Evaluation of Shared Autonomy and Tactile Feedback on a VR Realistic Physics Simulation of Upper Limb Prosthesis
Abstract
Despite their functional capabilities, myoelectric upper-limb prostheses still present high abandonment rates. Combining user-generated motor commands with algorithmic assistance for grasp control (shared autonomy), or providing tactile feedback about grasping force, may enhance users’ sense of agency over the prosthesis, a key component of prosthesis embodiment (integration of the prosthesis into the bodily self). Understanding these effects in intended users is therefore important for developing prosthetic assistance. We investigated these effects in 58 age- and gender-matched participants: 29 without and 29 with unilateral transradial amputation. Participants performed an object-repositioning task using a physics-based virtual reality simulation of a myoelectric prosthesis under four conditions combining two motor control modes (direct user control or algorithm-assisted shared autonomy) with two vibrotactile feedback modalities (continuous or pulse-width-modulated vibration). Sense of agency was assessed through self-report ratings and a perceived temporal compression task. Additional measures included placement accuracy, gaze behaviour, embodiment, and phantom-limb experience. Shared autonomy increased self-reported agency in controls but decreased it in participants with amputation. Tactile feedback modality showed no significant effect on sense of agency measures. Greater phantom-limb control and embodiment of the phantom limb were associated with higher self-reported agency. Placement accuracy was higher in participants with amputation and under direct control. Direct control outperformed shared autonomy with continuous vibration in controls and with pulse-width-modulated vibration in participants with amputation. In controls, pulse-width-modulated vibration also yielded higher accuracy than continuous vibration under shared autonomy. These population-specific responses may partly reflect previous myoelectric-control experience. Prosthesis development therefore requires consideration of functional performance and users’ perception of the device, alongside longitudinal clinical evaluation to determine how these responses change with continued use.
Related
- PharmaAgent: a real-world evaluation of high- and low-autonomy agentic systems for inpatient vancomycin dosing
- Finite element evaluation of a novel adjustable-length partial ossicular replacement prosthesis
- From sparse data to smart decisions: region-specific policy evaluation via simulation
- Virtual Operating Theatre Simulation: Development and Evaluation of a Web-Based Learning Tool for Medical Students
- Addressing Measurement Error of Machine-Learned Physical Activity in Nonlinear Dose-Response Survival Analysis: Development and Evaluation of Accelerated Failure Time, Spline, and Simulation-Extrapolation Method
BibTeX
@misc{hernandez2026evaluation,
title = {{Evaluation of Shared Autonomy and Tactile Feedback on a VR Realistic Physics Simulation of Upper Limb Prosthesis}},
author = {David Crevillén Hernández and Alex Pasquali and Davide Bargellini and Roberto Meattini and Cosimo Gentile and Camilla Arnaud and Laura Zapparoli and Ionela Bara and Fabio Egle and Claudio Castellini},
howpublished = {PsyArXiv},
year = {2026},
month = oct,
doi = {10.31234/osf.io/vn6a4_v1},
url = {https://osf.io/preprints/psyarxiv/vn6a4_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Evaluation of Shared Autonomy and Tactile Feedback on a VR Realistic Physics Simulation of Upper Limb Prosthesis
AU - David Crevillén Hernández
AU - Alex Pasquali
AU - Davide Bargellini
AU - Roberto Meattini
AU - Cosimo Gentile
AU - Camilla Arnaud
AU - Laura Zapparoli
AU - Ionela Bara
AU - Fabio Egle
AU - Claudio Castellini
PY - 2026
DA - 2026/10/21
PB - PsyArXiv
DO - 10.31234/osf.io/vn6a4_v1
UR - https://osf.io/preprints/psyarxiv/vn6a4_v1/
AB - Despite their functional capabilities, myoelectric upper-limb prostheses still present high abandonment rates. Combining user-generated motor commands with algorithmic assistance for grasp control (shared autonomy), or providing tactile feedback about grasping force, may enhance users’ sense of agency over the prosthesis, a key component of prosthesis embodiment (integration of the prosthesis into the bodily self). Understanding these effects in intended users is therefore important for developing prosthetic assistance. We investigated these effects in 58 age- and gender-matched participants: 29 without and 29 with unilateral transradial amputation. Participants performed an object-repositioning task using a physics-based virtual reality simulation of a myoelectric prosthesis under four conditions combining two motor control modes (direct user control or algorithm-assisted shared autonomy) with two vibrotactile feedback modalities (continuous or pulse-width-modulated vibration). Sense of agency was assessed through self-report ratings and a perceived temporal compression task. Additional measures included placement accuracy, gaze behaviour, embodiment, and phantom-limb experience. Shared autonomy increased self-reported agency in controls but decreased it in participants with amputation. Tactile feedback modality showed no significant effect on sense of agency measures. Greater phantom-limb control and embodiment of the phantom limb were associated with higher self-reported agency. Placement accuracy was higher in participants with amputation and under direct control. Direct control outperformed shared autonomy with continuous vibration in controls and with pulse-width-modulated vibration in participants with amputation. In controls, pulse-width-modulated vibration also yielded higher accuracy than continuous vibration under shared autonomy. These population-specific responses may partly reflect previous myoelectric-control experience. Prosthesis development therefore requires consideration of functional performance and users’ perception of the device, alongside longitudinal clinical evaluation to determine how these responses change with continued use.
ER -