(I-336) Promoting Functional Arm Movement Via 3-Dimensional Vibrotactile Kinesthetic Training

Introduction: Decreased sensations of limb position and movement are experienced by ~50% of stroke survivors, leading to a decrease in upper extremity motor control and difficulty performing common daily tasks¹. Although visual feedback of the arm and hand can partly compensate, resulting movements are typically slow and jerky. We propose an alternate compensatory approach: using supplemental vibrotactile feedback to enhance closed-loop control of the arm. Previous work has shown that healthy individuals and some survivors of stroke can use vibrotactile feedback about hand position to enhance the accuracy of 2-dimensional reaching tasks performed without visual feedback^2-5. However, functional interactions with the world commonly require 3-dimensional movements. For this study, we trained healthy participants to execute reach-to-grasp movements with 3-dimensional supplemental vibrotactile kinesthetic feedback. Based on previous results in 2-dimensions, we hypothesized that training with 3-dimensional vibrotactile feedback will lead to post-training improvements in reaching accuracy without concurrent vision compared to proprioception alone. We additionally hypothesized that participants would find the vibrotactile feedback useful, motivating, and satisfying, suggesting a higher probability of continued use.


Materials and Methods: Five participants held a squeeze ball with an integrated active infrared marker while seated in a chair that supported their dominant arm against gravity. A motion tracker (OPTOTRAK 3020) collected real-time reach-to-grasp movements and converted them to cursor motions on a vertical display centered on the participant’s midline. The screen showed a 5x5 grid of targets (±x, ±y) with each location having 5 concentric rings representing the depth (±z) requirement for the reach. Six eccentric rotating mass motors were placed on the stationary non-dominant arm; each motor produced a vibrational signal representing cursor deviations from the center of the workspace along one direction (±x, ±y, ±z). Participants were asked to reach towards the indicated target with their dominant hand holding the squeeze ball under four conditions: (i) with real-time visual feedback of their position to a visual target, (ii) without vision or vibrotactile feedback to a remembered target, (iii) with vibrotactile feedback to a remembered target, and (iv) with vibrotactile feedback to a remembered target with intertrial corrections performed with visual and vibrational feedback. Participants squeezed the ball when they believed they reached the target. Baseline measurements were taken with the first three trial conditions at the start of the first session. Three hours of training with condition four took place over three sessions separated by 12-36 hours. Post-training data with conditions two and three were collected at the end of the third session along with surveys to measure participants’ subjective user experiences.


Results, Conclusions, and Discussions: Preliminary results show that the vibrotactile feedback was effective in promoting increased accuracy of goal-directed reaches. Post-training vibrotactile feedback significantly reduced total Euclidean target capture error compared to post-training proprioception blocks (Figure 1). Additionally, errors in the ±x and ±y directions were significantly decreased between the baseline vibrotactile and post-training vibrotactile blocks. This improvement in accuracy was accompanied by a concomitant increase in reaching time and path length to targets.

Survey results suggest that participants found the device and training helpful, interesting, and enjoyable. Average scores for the System Usability Scale⁶, Intrinsic Motivation Inventory⁷, and the Quebec User Evaluation Satisfaction Assistive Technology⁸ surveys were all within positive subjective experience ranges. Additionally, many participants reported that they decomposed movements along each dimension to isolate a pair of vibration motors. This strategy was claimed to help participants learn and navigate the task space.

These results demonstrate the utility of 3-dimensional supplemental vibrotactile kinesthetic feedback. Importantly, this helps establish a training timeline for healthy participants. By characterizing the improvement of healthy participants over time, we will be able to optimize training protocols for post-stroke intervention.



Acknowledgements: This work was supported by the National Institute of Child Health and Human Development grant 2R15HD093086.


References: ¹Rand D. PLOS ONE. 2018.; ²Krueger A., et al. Journal of NeuroEngineering and Rehabilitation. 2017.; ³Risi N., et al. Journal of Neurophysiology. 2019.; ⁴Ballardini G., et al. Sensors. 2021; ⁵Shah V., et al. Experimental Brain Research. 2022.; ⁶Brooke, J. Usability Evaluation in Industry. 1996; ⁷McAuley, E. et al. Research Quarterly for Exercise and Sport. 1989; ⁸Demers, L., et al. Assistive Technology. 2000.