Coordination of the ankle joint complex during walking

A simple model of the joint between the leg and the rearfoot is that of two hinges each with one degree of freedom. The superior hinge at the talocrural joint caters for most of the sagittal plane motion due to its mediolateral orientation. Motion in the transverse and coronal planes is largely taken up by the subtalar joint which has an inclination of 42 degrees to the transverse plane and deviates 23 degrees to the medial side. It could be expected from the orientation of this joint that rearfoot motion in the coronal plane (inversion/eversion) would be tightly coupled to motion in the transverse plane (adduction/abduction). A simple, non-viscous mechanical hinge joint would produce an in-phase, fixed gain relationship between inversion/eversion and adduction/abduction. This relation could be adequately quantified by Pearson's product moment correlation (r(P)). In contrast, muscular control of the joint, which introduces visco-elastic elements and time-delayed control lines and feedback loops, could be characterised by more complex phase and gain relations among the frequency components of the signals. These relations can be characterised by a transfer function relation (gain and phase angle) computed using a dynamic analysis. In the present study, both Pearson's correlation and linear systems analysis were employed. For 43 normal adults, five trials of one walking stride (stance and swing phases) were videographed at 30 Hz with three markers placed on each of the leg and rearfoot. Using inversion/eversion as the input and adduction/abduction as the output, the proportion of the total variance accounted for by a dynamic relation was measured by the coherence square function, which is analogous to the correlation squared (r(P)(2)). The mean r(P)(2) was 0.26, whereas the mean overall coherence was 0.62, thus accounting for more than twice the variance. This result suggests that while there may be a degree of simple mechanical coupling between inversion/eversion and adduction/abduction during walking, the major control is via muscular and/or visco-elastic passive elements.