TY - GEN
T1 - Neural Responses to Altered Visual Feedback in Computerized Interfaces Driven by Force- or Motion-Control
AU - Dewil, Sophie
AU - Liu, Mingxiao
AU - Sanford, Sean
AU - Nataraj, Raviraj
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
PY - 2024
Y1 - 2024
N2 - Computerized interfaces, like virtual reality, are increasingly used to improve engagement in movement training tasks like in physical therapy. In this study, we examined how alterations in interface feedback can impact neural responses that affect motor learning. Neurotypical persons participated in simple motor training tasks (e.g., grasping, reaching) while visual performance feedback was systematically altered. We stratified neural response results across primarily force (grasp) and motion (reach) components for more fundamental analysis as complex movements typically require concurrent modulation of force and motion. Feedback alterations included adding noise to or automating the visual feedback in ways previously established to impair the sense of agency and performance. We analyzed the neural responses based on electroencephalography (EEG) recordings in two ways. First, we assessed EEG power changes in the alpha- and beta-band across the brain and in Brodmann area 6, given its role in planning and coordinating complex movements. Second, we did a preliminary analysis with neural networks to suggest how predictable motor errors were from neural response data. We observed significant increases in EEG power with noise-altered visual feedback in the force task, suggesting greater sensitivity of force tasks to training feedback. However, motion and force errors were both highly predictable (< 0.1% max target value) from neural response data, suggesting the potential for artificial intelligence tools to predict errors reliably and alter training feedback from computerized interfaces. In conclusion, computerized feedback may be optimized to leverage neural responses that accelerate movement outcomes.
AB - Computerized interfaces, like virtual reality, are increasingly used to improve engagement in movement training tasks like in physical therapy. In this study, we examined how alterations in interface feedback can impact neural responses that affect motor learning. Neurotypical persons participated in simple motor training tasks (e.g., grasping, reaching) while visual performance feedback was systematically altered. We stratified neural response results across primarily force (grasp) and motion (reach) components for more fundamental analysis as complex movements typically require concurrent modulation of force and motion. Feedback alterations included adding noise to or automating the visual feedback in ways previously established to impair the sense of agency and performance. We analyzed the neural responses based on electroencephalography (EEG) recordings in two ways. First, we assessed EEG power changes in the alpha- and beta-band across the brain and in Brodmann area 6, given its role in planning and coordinating complex movements. Second, we did a preliminary analysis with neural networks to suggest how predictable motor errors were from neural response data. We observed significant increases in EEG power with noise-altered visual feedback in the force task, suggesting greater sensitivity of force tasks to training feedback. However, motion and force errors were both highly predictable (< 0.1% max target value) from neural response data, suggesting the potential for artificial intelligence tools to predict errors reliably and alter training feedback from computerized interfaces. In conclusion, computerized feedback may be optimized to leverage neural responses that accelerate movement outcomes.
KW - Artificial intelligence
KW - Motor activity
KW - Rehabilitation
KW - Virtual reality
KW - Visual feedback
UR - http://www.scopus.com/inward/record.url?scp=85190670497&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85190670497&partnerID=8YFLogxK
U2 - 10.1007/978-981-99-9018-4_22
DO - 10.1007/978-981-99-9018-4_22
M3 - Conference contribution
AN - SCOPUS:85190670497
SN - 9789819990177
T3 - Smart Innovation, Systems and Technologies
SP - 299
EP - 312
BT - AI Technologies and Virtual Reality - Proceedings of 7th International Conference on Artificial Intelligence and Virtual Reality AIVR 2023
A2 - Nakamatsu, Kazumi
A2 - Patnaik, Srikanta
A2 - Kountchev, Roumen
T2 - 7th International Conference on Artificial Intelligence and Virtual Reality, AIVR 2023
Y2 - 21 July 2023 through 23 July 2023
ER -