This study focused on understanding the gait phase-specific generation of global leftward linear and angular momenta during straight-line gait, pre-planned, and late-cued 90° leftward turns. Supporting our first hypothesis, we found a significantly greater change in COM leftward velocity (analogous to normalized linear impulse) and greater average COM leftward linear acceleration during right single support phase than during any other gait phase during all three tasks. Supporting our second hypothesis, we found a significantly greater leftward change in transverse-plane angular momentum (analogous to normalized angular impulse) and greater average transverse-plane moment during the double support phase with the left foot in front (left double support) than during any other gait phase during all three tasks.
These results suggest that during turns, changes in body-facing direction are initiated most during the left double support and that changes in center of mass trajectory occur more during right single support phase. This occurred despite a diversity in the number of footfalls and turn strategies used to execute the turn. This behavior during turns aligns with the understanding that during straight-line gait, the change in linear momentum is leftward during right single support phases24 and that the change in transverse-plane angular momentum is leftward during left double support.30 This alignment provides emerging evidence of momenta control strategies that are shared across straight-line and turning gait.
During the right single support, the body generates the greatest leftward COM linear impulse and greatest average leftward COM linear acceleration. To review, during straight-line gait, the right foot's single support phase generates a leftward linear impulse (change in linear momentum), which facilitates the center of mass translating towards the left foot.24 This leftward linear impulse is matched with an (approximately) equal and opposite rightward linear impulse during left single support because the goal is to maintain net zero change in right-left linear momentum over repeating gait cycles. During 90° turns to the left, the goal is to generate leftward linear momentum, until the COM trajectory is in the new desired direction of travel. In this study, right single support generated approximately two to four times greater leftward linear impulse during pre-planned and late-cued turns, respectively, vs. straight-line gait. During turns, the other three gait phases also generated leftward linear impulse, in agreement with medially-directed impulses reported during circular gait.22,26 This study uncovers that the right single support is the dominant phase for generating leftward linear momenta in straight line gait and turns performed by healthy young adults.
During the left double support, the body generates the greatest transverse-plane leftward angular impulse and greatest average moment about vertical vs. the other three gait phases. To review, during straight-line gait, the double support phase with the left foot in front generates the greatest leftward angular impulse (positive change in transverse-plane angular momentum)30, which facilitates the following right leg swing phase. This angular impulse is matched with an (approximately) equal and opposite change in angular momentum towards the right during right double support because the goal in straight-line gait is to maintain net zero transverse-plane angular momentum over large numbers of gait cycles. In contrast, during leftward turns, the goal is to generate additional leftward transverse-plane angular momentum to initiate changing the body’s facing direction (e.g., by ~ 90° during a 90° turn), then arrest this additional angular momentum when the desired facing direction of the body is achieved. We found that during straight-line gait and turns, the greatest transverse-plane leftward moment applied during left double support. This is reasonable, as both feet can coordinate to generate rotation.16,17,34 When both feet are in contact with the ground, not only is the base of support larger which may facilitate balance maintenance, but ground reaction forces can apply a moment without applying extraneous linear impulse that can adversely affect balance or COM linear trajectory direction. In other words, the linear impulse generated by each leg’s transverse-plane ground reaction forces can cancel across legs, though each force can work together apply a moment about a vertical axis passing through the COM (e.g., illustrative example in Fig. 1).16,17,34
Though comparing across tasks was not the central research question for this paper, there are important contrasts to share. As reviewed, the patterns of the greatest leftward linear impulse generated during right single support and the greatest transverse-plane leftward angular impulse generated during left double support were consistent across straight line gait, pre-planned, and late-cued turns. However, we also observed during turn conditions, all four gait phases contribute to leftward linear impulse, but not all gait phases contribute to leftward angular impulse. Though future research can answer why we observed this difference in linear and angular momenta control, we suspect that this can be attributed to the following contexts that may limit the generation of leftward body rotation to primarily be the left double support phase. For one, the demand to rotate the body’s facing direction is transient, while the need to generate and sustain a leftward COM velocity is maintained throughout the turn phase as the person continues to walk in the new direction. Second, maintaining balance while generating leftward body rotation about a vertical axis is facilitated by doing so during double support, when the base of support is larger than single support and when both legs can mitigate extraneous momentum. Third, rotating the body towards the left prior to left single support can facilitate the transition to the right lower extremity swinging forward towards its next footfall (right swing phase). The second and third contexts can limit the opportunity to generate leftward rotation of the body primarily to the left double support phase.
