Dance training improves the CNS's ability to utilize the redundant degrees of freedom of the whole body

Humans maintain balance effectively despite internal and external perturbations. The central nervous system (CNS) coordinates multiple body segments to stabilize whole-body actions. Motor synergies, task-specific patterns of multi-DoFs, are proposed as a CNS control mechanism. The UCM approach quantifies DoF organization and coordination by analyzing variability in task-relevant and task-irrelevant spaces. In redundant motor systems, task-irrelevant variability, reflecting CNS solutions, is contrasted with task-relevant variability (error). Studies using UCM analysis in healthy populations show greater task-irrelevant than task-relevant variability, suggesting flexibility in DoF utilization. Pathological groups often exhibit reduced task-irrelevant variability. Professional dancers, known for superior balance control, may offer insights into CNS control mechanisms. While expert motor task performance has been studied, how dancers coordinate segments for whole-body stabilization remains largely unknown. Dynamic postural control involves maintaining the whole-body center of mass (CoM) over the base of support (BoS). The extrapolated center of mass (xCoM), a velocity-adjusted CoM, is often used in dynamic tasks. This study investigates dancers' ability to control multiple segments during a mechanical perturbation, analyzing two phases: xCoM within and outside the BoS. We hypothesized that dancers would show smaller task-relevant variability and greater task-irrelevant variability than non-dancers, particularly in the unstable phase.

Existing research highlights the CNS's role in coordinating multiple body segments for balance control, often explained by the concept of motor synergies. The Uncontrolled Manifold (UCM) analysis is a prominent method to quantify these synergies by separating variability into task-relevant and task-irrelevant components. Studies in healthy individuals generally show a predominance of task-irrelevant variability, reflecting flexibility in utilizing redundant degrees of freedom. Conversely, pathological populations often exhibit reduced task-irrelevant variability. While expertise in motor tasks has been extensively researched, the specific coordination strategies employed by dancers, known for their remarkable balance, are less understood. Previous studies have investigated postural control using different methods, including analysis of the whole-body center of mass (CoM) and its relationship to the base of support (BoS). However, the CNS control strategies underlying multi-segmental synergistic patterns in dynamic behaviors like perturbation-evoked stepping have not been thoroughly explored.

Ten female professional dancers (15.8 ± 4.26 years of dance training) and ten female non-dancers participated. Nineteen reflective markers were placed on various body segments. Participants stood barefoot in front of a custom-made pulling apparatus, with a belt around their waist connected to a spring mechanism. A pulling impulse was delivered at random times within a 5-second period, inducing a forward step. Whole-body kinematics were recorded using a motion capture system at 100 Hz. Data analysis was performed using customized MATLAB code. The extrapolated center of mass (xCoM) was calculated using a linearized inverted pendulum model, incorporating whole-body CoM position and velocity. Whole-body CoM and velocity were computed as weighted sums of individual segment CoMs and velocities. Relative individual segmental xCoMs (iRxCoM) were calculated by subtracting proximal segment xCoM from distal segment xCoM, yielding independent variables for UCM analysis. The UCM analysis decomposed iRxCoM into task-relevant and task-irrelevant components. Two phases were defined: Phase I (xCoM inside BoS) and Phase II (xCoM outside BoS). Mean squared deviations (MSD) were calculated for task-relevant (MSDTR) and task-irrelevant (MSDTIR) spaces. The index of synergy (SYN) was computed to quantify multi-segmental coordination. Two-way repeated measures ANOVA was used, with factors Group (dancers vs. non-dancers) and Segment (14 segments). Post-hoc tests with Bonferroni correction were applied. Individual segment contributions to MSDTR, MSDTIR, and SYN were also analyzed.

All participants took one forward step in response to the perturbation, with the stepping foot varying among individuals. No significant differences were found between dancers and non-dancers in the medial-lateral (ML) direction during either phase. In the anterior-posterior (AP) direction, no differences were observed in Phase I (xCoM within BoS). In Phase II (xCoM outside BoS), dancers exhibited significantly greater MSDTIR (F1,18 = 5.986, p = 0.025) than non-dancers. This was supported by a significant Group × Segment interaction (F1,18 = 2.399, p = 0.005). Individual segment comparisons showed significant differences in iMSDTIR at the head, trunk, contralateral side (CS) thigh, and ipsilateral side (IS) shank. There was no significant difference in MSDTR between groups. Dancers also showed significantly greater SYN (F1,18 = 5.152, p = 0.036) than non-dancers in Phase II, with a significant Group × Segment interaction (F1,18 = 3.069, p < 0.001). Pairwise comparisons revealed significant differences in SYN at the head, CS thigh, and IS shank. Figure 3A displays the mean MSDTR, MSDTIR, and SYN for both groups, while Figure 3B shows the percentage contribution of individual segments to each variable. The greater MSDTIR and SYN in dancers were mainly attributed to higher values in these variables at the head, trunk, CS thigh, and IS shank.

The findings of greater motor synergy (SYN) in dancers during Phase II suggest distinct postural control strategies compared to non-dancers, effectively exploiting redundant DoFs. The increased task-irrelevant deviation (MSDTIR) indicates that dancers utilize greater variability in individual segment movements while maintaining a consistent whole-body CoM position. The absence of differences in task-relevant deviation (MSDTR) suggests that both groups maintained similar whole-body CoM trajectories. While some studies show superior balance control in dancers, our results don’t directly demonstrate superior balance performance due to similar MSDTR. Our results align with studies showing experts in various fields utilizing motor equivalence with greater task-irrelevant variability without compromising performance. The greater MSDTIR in dancers at the head, trunk, CS thigh, and IS shank supports the idea of compensatory mechanisms for head-trunk coordination and the importance of hip and ankle joints in postural control, consistent with previous research. This study provides evidence of superior body segment coordination and dynamic postural control in dancers, particularly during unstable conditions. The increased task-irrelevant variability suggests greater flexibility in DoF utilization, which could be beneficial in handling unexpected perturbations and performing secondary tasks.

This study introduces a novel approach to quantify kinematic synergy during postural control, focusing on within-trial variability. Dancers demonstrated superior coordination ability during dynamic postural control, particularly reflected in increased task-irrelevant variability at specific body segments. This suggests that long-term dance training enhances the CNS's ability to exploit redundant DoFs for whole-body stabilization. Future research could investigate trial-to-trial movement variability to further elucidate CNS control mechanisms in dancers.

The UCM analysis used in this study does not incorporate segment orientation information, which might play a role in postural coordination, especially during rotational movements. While the study uses segment CoMs for whole-body CoM estimation, which may be more accurate than using joint angles, it still involves estimations of inertial properties that could propagate errors. Future studies could incorporate more sophisticated methods to address these limitations.