Cognitive performance and dynamic spinopelvic kinematics across walking speeds in healthy young adults

Although gait is often considered highly automated, cognitive performance may be associated with dynamic trunk–pelvis kinematics as locomotor demand increases. How these associations vary across standardized walking speeds remains unclear. This study examines speed-dependent associations between cognitive performance and DIERS-derived trunk–pelvis kinematics. Sixty-four healthy young adults completed four cognitive tests and DIERS Formetric 4Dmotion assessment during static standing and treadmill walking at 1, 2, 4, and 5 km/h. Eight trunk–pelvis kinematic variables were examined using correlation, trend, linear mixed-effects, principal component, and canonical correlation analyses. Cognition–kinematic associations were predominantly observed at the faster walking conditions of 4 and 5 km/h, with correlations involving PASAT and spinal-curvature measures reaching r = − 0.700 and r = − 0.709, respectively. Linear mixed-effects models identified significant Speed × SDMT interactions for sagittal imbalance (β = − 0.10, 95% CI − 0.14 to − 0.06; p < 0.001) and pelvic drop (β = − 0.12, 95% CI − 0.16 to − 0.08; p < 0.001), indicating smaller speed-related increases among participants with higher SDMT scores. Pelvic drop increased from 1.59 ± 0.63° during standing to 19.40 ± 7.89° at 5 km/h. The first principal component accounted for 53.4% and 68.7% of kinematic variance at 4 and 5 km/h, respectively. Exploratory multivariate analysis also identified a significant first cognitive–kinematic canonical function (r < sub > c < /sub > = 0.855, Wilks’ Λ = 0.185, F = 3.456, p < 0.001). Cognitive performance was associated with trunk–pelvis kinematic behavior primarily at faster walking speeds. Walking-speed variation may provide a useful framework for examining cognition–kinematic coupling; however, these findings do not establish causal cognitive control or direct mechanical gait instability.