Study at a glance

In 29 typically developing children spanning toddlers to adolescents, Theia3D-derived kinematic and ground reaction force data showed that the body's "virtual pivot point," a key marker of upper-body stabilization during walking, is present even in children who have only recently learned to walk, but its position and variability change substantially with age.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor's note: Theia3D version not specified in the paper. This study uses Theia3D as a measurement tool to derive center-of-mass trajectory for calculating ground reaction force focusing behavior, rather than evaluating Theia3D's own kinematic accuracy.

Study Overview

Design

A retrospective, cross-sectional study comparing virtual pivot point position, its step-to-step variability, and the degree of ground-reaction-force focusing toward it, across three pediatric and adolescent age groups using synchronized markerless motion capture and force plate data.

Participants

Twenty-nine typically developing children drawn from a motion laboratory's normative reference database, divided into three groups: one-year-olds who had learned to walk independently within the prior 12 months (mean age 1.5 years), two- to three-year-olds who had been walking independently for more than 12 months (mean age 3.0 years), and a ten- to fifteen-year-old adolescent reference group with adult-like gait patterns (mean age 12.3 years). Inclusion required age-appropriate development and no history of orthopedic or neurological conditions affecting gait; each child additionally underwent a physical movement assessment performed jointly by two experienced movement specialists trained in gait analysis.

Data Collection

Participants walked barefoot at a self-selected comfortable speed in tight-fitting clothing. Kinematic data were collected using 11 synchronized Qualisys Miqus cameras (100 Hz) processed through Theia3D, which used deep-learning-based computer vision to identify up to 124 anatomical landmarks and reconstruct them in 3D; these data were exported to Visual3D to build each child's biomechanical model and calculate center-of-mass position. Ground reaction forces were recorded synchronously at 1000 Hz using two floor-integrated force plates. The virtual pivot point was calculated in the sagittal plane during the single support phase of each step as the point minimizing the perpendicular distance from the measured ground reaction force vectors, with the strength of this convergence quantified using a coefficient of determination (R²).

Analysis

Group differences in walking speed, relative single support phase duration, ground-reaction-force focusing (R²), and normalized horizontal and vertical virtual pivot point position were assessed using one-way ANOVA with post hoc testing, with effect sizes reported as eta squared. Levene's test assessed variance homogeneity between groups, and 95% confidence ellipses characterized the spatial variability of virtual pivot point position within each group. Step-to-step variability in vertical virtual pivot point position was additionally correlated with age in months for the two youngest groups using Pearson's r.

Key Findings

  • Ground reaction forces converged strongly toward a single intersection point across all three age groups, with group-mean R² values exceeding 0.95 in every group and no significant age-related difference in this focusing behavior. This indicates that the basic mechanism underlying virtual-pivot-point stabilization is already present in children who have only recently learned to walk independently.
  • The normalized vertical position of the virtual pivot point increased progressively and significantly with age, from 7.58 cm in one-year-olds to 11.43 cm in two- to three-year-olds to 14.79 cm in the adolescent reference group (moderate effect size), with the one-year-old group's position roughly half that of the adolescent group.
  • In several one-year-olds, and occasionally in two- to three-year-olds, the virtual pivot point was located at or below the center of mass, a pattern not observed in any adolescent participant, where the point was consistently positioned above it. A pivot point located above the center of mass is thought to generate a passive restoring moment that helps stabilize the upper body, while one at or below the center of mass does not contribute to this passive mechanism.
  • Variability in vertical virtual pivot point position was substantially higher in the youngest group and decreased progressively with age, showing a strong negative correlation with age in months among the two youngest groups combined (r = -0.586, p < 0.05), indicating a steady increase in step-to-step consistency as children matured.
  • Younger children showed a reduced or largely absent second peak in the vertical ground reaction force profile during single support, compared with the characteristic double-peaked pattern observed in the adolescent group, a difference the authors note may be mechanically related to the altered vertical pivot point position seen in the youngest children.

What This Means for Pediatric Movement Researchers and Developmental Assessment Teams

For researchers studying early gait development, this study offers a normative developmental benchmark that did not previously exist: the mechanism of ground-reaction-force focusing toward a stabilizing pivot point is present from the earliest stages of independent walking, but its position and reliability mature substantially over childhood, with toddlers showing a markedly different, more variable, and sometimes centrally or inferiorly positioned pivot point compared with the consistently superior position seen by adolescence. Because trunk stabilization strategies are frequently altered in children with developmental motor differences, this normative trajectory may offer a useful reference point for interpreting pediatric gait assessments and for future research examining how this specific marker of postural control differs in children whose gait development follows an atypical path.

The Role of Theia3D in Movement Analysis

Theia3D was used as the markerless motion capture processing system throughout this study, providing the kinematic data used to calculate each child's center-of-mass trajectory, which in turn served as the reference point for computing ground-reaction-force focusing and virtual pivot point position. The authors note that markerless motion capture is not yet considered a fully established substitute for marker-based systems, citing prior work indicating that accuracy is strongest in the sagittal plane, which aligns with their decision to restrict this study's kinematic analysis to the sagittal plane specifically. Because this study used Theia3D as an established measurement tool to investigate a developmental biomechanics question rather than to test the software's own accuracy, it offers a practical example of markerless capture enabling a line of pediatric research, virtual pivot point development across early childhood, that had not previously been explored.

Study Limitations and Future Directions

The authors note that the relatively small sample size limits statistical power, though the moderate-to-strong effect sizes observed suggest the age-related differences identified are likely meaningful; they recommend future studies with larger samples and a priori power analysis to investigate additional potential influences on virtual pivot point position, such as kinetics and sex. Because this is, to the authors' knowledge, the first study of virtual pivot point development to use markerless motion capture, they note that their adolescent group's values were comparable to previously published marker-based reference data in healthy adolescents, lending some external support to the findings despite the absence of a direct marker-based comparison within the study itself. The authors also note that their center-of-mass calculations relied on adult-derived anthropometric scaling parameters applied uniformly across all age groups, even though body segment mass distribution is known to differ meaningfully between children and adults; they hypothesize that using age-appropriate mass distributions would shift the calculated pivot point position in the youngest groups, and identify this as an important area for methodological refinement. Finally, because this study's cross-sectional design captures group differences at a single time point, the authors identify longitudinal research tracking individual children over time as a necessary next step to characterize developmental trajectories directly and to establish whether these virtual pivot point parameters hold value as indicators of typical motor development.

Citation: Schreff, L., Nirmaier, K., Blank, C., Abel, R., & Müller, R. (2026). Age-related changes in virtual pivot point position and variability during pediatric gait development. Children, 13(3), 363. https://doi.org/10.3390/children13030363

Link: Read the full study

How Does the Body's Walking "Pivot Point" Develop in Early Childhood?

This study uses Theia3D markerless motion capture to show that the virtual pivot point, a key feature of upper-body stabilization during walking, is present from the earliest stages of independent walking but continues to mature in position and consistency well into childhood. If your work involves pediatric gait development or postural control research and you want to study how stabilization strategies evolve across childhood, talk to our team about applying Theia3D to your own developmental biomechanics research.

Recent Posts