Relearning Upper Limb Proprioception After Stroke Through Robotic Therapy: A Feasibility Analysis

Abstract

Background: Motor learning can occur through active reaching with the arm hidden from view, leading to improvements in somatosensory acuity and modulation of functional connectivity in sensorimotor and reward networks. In this proof-of-principle study, we assess if the same paradigm benefits stroke survivors using a compact end-effector robot with integrated gaming elements. Methods: Nine community-dwelling chronic hemiplegic stroke survivors with persistent somatosensory deficits participated in 15 training sessions, each lasting 1 h. Every session comprised a robotic-based joint approximation block, followed by 240 repetitions of training using a forward-reaching task with the affected forearm covered from view. During movement, the robot provided haptic guidance along the movement path as enhanced sensory cues. Augmented reward feedback was given following every successful movement as positive reinforcement. Baseline, post-intervention, and 1-month follow-up assessments were conducted, with the latter two sessions occurring after the final training day. Results: Training led to reliable improvements in endpoint accuracy, faster completion times, and smoother movements. Acceptability and feasibility analyses were performed to understand the viability of the intervention. Significant improvement was observed mainly in robotic-based sensory outcomes up to a month post training, suggesting that training effects were predominantly sensory, rather than motor. Conclusions: The study outcomes provide preliminary evidence supporting the feasibility of this intervention for future adoption in neurorehabilitation.

Keywords: neurological rehabilitation; proprioception; reward; robotics; sensory; stroke.

Figure A1

Panel (a) shows typical reaching patterns of a representative stroke participant during the first and last training sessions; (b–d) depict the group average (n = 9) of the endpoint error in each reaching direction, during training, motor, and sensory tests.

Figure 1

(a) Block diagram of the study, comprising a baseline assessment, 15 training sessions, and 2 post-training assessments; (b) in the first warm-up block, the accompanying therapist pointed at an item on the LCD screen, prompting the participant to move the robotic handle toward it; (c) the second warm-up block involved the joint approximation task, where participants moved the handle from a start position to a designated location while encountering a position-dependent resistive force (green arrow), which they were instructed to observe; (d) during training, participants performed outward reaching movements while their affected arm was occluded by an arm cover. The inset illustrates the front view of the hand gripping the robotic handle. The next diagram on the right shows an example of a virtual channel while the participant is reaching 120° towards the left.

Figure 2

Changes in robotic assessment indices of motor and sensory tests, and upper limb clinical scores for the sensory (Em-NSA), motor impairment (FMA-UE), and functional ability (functional ability score—FAS and movement time of the streamlined WMFT), before and after training for each participant. The higher the ordinal scores, the better the performance outcome. As for the MT, a smaller number (in seconds) means better improvement. The * in the top panel denotes p < 0.05.

Figure 3

Deviation between the actual and perceived target locations during training, with Pxx representing individual participants. A linear fit was applied to the data points, where the purple line indicates a reduction in endpoint error (signifying improved movement accuracy), and red line denotes no obvious improvement over time.

Figure 4

The perception of stroke participants about the training sessions they attended using a Likert scale questionnaire. The rating ranges from 3 (the most agreeable) to −3 (least agreeable).