Within each turn condition, participants used a variety of footfalls that led to identifying both “step” and “spin” turn strategies, which is likely representative of what is used in daily life. In spin turns, the inside (left) foot was closest to the center of the intersection, and vice-versa for step turns.37 When we performed a preliminary analysis stratified by turn strategy, we found that during spin turns, leftward linear momenta generation was not significantly greatest during right single support. We offer possible explanations to be further explored in future studies. Spin turns were identified when the left (inside) foot was closest to the middle of the intersection. During spin turns, we found that these participants leaned such that their COM was often leftward of the left foot base of support (as in Fig. 2a; Fig. 7).2 This can relate to the finding in this study that right single support did not generate significantly larger changes in COM leftward velocity or average leftward COM acceleration than during left double support or left single support. When the COM is allowed to traverse left of the left foot’s base of support, as it did often during spin turns, the left single and double support phases can generate leftward linear momenta. Additionally, during late-cued turns, we found fewer (two to four) footfalls vs. three to five footfalls used in pre-planned turns.2 With fewer gait phases, during spin turns, the left single and double support phases could greatly contribute to generating leftward linear momenta due to the opportunity presented by the left foot’s placement within the center of the intersection (Fig. 7). In contrast, when the right foot was closest to the middle of the intersection during late-cued step turns (as in Fig. 1b, Fig. 7), the right single support can still be used as the primary gait phase to generate leftward linear momentum. Future research should be conducted to explore these possibilities. A limitation of our analysis stratifying by turn type includes that we observed 27% spin turns during late-cued turns, which could be limiting statistical power in comparing across gait phases within the spin turn strategy of late-cued turns. Thus, this stratified analysis and interpretation are considered preliminary.
This study has several limitations to note. First, although we expect left and right turns to be controlled similarly, we only examined left turns. We administered a visual late cue to turn, but future research can uncover how this compares to auditory cues to turn. It is unknown how turning the head towards a visual cue prior to the turn phase affects the overall transverse-plane leftward angular impulse generated during the turn phase. This study used blocks from straight-line gait to pre-planned and late-cued turns, but future research should randomize these blocks or trials. Also, our turn phase definition relied on pelvis rotation, when other preparatory adjustments may have preceded pelvis rotation, such as rotation of other body segments (like the head) or changes in the COM trajectory. Relatedly, although we explore how turns are initiated, our turn phase may consist of both turn initiation and termination, which may include arresting transverse-plane angular momenta generated earlier. Future research can compare how turn phases are defined and further parsed into sub-phases (initiation and termination, for example). Finally, we plan to expand our sample to include more females because this sample only included three females and was therefore too small to investigate if there were differences in the studied relationships between sexes.
This study opens many future research questions. For example, we would like to understand how the generation of transverse plane momenta relates to frontal-plane balance maintenance2 and how these momenta generation strategies are used by different balance-impaired populations. Future research will investigate temporal patterns in momenta control, for example, if generating the requisite momenta tended to occur in a sequence (e.g., rotation prior to translation). By more fully understanding the role of each gait phase towards the required generation of momenta during different types of turns, we can better assist people who struggle to balance during turns.
In summary, we found consistent transverse-plane momenta generation patterns across straight-line gait, pre-planned and late-cued turns. Specifically, in straight-line gait and turns, the greatest change in COM linear momentum and average COM acceleration towards the new direction of travel occurred during right single support. Additionally, during straight-line gait and turns, the greatest change in transverse-plane angular momentum and average moment to rotate the body to the left occurred during left double support. These patterns suggest that healthy young adults leverage straight-line gait momenta generation strategies during pre-planned and late-cued 90° turns.