Adaptation to large-magnitude treadmill-based perturbations: improvements in reactive balance response

Abstract

We aimed to examine the trial-to-trial changes in the reactive balance response to large magnitude slip-like treadmill perturbations in stance and whether the acquired adaptive changes could be appropriately scaled to a higher intensity perturbation. Seventeen young adults experienced 15 slips for training on level I intensity. Pre- and post-training slips were delivered at a higher intensity (20% > level I). Pre- and post-slip onset stability (at liftoff and touchdown of stepping limb) was measured as the shortest distance of the center of mass (COM) position (XCOM/BOS) and velocity (ẊCOM/BOS) relative to base of support (BOS) from a predicted threshold for backward loss of balance. The number of steps to recover balance, compensatory step length and peak trunk angle were recorded. The post-slip onset stability (at liftoff and touchdown) significantly increased across the trials with no change in preslip stability. Improvement in stability at touchdown positively correlated with an anterior shift in XCOM/BOS but not with ẊCOM/BOS. Consequently, the number of steps required to recover balance declined. The adaptive change in XCOM/BOS resulted from an increase in compensatory step length and reduced trunk extension. Individuals also improved post-slip onset stability on a higher intensity perturbation post-training compared with the pre-training trial. The results support that the CNS adapts to fixed intensity slip-like perturbations primarily by improving the reactive stability via modulation in compensatory step length and trunk extension. Furthermore, based on prior experience from the training phase, the acquired adaptive response can be successfully calibrated to a higher intensity perturbation.

Keywords: Compensatory stepping; fall prevention; learning; slips.

Figure 1

(A) Schematic representation of the experimental set up and (B) compensatory stepping response to slip-like perturbation of a representative subject during the time events of B1) slip onset, B2) liftoff of the stepping limb (LO), and B3) touchdown (TD) of the stepping limb. During LO and TD, the initial position of the belt marker is indicated by the empty circle and the arrow represents direction of the belt displacement.

Figure 2

Figure showing (A) Experimental protocol – trials S2 to S16 constituted the training session, and trials S1 and S17 were the pre- and post-training slips delivered at a higher intensity compared to the training session. The S0 and S1 trials were separated by a regular walking trial (NS). (B) A typical trajectory of belt displacement during perturbation (top) and center of mass displacement relative to base of support normalized to the foot length (X_COM/BOS/foot length, bottom).The vertical lines represent the events of belt displacement onset (dotted) and liftoff of the stepping limb (solid).

Figure 3

Figure showing (A) adaptation in proactive, preslip onset (Pre-SO) and reactive, post-slip onset stability at liftoff (Post-LO) and touchdown (Post-TD) of the stepping limb for trial 1 (T1), trial 10 (T10), and trial 15 (T15) and (B) relationship between the stability at touchdown (TD) and the center of mass position and velocity relative to the base of support (X_COM/BOS and Ẋ_COM/BOS respectively). The stability was defined as the shortest distance of the instantaneous COM state (position and velocity) from a theoretically predicted threshold for backward loss of balance. Stability values <1 indicate instability in the backward direction. Significant differences are indicated by *P < 0.05.

Figure 4

Scatter plot showing the relationship between stability at touchdown (X_COM/BOS at TD) with (A) compensatory step length and (B) trunk angle. Significant associations between the variables are denoted by *P < 0.05.

Figure 5

Figure demonstrating mean differences in post-slip onset stability between the pre-training (Pre) and post-training trials on a higher intensity perturbation. *P < 0.05 represents the differences in stability at (A) liftoff (LO) and (B) touchdown (TD).