Presented at NACOB 98: North American Congress on Biomechanics Canadian Society for Biomechanics - American Society of Biomechanics University of Waterloo Waterloo, Ontario, Canada August 14-18, 1998

---

CONTROL OF FRONTAL PLANE BODY MOTION DURING
INDUCED FORWARD STEPPING

M.W. Rogers ^1,2 , T.D. Cain ³ , T.A. Hanke ^4
1 Programs in Physical Therapy and ² Department of Physical Medicine and Rehabilitation,
Northwestern University Medical School, Chicago, IL 60611
³ Geriatric Medicine Associates, Denver, CO 80218
⁴ Marianjoy Rehabilitation Hospital and Clinics, Wheaton, IL 60189

INTRODUCTION

This study investigated the strategies used by healthy younger adults to control body motion in the frontal plane during induced forward stepping for balance recovery. These responses provide a basis for examining age-associated changes in neural and musculoskeletal factors related to postural balance dysfunction in the elderly.

REVIEW AND THEORY

The initiation of protective stepping to alter the relationship between the body center of mass (COM) and the base of support (BOS) is a common solution for balance recovery following externally applied postural disturbances. A fundamental problem common to both the initiation of gait and protective stepping is the control of frontal plane body motion during transitions from bipedal to single limb support. Previous studies of "volitional" step initiation (e.g., Breniere et al., 1987) have consistently observed preparatory postural adjustments that minimized the tendency for the body to fall laterally beginning at liftoff by reducing in advance the medial location of the COM relative to the BOS. However, imposed constraints on the BOS-COM relationship during perturbation induced stepping might require alternative strategies for controlling lateral body motion. For example, limitations in the time to successfully recover balance by stepping could reduce the effectiveness of an anticipatory strategy. This might require execution adjustments in the BOS through medio-lateral (M-L) foot placement as one alternative control strategy (MacKinnon, et al., 1993). Accordingly, this study examined the operational characteristics of COM motion in the frontal plane in relation to the response features of induced forward steps. We hypothesized that subjects would adapt their control of lateral body motion in response to imposed alterations in the positional and timing constraints on balance recovery during induced forward stepping.

PROCEDURES

Twelve healthy (nine females and three males) adults (mean age = 31years ± 7 S.D.) participated in the study. Subjects stood with each foot on a separate force platform (AMTI) in standardized positions. A visual display of the BOS area and net center of pressure (COP) was provided. Data collection began only after subjects positioned the COP in a target region (± 20% of the antero-posterior and M-L support area) for a 500 ms period. Stepping was evoked by a closed-loop stepper motor waist-pull system that allowed for independent control of pulling acceleration, velocity, and displacement (Pidcoe et al., in press). The system included a linear drive table, cable and pelvic belt connection, electromechanical latch with floating connection and spring pre-tensioner, in-line load cell, and position transducer. A safety harness was worn. Forward pulls were administered at five different combinations of pulling displacement (4.5 to 22.5 cm), velocity (9.0 to 45.0 cm/s), and acceleration (180.0 to 900.0 cm/s/s). Blocks of three trials at each of the five levels of pulling magnitude were randomly presented. Subjects were instructed to "react naturally" in response to the pulls.

A two-camera video based 60 Hz motion analysis system (Peak Performance Co.) recorded the 3-D positions of reflective markers placed bilaterally on the foot, ankle, knee, hip, shoulder, and wrist. A nine-segment model was used to estimate the motion of the COM. Central difference methods were applied to compute velocities. All data were digitally sampled at 120 Hz for five seconds during each trial. An ANOVA was used to determine the effects of pulling magnitude on kinematic variables. Relationships among pairs of variables were evaluated by linear regression.

RESULTS

The incidence of stepping, and the forward COM and initial step displacement and velocity (all comparisons: p < .05), increased between the lowest level pull magnitude (P1) and the highest level (P5). Only one subject stepped during one trial at P1.The mean displacements of the initial step foot are represented in Figure 1. Overall, subjects stepped medially towards the midline of the body. A significant (p< .01) effect of pull condition for P3 to P5 at which all subjects stepped indicated a more lateral placement for the lower intensity P3 pulls versus P4 (p < .03) and P5 (p < .01). P2 steps taken by eight subjects approximated the P3 foot placement. The M-L step displacement was highly significantly related to COM lateral displacement at touchdown (r = 0.57, r ² = 0.32, p < .001). Conditions P3 to P5 displayed very small (< 1 cm) initial COM displacements towards the upcoming stance side prior to liftoff. Only for P2 did peak preparatory displacements increase (mean = 1.3 cm ± .22 S.E.M.). COM motion towards the initial step side at touchdown was reduced (p < .02) for P2 versus P3 to P5 which were comparable (p > 05).

Figure 1: Group mean displacements of the initial step foot in the antero-posterior (A-P) and medio-lateral (M-L) directions relative to the mean starting position as a function of pulling magnitude.

Highly significant (p< .001) linear correlations were found between the COM frontal plane velocity at swing limb liftoff and the COM displacement at touchdown (r = 0.78, r ² = 0.62), and between the frontal plane angular displacements of the stance leg and the trunk recorded at touchdown (r = -0.73, r ² = 0.53).

DISCUSSION

The foot placement results indicated that subjects adapted their responses in relation to the magnitude of postural disturbance not only in the sagittal plane (Maki et al., 1996), but also in the frontal plane. Thus, steps were directed more medially for the larger disturbances than for the smaller disturbances. As foot medial displacement increased the COM displacement to the step side decreased. An anticipatory strategy for controlling lateral body motion was apparently not effectively used. However, the smaller COM displacement to the step side at touchdown during the P2 condition suggested some effectiveness in that case since subjects also stepped more laterally. Similar to "volitional" stepping (Lyon, et al. ,1997), the COM velocity at liftoff was related to the subsequent lateral "fall": the greater the velocity toward the swing side the greater the COM displacement to that side at touchdown. Given the limited preparatory COM motion, it is likely that a stabilizing action of the single stance hip musculature contributed to regulating the COM momentum (unpublished observations). The counter-rotation of the trunk suggested an ongoing postural correction for the tendency for the body to "fall" about the stance foot (MacKinnon, et al., 1993), and/or for maintaining a stable head position. Overall, the results indicated an interaction of several processes for controlling COM lateral motion.

REFERENCES

Breniere Y. et al., J. Motor Behav., 19, 62-76, 1987.

Lyon I. et al., Exp. Brain Res., 115, 345-356, 1997.

MacKinnon C. et al., J. Biomech., 26, 633-644, 1993.

Maki B. et al., J. Biomech., 29, 343-353, 1996.

Pidcoe P. et al., J. Biomech., in press.

ACKNOWLEDGMENTS

This research was supported by the NIA, Grant # K01 AG 00581. The assistance of F. Gao, L. Hedman, S. Redman, and J. Zhang is gratefully acknowledged.