Humans and other legged animals meet locomotor challenges by modifying the frequency, amplitude, and segment interactions of rhythmic flexion/extension patterns (Grillner, 2011). The coordinated patterns of spatiotemporal interaction between lower limb segments determine the expression of gait (Alexander, 1989, Diedrich and Warren, 1995). Modern approaches to locomotor control analysis employ tools from dynamic systems such as the relative phasing of limbs to study coordination and changes with disease, injury, and skill (van Emmerik et al., 2016). Many investigations have studied walking and jogging, leaving coordination in high-speed running relatively unexplored. Understanding differences in coordination in sprinting may provide valuable insight into the coordinative constraints at the limits of running performance.

The mechanics of high-speed running have been studied extensively and gives insight into how coordination might change with speed and skill. Faster steady-speed running is primarily accomplished via higher step frequencies, mostly due to decreases in ground contact time (GCT), as flight time (FT) plateaus at faster speeds (Nummela et al., 2007; Weyand et al., 2000). Reductions in either GCT or FT necessitate faster limb movement, and reduction in GCT requires greater peak vertical ground reaction forces (GRFs), increased vertical stiffness, and changes in limb contact geometry (Blickhan, 1989). Substantial mechanical demands accompany reduced GCT, as peak vertical GRFs can exceed four times bodyweight in sprinters (Clark and Weyand, 2014). Although swing and stance phases have distinct requirements, they are mechanically linked as GRFs applied in stance depend on limb repositioning in swing (Clark et al., 2017, Clark et al., 2020). Recent research has highlighted differences in swing-leg mechanics in faster versus slower runners, including position of the swing thigh at contralateral limb touchdown (Miyashiro et al., 2019, Rottier and Allen, 2021). Thus, determinants of sprinting performance include the ability to apply large vertical forces during brief contact times, and the ability to rapidly reposition the limbs between ground contacts. Changes in contact geometry and stiffness imply a role for coordination. More importantly, differences in swing-leg mechanics with training directly associate coordination with faster sprint speeds. But how can coordination be measured and interpreted?

The classic approach to coordination is to model the limbs as coupled pendulums, where relative phasing of limb segments serves as an order parameter (characterizing coordination) and movement speed acts as the control parameter (Jirsa and Kelso, 2004, Schoner et al., 1990, Turvey, 1990). Relative phasing of intralimb segments captures coordination of within-limb segments, such as the thigh and shank. This type of analysis has been utilized to measure change in intralimb coordination at the walk-to-run transition (Diedrich and Warren, 1995) and coordination during the clearance of vertical obstacles while running (Stergiou et al., 2001). Although speed and task dependent, intralimb coordination characterizes the coordinated behavior of the leg as it reorients during swing and meets the force requirements during stance, as well as the transition from inverted pendulum mechanics in walking to spring-mass mechanics in running (Diedrich and Warren, 1995). Relative phasing of interlimb segments (e.g. between thigh segments) captures the symmetric motion of the limbs during the gait cycle. In walking and running, this manifests as some degree of antiphase coordination. Presumably, highly antiphase coordination could simplify control (Raibert, 1986), though the degree to which antiphase coordination patterns are preserved during bipedal locomotion may depend on the specific task constraints (Haddad et al., 2010, Reisman et al., 2005).

Limb coordination at faster running speeds has not been explored as comprehensively as walking and submaximal running. Recent research on control in sprint running has shown that sprinting speeds use less functional muscle groupings (synergies), thus reducing movement complexity to optimize forward velocity and stabilize gait (Santuz et al., 2020). Using deviation in limb relative phasing, Wang et al. (2021) showed that athletes demonstrated greater variability in coordination patterns at maximum speed compared to non-athletes, although the actual nature of coordination was not reported. Thus, the nature of interlimb and intralimb relative phasing remains to be characterized at faster running speeds. Similarly, whether sprint-trained individuals differ in coordination compared to those with other athletic backgrounds at such speeds remains unknown. Evaluating the changes in coordination patterns across the spectrum of running speeds, and at the limits of maximal sprinting, would provide insight into locomotor control. Particularly, whether track and field athletes trained to optimize forward locomotor speed differ in their limb coordination from other athletes, would provide insight at the intersection of skill and control.

Here we evaluated the interlimb and intralimb coordination of leg segments during stance and flight/swing as a function of running speed in different athletic populations. Our questions were three-fold. First, how does coordination change as a function of speed from submaximal to maximal speed? For all limb couplings examined, we predicted increasingly antiphase coordination with faster speeds, with changes being more pronounced in stance than during swing due to differential changes in stance duration with speed. Second, are there differences in coordination and the coordination-speed relationship between trained and untrained sprinting populations? We hypothesized that trained track and field athletes would use more antiphase coordination compared to untrained counterparts at slower speeds, resulting in less change in coordination with speed compared to the untrained group. Third, how does coordination differ at the individual limits of running performance? Similar trends were expected across groups but, due to faster speeds in track and field athletes, it was hypothesized that track and field athletes would display more antiphase coordination at faster top speeds.