model adjusted r² values from 0.612 to 0.646 (low velocity trials) and 0.661 to 0.689 (high velocity trials) for normalized step length. Minimum xMOS adjusted r² values were increased from 0.375 to 0.419 (low velocity trials) and 0.466 to 0.507 (high velocity trials).

Discussion/Conclusion:​ Variation in platform parameters resulted in significant changes to measures of step length and minimum xMOS. The increase in overall perturbation magnitude resulted in theoretically more stable responses (increased minimum xMOS) which was driven by the increased step length. As the external surface translation parameters are such important drivers of dynamic stepping responses, their effects should be considered when comparing studies which utilize different perturbation parameters. The statistically significant associations between personal pre-perturbation factors (particularly AP COM and COP location) on step length and xMOS align with mechanical models which suggest they play important roles in balance control. Interestingly though, these pre-perturbation factors explained only a small degree of variance beyond that provided by factors such as repeated measures and external perturbation characteristics. These two studies provide insights for researchers to more appropriately compare previous literature as well as provide recommendations for future study design during sudden support surface translations.