Elsevier

Journal of Biomechanics

Volume 38, Issue 9, September 2005, Pages 1886-1894
Journal of Biomechanics

A comparison of forefoot stiffness in running and running shoe bending stiffness

https://doi.org/10.1016/j.jbiomech.2004.08.014Get rights and content

Abstract

This study characterizes the stiffness of the human forefoot during running. The forefoot stiffness, defined as the ratio of ground reaction moment to angular deflection of the metatarsophalangeal joint, is measured for subjects running barefoot. The joint deflection is obtained from video data, while the ground reaction moment is obtained from force plate and video data. The experiments show that during push-off, the forefoot stiffness rises sharply and then decreases steadily, showing that the forefoot behaves not as a simple spring, but rather as an active mechanism that exhibits a highly time-dependent stiffness. The forefoot stiffness is compared with the bending stiffness of running shoes. For each of four shoes tested, the shoe stiffness is relatively constant and generally much lower than the mean human forefoot stiffness. Since forefoot stiffness and shoe bending stiffness act in parallel (i.e., are additive), the total forefoot stiffness of the shod foot is dominated by that of the human foot.

Introduction

This study quantifies the stiffness of the human forefoot during running and compares it with the bending stiffness of running shoes. The human stiffness results suggest a need for more sophisticated foot models. The shoe stiffness results pertain to shoe design, since a shoe's relative contribution to forefoot stiffness may affect running performance.

Human joints are often modeled as stiffness elements. Stefanyshyn and Nigg (1998) modeled the ankle joint as an activity-dependent stiffness element, and compared their experimental results with those reported by Davis and Deluca (1995), who measured ankle stiffness during the stance phase of walking. In their study, Stefanyshyn and Nigg defined stiffness as the joint moment divided by the angular displacement.

Isokinetic devices have been used to measure the stiffness of certain joints as a function of time (Hunter and Kearney, 1982). However, dynamic stiffness measurements have not been applied to the forefoot's metatarsophalangeal (MP) joints, which permit the bending of the toes relative to the metatarsal bones during push-off. Dozzi et al. (1989) investigated the MP joint power required during jumping in ballet and found that the muscles near the MP joint provide a great deal of power relative to the joint cross-sectional area.

Early research by Joseph (1954) focused on the first MP joint due to its superior size and strength. This study found that the range of motion of this joint was highly subject-dependent, but in all cases was much larger in dorsiflexion than in plantar flexion. This research further showed that the range of the MP joint in dorsiflexion is significantly greater than that of the interphalangeal joints, suggesting that the portion of the foot distal to the MP joints may be modeled as a single body.

Mann and Hagy (1979) found that the toe muscles are much more active in running than in walking, assisting in forward propulsion of the body. A similar study by Bojsen-Moller and Lamoreux (1979) correlated forefoot motion with gait. Hetherington et al. (1990) determined the amount of dorsiflexion required for forward propulsion to take place. Fulford et al. (1989) included inversion and eversion within their study of MP joint dorsiflexion. Lee (1997) and Davies (1999) investigated additional motions of the forefoot.

Foot models have been used for simulation and indirect force determination (Morlock, 1989). Stokes et al. (1979) proposed a two-dimensional MP-inclusive foot model to estimate the forces acting on the MP joint during normal walking. This model was developed further by Salathe and Arangio (1986) to include four rigid bodies. They had concluded that pronation and supination greatly affect the force distribution on the metatarsal heads. Morlock (1989) increased the number of rigid bodies in his MP-inclusive model to six and found that there was no benefit in increasing the degrees of freedom past six. Scott and Winter (1993) created a foot model involving 8 rigid bodies, but limited the degrees of freedom between two connecting bodies to one. Chan and Rudins (1994) used an MP-inclusive model to describe the biomechanics of the foot during running and walking.

The model used in this study is similar to those suggested by Stefanyshyn and Nigg (1997). They quantified the mechanical energy used by the MP joint during running and sprinting and determined that the MP joint was a large energy absorber. Their model considered the distal side of the MP as a single rigid body.

This paper is organized as follows. Section 2 describes the human MP joint experiment and presents the key results. Section 3 presents an experiment evaluating shoe bending stiffness about the MP axis as well as the results of these experiments and their implications. Conclusions are summarized in Section 4.

Section snippets

Human forefoot stiffness

In the analysis of the gait cycle, the foot is often modeled as a rigid body. However, this model is not suitable for analyzing the `push-off' phase of running, in which the MP joint plays a major role (Mann and Hagy, 1979, Morlock, 1989). This section describes an experiment to measure the dynamic stiffness of the MP joint during running.

Shoe bending stiffness

In this section, the effective shoe bending stiffness about the MP joint is found for four running shoes and compared with the forefoot stiffness determined in Section 2. The shoe bending stiffness acts in parallel to the forefoot stiffness, thus adding to it. Although runners seem to prefer flexible footwear, the need for flexibility within the sole of the shoe has yet to be proven scientifically. A sport shoe is designed such that the shoe can bend at least 30 degrees at a point just behind

Summary and conclusion

The stiffness of the human forefoot was measured for six subjects running barefoot. It was shown that as the joint deflects, the effective forefoot stiffness rises sharply and then decreases steadily. This indicates that the forefoot does not behave as a simple spring, but rather as an active mechanism with a highly time-dependent stiffness. The forefoot stiffness was compared with running shoe bending stiffness about the same axis. For each of the four shoes tested, the shoe stiffness was much

Acknowledgements

The authors wish to thank Dr. Benno Nigg and staff of the Human Performance Lab and the Health Sciences Centre at Foothills Hospital for their assistance and the use of experimental facilities. Funding support was provided by Biomechanigg Research Inc., NSERC, and the Alberta Heritage Foundation for Medical Research.

References (18)

There are more references available in the full text version of this article.

Cited by (66)

  • Does running speed affect the response of joint level mechanics in non-rearfoot strike runners to footwear of varying longitudinal bending stiffness?

    2021, Gait and Posture
    Citation Excerpt :

    Though MTPJ dorsiflexion was restricted, there was still an increase in shoe extension torque with increased LBS (Fig. 3). Because the foot and shoe act in parallel about the MTPJ axis [30], another reason for the lack of observed increase in MTPJ moment may be that the body is resolving optimal angular stiffness of the foot-shoe complex in-series, similar to deformation of the longitudinal arch and footwear midsole [31]. Shoe extension torque at maximum dorsiflexion accounts for one-third to one-half of total MTPJ moment (Fig. 3).

  • A comparison of metatarsophalangeal joint center locations on estimated joint moments during running

    2019, Journal of Biomechanics
    Citation Excerpt :

    Three separate methods using fixed joint center locations were defined according to previous studies (Fig. 1). These methods included joint center locations at the head of the fifth metatarsal (Bezodis et al., 2012; Stefanyshyn and Nigg, 2000, 1998, 1997), the midpoint between the first and fifth metatarsals (Hoogkamer et al., 2018; McDonald et al., 2016; Oh and Park, 2017; Oleson et al., 2005; Roy and Stefanyshyn, 2006; Willems and Gosseye, 2013), and at a point that was in the plane of the head of the first metatarsal in the anterior-posterior direction and the long axis of the foot in the medial-lateral direction (Miyazaki and Yamamoto, 1993; Willwacher et al., 2013), referred to henceforth as the second metatarsal head joint center. Kinetic estimations for the fifth metatarsal location were solved in a two-dimensional model.

View all citing articles on Scopus
View full text