Evaluating self-assistance during functional reach with a passive hydrostatic exoskeleton under artificial impairment.

Abstract

Background: Practicing functional upper-extremity tasks with manual self-assistance may promote motor recovery and restore voluntary control to an impaired limb, reducing reliance on external aid. However, most evidence comes from studies involving tasks with limited coordinative demands. In a functional task like reaching for and lifting an object, learning to generate coordinated assistive forces with an external device may pose bilateral sensorimotor challenges that limit motor learning in the impaired limb. To address this question, we developed a passive hydrostatic exoskeleton (hEXO) that enables self-assistance and paired it with an artificial impairment paradigm using Dysfunctional Electrical Stimulation (DFES), which induces involuntary hand closure during reaching.

Methods: Twenty neurologically typical adults (26 ± 3 yrs) performed a reach-to-grasp and object lift task under challenging sensorimotor conditions: as fast as possible with their non-dominant hand while experiencing an artificial impairment induced by DFES. The stimulation functionally mimicked deficits related to a flexion synergy after neurological injury by making it difficult for participants to extend their fingers while reaching for an object. Experiment 1 assessed the short-term effects of DFES and wearing the hEXO. In Experiment 2, participants were randomly assigned to either a group that could self-assist with the hEXO (n = 10) or a control group that could not self-assist (n = 10) to investigate adaptation to self-assistance and transfer of motor performance to unassisted conditions.

Results: DFES created a sensorimotor challenge and increased reach-to-grasp time by about 50% during early exposure. The self-assist group improved their reach-to-grasp times faster than controls (p = 0.008), achieved comparable reaching times (p = 0.060), and had a slightly higher incidence of unsuccessful attempts (about one in 20 attempts; p < 0.001). Reach-to-grasp performance did not decline following the removal of self-assistance, indicating no performance dependency. Both groups had similar movement times and success rates in the final unassisted practice block.

Conclusions: In this sample of adults with an artificial impairment, self-assistance using a passive hydrostatic exoskeleton accelerated motor performance improvements without creating a dependency on the assistance. If replicated in clinical populations, this approach may help promote upper-limb functional independence.