Robotic Assessment of Thumb Proprioception after Stroke: Reliability, Validity, and Sensitivity
Related Experiment Video
Updated: Oct 31, 2026
Isokinetic Robotic Device to Improve Test-Retest and Inter-Rater Reliability for Stretch Reflex Measurements in Stroke Patients with Spasticity
Published on: June 12, 2019
Luis Garcia-Fernandez1, Andria J Farrens1, Vicky Chan1
1Mechanical and Aerospace Engineering Dept., University of California Irvine, Irvine, California.
Objective:
Despite the thumb's critical role in dexterity and its dense sensory innervation, thumb proprioceptive deficits after stroke remain limited due to the limited nature of clinical assessments. This study evaluated the reliability, validity, and sensitivity of a novel robotic assessment of thumb proprioception, SomatoCircleJump (SCJ), in individuals with chronic stroke.
Design:
Cross-sectional psychometric study evaluating the reliability, validity and sensitivity of SCJ using baseline data from a clinical trial, with assessments conducted twice 5-10 days apart. In SCJ, a robot moved the visually occluded thumb, and participants indicated when the thumb passed a visual target.
Setting:
University-based rehabilitation research laboratory.
Participants:
Thirty-five stroke survivors with residual unilateral hand weakness and 24 healthy age-matched controls.
Interventions:
Not applicable.
Main Outcome Measures:
SCJ proprioceptive error, defined as the angular difference between actual thumb position and the visually presented target. Test-retest reliability was assessed with Intraclass Correlation Coefficient (ICC). Validity and sensitivity were evaluated by comparing SCJ with the Fugl-Meyer Assessment Sensory Scale for the upper extremity (FMA-S) thumb position item, and two robotic proprioception tests, one for the thumb (ThumbSense) and one for the fingers (Crisscross).
Results:
SCJ demonstrated good test-retest reliability (ICC = 0.82, 95% CI [0.72-0.87]). SCJ error correlated significantly with FMA-S (ρ = -0.56, p < 0.001), ThumbSense accuracy (ρ = -0.62, p < 0.001), and Crisscross error (ρ = 0.61, p < 0.001). Using a 2SD criterion relative to controls, SCJ identified thumb proprioceptive impairment in 61% of stroke participants, compared with 23% identified by FMA-S.
Conclusions:
SCJ is a reliable, valid, and sensitive assessment of dynamic thumb proprioception after stroke. It revealed that thumb proprioceptive deficits are common and related to finger proprioception deficits, supporting its use in future studies of sensory impairment and rehabilitation outcomes.
