The dynamics of rhythmical aiming in 2D task space: relation between geometry and kinematics under examination.

We explored a two-dimensional task space variant of the classical rhythmical Fitts' task in which participants were asked to sequentially cross four targets arranged around the extreme points of the major axes of an ellipse. Fitts' law was found to adequately describe the changes in movement time with the variations in task difficulty (ID), but the 1/3 power-law relating curvature and tangential velocity of the trajectory did not resist the increase in ID. Kinematic analyses showed that the behavioral adaptation to the ID resulted in an increase in the contribution of non-linear terms to the kinematics along the two axes of task space. Moreover, a limit cycle model (combining Rayleigh damping and Duffing stiffness, as in one-dimensional Fitts' task) captured such a behavior. In such a context, Fitts' law and the 1/3 power law appear as surface relations that emerge from parametric changes in a dynamical structure that captures the nature of Fitts' task.