The use of immersive environments is increasing, especially in rehabilitation and other areas, like sports and gaming. Tasks performed in virtual/augmented reality (VR/AR) often have poorer motor performance, in part due to reduced tactile and proprioceptive inputs. We investigated whether vibrotactile haptic feedback could enhance motor performance within AR. We used the Box and Block Test (BBT) to investigate manual dexterity, where participants moved cubes over a partition using one hand. Participants performed the task in 8 conditions: Real, AR without feedback, and AR with haptic wristband or ring vibrotactile feedback using different vibration patterns (Impulse, Continuous, Hybrid). Movement features were extracted and machine learning was used to classify the experimental conditions. Results revealed that performance was consistently lower in the AR environment compared to the physical BBT. The conditions with the haptic wristband feedback gave similar performance, compared to the AR condition without haptic feedback, in terms of number of cubes moved and successful grasp rate. Conversely, conditions using the ring showed lower performance compared to AR conditions without haptics and the wristband conditions. Motion analysis of the movement trajectory revealed that using vibrotactile devices, particularly the ring, produced slower and shorter movements, as compared to AR alone. A Gradient Boosting machine learning model did not distinguish between all conditions, showing only some accuracy between AR, wristband, and ring conditions. Our study highlights the challenges of integrating haptic sensory enhancements in AR training environments, advocating a nuanced approach to developing haptic feedback systems that complement AR technologies more effectively.
Bonnet et al. (2026) studied this question.