Movement strategies and sensory reweighting in tandem stance: differences between trained tightrope walkers and untrained subjects.

Does skill with a difficult task, such as tightrope walking, lead to improved balance through altered movement strategies or through altered weighting of sensory inputs? We approached this question by comparing tandem stance (TS) data between seven tightrope walkers and 12 untrained control subjects collected under different sensory conditions. All subjects performed four TS tasks with eyes open or closed, on a normal firm or foam surface (EON, ECN, EOF, ECF); tightrope walkers were also tested on a tightrope (EOR). Head, upper trunk and pelvis angular velocities were measured with gyroscopes in pitch and roll. Power spectral densities (PSDs) ratios, and transfer function gains (TFG) between these body segments were calculated. Center of mass (CoM) excursions and its virtual time to contact a virtual base of support boundary (VTVBS) were also estimated. Gain nonlinearities, in the form of decreased trunk to head and trunk to pelvis PSD ratios and TFGs, were present with increasing sensory task difficulty for both groups. PSD ratios and TFGs were less in trained subjects, though, in absolute terms, trained subjects moved their head, trunk, pelvis and CoM faster than controls, and had decreased VTVBS. Head roll amplitudes were unchanged with task or training, except above 3Hz. CoM amplitude deviations were not less for trained subjects. For the trained subjects, EOR measures were similar to those of ECF. Training standing on a tightrope induces a velocity modification of the same TS movement strategy used by untrained controls. More time is spent exploring the limits of the base of support with an increased use of fast trunk movements to control balance. Our evidence indicates an increased reliance on neck and pelvis proprioceptive inputs. The similarity of TS on foam to that on the tightrope suggests that the foam tasks are useful for effective training of tightrope walking.