Assessment of the visual noise influence on muscle activation during a tracking task.

It is known how humans exploit their muscle contraction to increase the accuracy of their movements, in order to compensate for haptic disturbance or a combination of haptic and visual noise, but it is unclear the individual contribution of visual noise to muscle activity. Here, we aim to investigate the activation of the arm muscle during a 3-DoFs tracking task in presence of visual noise without any haptic feedback using a robotic interface. We evaluated four different levels of visual noise together with a no-noise condition. Tracking performance was assessed in terms of position and orientation error. An index of global muscle activation was obtained through data recorded with six EMG sensors placed on the participant’s arm used in the tracking task. We observed that muscle activation decreased with the increment in visual noise for the first three levels. On the other hand, the muscle activation for the fourth noise level increased with respect to the third one, reaching values similar to the no-noise condition. As in previous studies, we obtained a linear relationship between tracking errors and visual noise with bigger errors corresponding to higher standard deviation values of the noise, while muscle co-contraction showed a more complex behaviour.