Unexpected Complexity Of Everyday Manual Behaviors

Abstract

AO_SCPLOWBSTRACTC_SCPLOWHow does the brain control an effector as complex and versatile as the hand? One possibility is that the neural control of the hand is simplified by limiting the space of achievable hand postures. Indeed, hand kinematics can be largely accounted for within a small subspace of postures. This oft replicated finding has been interpreted as evidence that hand postures are confined to this subspace, and that leaving it volitionally is impossible. A prediction from this hypothesis is that measured hand movements that fall outside of this subspace reflect motor or measurement noise. To address this question, we track hand postures of human participants as they perform two distinct tasks - grasping and signing in American Sign Language. We then apply a standard dimensionality reduction technique - principal components analysis - and replicate the finding that hand movements can be largely described within a reduced subspace. However, we show that postural dimensions that fall outside of this subspace are highly structured and task dependent, suggesting that they too are under volitional control. We conclude that hand control occupies a higher dimensional space than previously considered, and propose that controlling the complexity of hand movements is well within the scope of the brains computational power.