Accurate production of time-varying patterns of the moment of force in multi-finger tasks.

We investigated the production of time profiles of the total moment of force produced in isometric conditions by the four fingers of a hand. We hypothesized that these tasks would be associated with multi-finger synergies stabilizing the time profile of the total moment across trials but not necessarily stabilizing the time profile of the total force produced by the fingers. We also expected the multi-finger synergies to prevent an increase in the moment variability with its magnitude. Seated subjects pressed on force sensors with the four fingers of the right hand and produced two time profiles of the total moment of force, starting from a certain pronation effort, leading to a similar supination effort, and back to the initial pronation effort. One of the profiles was a sequence of straight lines (M-Ramp) while the other was a smooth curve (M-Sine). The subjects showed an increase in the total force during each task. This was accompanied by an increase in the force produced by the fingers opposing the required direction of the total moment-antagonist fingers. Variability of the total force and of the total moment showed complex, non-monotonic changes with the magnitude of the force and moment, respectively. In both tasks, the subjects showed patterns of co-variation of commands to fingers that stabilized the required moment profile over trials. The time profile of the total force was stabilized to a lesser degree or not stabilized at all. The share of fingers with larger moment arms (index finger for pronation efforts and little finger for supination efforts) was higher when the fingers acted to produce moments in a required direction but not necessarily when they acted as antagonists. The results demonstrate the existence of multi-finger synergies stabilizing the combined rotational action. They fit a hypothesis that stabilization of rotational actions may be a default strategy conditioned by everyday experience. The data also suggest that the mechanical advantage hypothesis is valid for sets of effectors that act in the required direction but not for sets of effectors that act as antagonists.