Adaptive feedback control in bimanual cooperative tasks

This study explores the question of adaptation in bimanual cooperative tasks. Previous studies of Crevecoeur et al. (2020a and 2020b) hypothesized the presence of online corrections within the movement in unimanual reaching movement experiments. These online corrections gradually improved even when the external disturbances applied were unexpected. They extracted from muscles recordings the latency of this feedback adaptation, which was approximately 250 ms after the reach onset. This master thesis is part of a larger study aiming to investigate the feedback adaption in the specific case of bimanual coordination. Both limbs may display this within movement feedback adaptation and, the latency could be similar or distinct compared to the unimanual case. Therefore, we conducted a bimanual cooperative reaching movement experiment to search some insights to these questions. A first objective of our work was to reproduce previous findings from Diedrichsen et al. (2007) suggesting that both limbs adapted to a force field perturbation even if one limb remained unperturbed when the goal task was shared between them. Our analysis corroborated their conclusion and strengthened it by the examination of muscles recordings. We also searched for the presence of online corrections during catch trials where the perturbation was unexpected removed. We found a reduced terminal peak end force across catch trials for the perturbed hand that could explained the decreasing path length observed. The muscles recordings did not highlight a potential feedback adaptation latency. We also extended the adaptive control model proposed by Crevecoeur et al. (2020a) to the bimanual case with two different controllers: the linear–quadratic–Gaussian (LQG) and the robust controllers.