Using a self-modulated treadmill as a novel approach to study cognitive-motor and biomechanical outcomes during dual-task walking in individuals with and without lower limb loss

TL;DR

This study uses a self-modulated treadmill to compare cognitive and biomechanical responses during dual-task walking in people with and without lower limb loss.

Contribution

The novel use of a self-modulated treadmill allows for a more realistic assessment of cognitive-motor interactions during walking.

Findings

Both groups maintained walking mechanics during dual-task walking but showed increased neurocognitive engagement.

Individuals with lower-limb loss showed reduced cognitive task performance despite similar cortical activity.

Uninjured individuals robustly engaged neurocognitive processes during dual-task walking, unlike those with limb loss.

Abstract

Combined examination of mental workload and biomechanics during dual-task walking in individuals with lower-limb loss is limited to fixed, but not self-modulated walking pace, for which the latter enables dynamic cognitive-motor behavior as typically engaged during community ambulation. By assessing electroencephalography (EEG) (theta, low/high-alpha power) and biomechanics (gait speed, double limb support, stride width), the cerebral cortical activity underlying mental workload and walking mechanics were examined when individuals with and without lower-limb loss executed a cognitive task (assessed via response time and accuracy) under variable demand (seated and walking). Both populations maintained walking mechanics (unchanged gait speed, double limb support, stride width) during dual-task walking across demand and exhibited similarly elevated neurocognitive engagement (e.g., attention, action monitoring) indicated by similar theta power increase and low/high-alpha power decrease when facing greater demand. However, injured individuals exhibited relative performance decrement (degraded response time/accuracy), which suggests attenuated cognitive-motor efficiency relative to uninjured (i.e., similar cortical activity across groups with degraded performance). Moreover, while uninjured individuals during dual-task walking could robustly engage neurocognitive processes to maintain walking mechanics and successfully attend to the concurrent cognitive task, those with lower-limb loss did not exhibit such a robust recruitment (i.e., unchanged frontal/temporal high-alpha power). Such alterations in individuals with lower-limb loss leads to maintenance of walking at the cost of a concurrent task. The present work informs rehabilitation practice and reveals specific cognitive-motor outcomes for individuals with lower-limb loss in an enhanced ecological context.

The online version contains supplementary material available at 10.1007/s00221-025-07209-2.

Figures