Shared Data for Reaching in Altered Gravity

The shared data, scripts, and tables support the reproduction of statistical analysis and associated tables and figures for the Gritsenko et al., 2026 manuscript.Folder Stats on torques contains Statistical analysis of torque changes across dynamic conditions. To test whether torque scaling with gravity differed between gravity-related and dynamic-related components, cycle-integrated and peak-to-peak torque magnitudes were computed for each participant, degree of freedom, movement direction, amplitude, gravity condition, and method (ID- or EMG-derived). These values were analyzed using linear mixed-effects models implemented in MATLAB (fitlme) as follows: nDataGStat ~ 1 + nDistance + sMethod*nGravityLevel + nGravityLevel*sDOF + (1 + nGravityLevel | idSubject). Random effects were random intercept and random slope for gravity level within subject. In complementary analyses, subject-specific linear slopes relating integrated torque magnitude to gravity level were estimated for each component and compared across components and methods.Folder Stats on EMG predictions contains Statistical analysis of regression coefficients. Regression coefficients were analyzed using linear mixed-effects models implemented in MATLAB (fitlme). For each coefficient type separately (gravity torque, velocity, extension torque derivative, and flexion torque derivative), coefficient values from all participants were reshaped from subject-by-condition matrices into a long-format table. The fixed effects were gravity condition (0g, 1g, 2g), movement direction (down, up, averaged across movement amplitudes), muscle (DeltA, DeltP, Trap, Pec), and all their interactions, whereas subject was included as a random intercept to account for repeated measurements within individuals. Models were fit using the formula Response ~ GCond*Dir*Muscle + (1|Subject), and fixed-effect significance was assessed by ANOVA on the fitted mixed model using Satterthwaite’s approximation for denominator degrees of freedom. For each coefficient, the resulting ANOVA table was used to extract F statistics, numerator and denominator degrees of freedom, and p values for the main effects of gravity condition, movement direction, and muscle, as well as for their two-way and three-way interactions. When effects involving gravity condition were significant, post hoc pairwise comparisons between gravity levels (0g vs 1g, 0g vs 2g, and 1g vs 2g) were performed using reduced mixed-effects models within the relevant subgrouping factor(s), and p values were Bonferroni-adjusted within each comparison set.