Neural correlates of performance and mental workload dynamics during learning of upper-limb body-powered and...

Objective.Despite the technological advancement of upper limb prostheses, it is unclear how device design affects the cognitive-motor processes and mental workload underlying the acquisition of dexterity skills. This study examined changes in performance and mental workload as individuals with intact upper limbs learned to use either a body-powered (BP) or a myoelectric (MYO) bypass prosthesis.Approach.Motor performance (number of blocks transported in one minute), movement smoothness, perceived mental workload (NASA-TLX), and electroencephalography theta, low-alpha, high-alpha, and beta power were examined across a 10-session prosthetic training program to assess the engagement of cognitive-motor and sensorimotor processes.Main results.While both groups revealed enhanced dexterity and movement smoothness and a decrease in perceived mental workload due to training, differences in cortical dynamics were revealed. From early to late practice, BP prosthesis users revealed an elevation of low- and high-alpha and beta power, whereas no such change was observed for MYO prosthesis users. Also, during both practice and retention, MYO prosthesis users revealed attenuated low- and high-alpha and beta power compared to the BP prosthesis users.Significance.These findings suggest that cortical dynamics for BP but not MYO device users became more refined from early to late practice, along with the latter experiencing greater arousal and recruitment of cognitive-motor and sensorimotor resources to produce a similar box and block test performance relative to the former. This work provides novel evidence of neural correlates of prosthesis-specific cognitive-motor processes underlying mental workload, offering a neurophysiological framework for improving the training and assessment of prosthesis users.