Enhancing Grip Force Control Through Haptic Feedback Training: Development and Evaluation of a Hand-Grip Device

Abstract

Prior investigations have explored the correlation between hand load force (LF) and grip force (GF) during object manipulation and have established a comprehensive understanding of the characteristics associated with healthy force control. Through these studies, various methodologies have been developed to assess task performance and detect hand dysfunction. Building on these methods and insights from a pilot study discussed in later chapters, the current study aims to develop assessment and training methods to improve hand-grip and load force coordination by incorporating haptic feedback to improve motor function. The haptic feedback system in this study specifically addresses the issue of over-gripping by encouraging a reduction in GF/LF ratio in healthy subjects, with the potential for future application in patients. Since a lower GF/LF ratio, particularly closer to 1, is indicative of good hand coordination, the feedback is designed to guide users toward improving force control during object manipulation. Utilizing a hand-grip assessment device with real-time haptic feedback may be an effective method for improving grip control over time.

An initial pilot study was carried out to gain preliminary insights into grip force reduction. The findings from this earlier phase guided the design of a second user study with haptic feedback, including task protocols. In the follow-up user study, ten healthy individuals engaged in a series of static and dynamic manipulation tasks with a hand-grip device that measures the applied load force and grip force to evaluate the effectiveness of the device and the training in reducing over grip, and thereby their GF/LF ratio. Upon obtaining the individuals’ baseline grip forces for 3 different weight conditions, they were then subjected to training sessions with haptic feedback with the aim of reducing their baseline grip force by 10%. This reduction was calculated based on the original baseline grip forces for each of the three weight conditions. The adjusted values served as maximum grip force thresholds during the training sessions with haptic feedback, guiding participants to reduce their grip force accordingly. The subjects subsequently replicated the initial static and dynamic tasks (without haptic feedback) to compare their performance before and after the training. By reducing their grip force by at least 10% in 28 out of 30 trials, participants also lowered their GF/LF ratio, demonstrating a more efficient application of grip force relative to the load. Furthermore, the study revealed that 9 out of 10 participants experienced a reduction in Root Mean Square Error (RMSE) during the static task post-training, suggesting a positive effect of the training on task performance.