GROUND REACTION FORCE CHARACTERISTICS AND RUNNING ECONOMY

G.D. Heise (1), P.E. Martin (2), and P.S. Carroll (1)
(1) School of Kinesiology and Physical Education
University of Northern Colorado, Greeley, CO 80631
(2) Exercise and Sport Research Institute
Arizona State University, Tempe, AZ 85287

Presented at the 20th Annual Meeting of the American Society of Biomechanics
Atlanta, Georgia. October 17-19, 1996

INTRODUCTION

It was hypothesized that ground reaction force (GRF) characteristics that describe certain functional and mechanical requirements during stance would be positively correlated with the aerobic demand at a single speed of running (i.e., running economy). These characteristics describe potentially wasteful motion in terms of metabolic energy requirements. Recreational runners (n=16) ran on a treadmill at 3.35 m.s-1 for physiological measures and overground for biomechanical measures. Total vertical impulse, net impulses in three orthogonal directions, average vertical force, and descriptors of the free moment were correlated with . Total vertical impulse was the only GRF characteristic significantly correlated to (r=0.62). This relationship suggests that the greater overall muscle support requirements during ground contact are responsible for greater aerobic demand.

REVIEW AND THEORY

For a given aerobic activity, such as distance running, some individuals are more economical than others. Biomechanists recently have hypothesized that variability in the aerobic demand of running can be partly explained by measures that describe runners' contact with the ground. For example, Kram & Taylor (1990) presented a simple inverse relationship between aerobic demand and stance time. They suggested that "it is primarily the cost of supporting the animal's weight and the time course of generating this force that determines the cost of running" (p. 265). They examined different animal species over a range of running speeds.

During ground contact, a runner activates muscles for purposes of stability and for the development of forward momentum. These functional and mechanical requirements during stance are described by characteristics of the ground reaction force (GRF). Linear momentum is produced in the vertical and mediolateral directions as well as the forward direction, and could be considered wasteful motion in terms of metabolic energy requirements. Linear impulse measures the change in momentum and quantifies "the time course" of the GRF. In the present study, it was hypothesized that less economical runners (i.e., those with a higher aerobic demand) would exhibit greater support requirements during foot contact, as indicated by the total vertical impulse, and greater changes in momentum, as indicated by the net impulse. It was reasoned that a greater net impulse, or a greater change in linear velocity of the center-of-mass, would be metabolically wasteful. It also was hypothesized that the average vertical GRF, the rate of vertical force application, and the free moment would be correlated positively with aerobic demand.

PROCEDURES

Sixteen well-trained men performed treadmill running for determination of running economy and overground running for which biomechanical measures were determined (running speed = 3.35 m.s-1). Specifically, time of contact (tc), average vertical ground reaction force (Favg), the rate of vertical force application, linear impulses for each of three orthogonal directions (vertical, Fz; anterior-posterior, Fap; medial-lateral, Fml), and descriptors of the free moment (Mz') applied to the force platform were calculated.

Five successful, right foot contacts with an AMTI force platform were chosen for analysis for each subject. Mean values of GRF characteristics were then calculated. The rate of vertical force application was the slope of the initial, linear portion of the Fz- time curve. This portion of the Fz-time curve was manually selected with the aid of an interactive, computer-graphics program.

for vertical, anterior-posterior, and medial-lateral directions, respectively. The unit of measure for NVI was BW.s while other impulse measures were expressed in BW.ms.

Maximum and minimum values of the free moment applied to the force platform and the net angular impulse (NAIMz') associated with the free moment were determined and normalized based on calculations presented by Holden & Cavanagh (1991). Mean GRF characteristics were then correlated with submaximal VO2.

RESULTS AND DISCUSSION

Mean data and correlation coefficients between aerobic demand and GRF characteristics are shown in Table 1. The difference between the least and most economical subjects expressed as a percentage of the mean aerobic demand, 26.5%, agrees well with results of previous studies (see Martin & Morgan, 1992 for review). There existed no distinct pattern between subjects' VO2 max and their submaximal aerobic demand as was verified by the poor correlation between these two variables (r = -0.08). The most economical runner used only 59% of his aerobic capacity, whereas the least economical subject used 83%.

The magnitudes of GRF characteristics were similar to previously reported values (Holden & Cavanagh, 1991; Munro et al., 1987, Williams, 1980). TVI was the only GRF characteristic to be significantly correlated with aerobic demand. More economical runners exhibited a lower TVI. Positive correlation coefficients, although not significant, were also noted between aerobic demand and Favg, Fz rate, NAPI, and NMLI.

In the present study, the combined influence of Fz and the "time course" of that force application, as represented by TVI, explains a significant amount of inter-individual variability in running economy. Kram & Taylor (1990) examined several speeds of running-type gait (e.g., trotting, galloping) in different species of animals and reported a significant relationship between tc and aerobic demand. In contrast, tc was not related to aerobic demand in the present study (see Table 1). Williams (1980) showed no trend between low, medium, and high economy groups and TVI.

TVI can be considered an indication of overall muscular support during ground contact. Differences in muscle activation patterns during stance may help explain the significant relationship reported here. A recent study that examined this question found a trend that indicated economical runners exhibited more coactivation between two-joint muscles of the leg during stance (Heise et al., 1996). These neuromuscular differences in runners need to be examined further.

Of the impulse measures that reflected momentum changes, NAPI and NMLI exhibited the expected positive correlation coefficients with VO2 , however, the relationships were not statistically significant. It is well documented that the mediolateral component of the GRF is quite variable (e.g., Miller, 1990).

The variables describing the free moment applied to the platform historically have not been examined by researchers examining GRF characteristics. Here it was hypothesized that any force application that detracted from forward motion may be metabolically wasteful. The free moment, thought of as a "twisting moment" applied by the foot to the ground does not influence aerobic demand. Its low magnitude prior to normalization (mean NAIMz' = 1.58 N.m.s) may partly explain the lack of a relationship.

In conclusion, only one of the GRF characteristics (TVI) exhibited a significant, positive correlation with aerobic demand. Several other correlation coefficients were positive as hypothesized, but tc, NVI, and descriptors of the free moment showed no correlation with VO2.

REFERENCES

Holden, J.P. & Cavanagh, P.R. J. Biomech, 24, 887-897, 1991.

Heise, G.D. et al. Int. J. Sports Med., 17, 128-133, 1996.

Heise, G.D. & Martin, P.E. ASB Proceedings, Palo Alto, CA, pp. 111-112, 1995.

Kram, R. & Taylor, C.R. Nature, 346, 265-267, 1990.

Martin, P.E. & Morgan, D.W. Med. Sci. Sports Exerc., 24, 467-474, 1992.

McMahon, T.A. & Cheng, G.C. J. Biomech., 23(Supp), 65-78, 1990.

Miller, D. In Biomechanics of Distance Running (Edited by P.R. Cavanagh), Human Kinetics, Champaign, IL, 1990.

Munro, C.F. et al. J. Biomech., 20, 147-155, 1987.

Williams, K. Ph.D. dissertation, The Pennsylvania State University, 1980.