Sensitivity of Modeled L5–S1 Joint Reaction to Virtual Hand-Wrench Translation During a Squat: A Single-Trial OpenSim Study

Abstract Purpose. To quantify how a prescribed virtual translation of hand-applied external wrenches changes the modeled L5–S1 joint-reaction resultant when the measured kinematics and ground reactions of a squat trial are held fixed. Methods. A single two-handed, 7-kg box-squat trial (participant P2; FATLS APP1 model) was processed using OpenSim 4.4.1. A free moment equal to ΔM = r × F was added while preserving the recorded hand-force vectors, application points, kinematics, and ground reactions. Fourteen accepted model conditions evaluated bilateral global-X (± 50 and ± 100 mm), global-Z (± 50 mm), pelvis-axis (+ 50 mm), unilateral X − 50 mm offsets, and reserve-actuator scaling. Outcomes were descriptive and model-specific. Results. The baseline peak resultant was 5,643.59 N. X − 50 and X + 50 mm altered that peak by + 88.38 and − 88.39 N; X − 100 and X + 100 mm altered it by + 176.79 and − 177.11 N. The five-point global-X response was nearly linear (descriptive R² = 0.999999). Left- and right-hand X − 50 mm changes were + 44.08 and + 44.29 N, with an additivity index of 5.90 × 10⁻⁵. Reserve-scale slope ratios ranged from 0.970 to 1.042. The baseline residual-moment quality ratio exceeded the adopted reference. Conclusion. The analysis demonstrates an internally consistent virtual-wrench sensitivity for one fixed-input model instance. It neither estimates physical load relocation nor provides evidence about disc/contact forces, injury risk, other squat techniques, or population-level effects.