Coordination dynamics provides a theoretical framework for understanding how stable patterns of coordinated action emerge, stabilize, and change under varying conditions as a result of the interaction between neuromuscular, environmental, and task constraints. The Haken-Kelso-Bunz (HKB) model predicts that when coordinating actions between two limbs, in-phase (0°) is the most stable coordination mode, anti-phase (180°) is less stable, and 90° is unstable, reflecting the attractor-repeller architecture of the coordination landscape. Although sensory factors such as visual and tactile feedback have been well studied, the role of gravity in shaping these dynamics remains largely unexplored. Here, we tested whether gravity functions as a contextual control parameter in coordination dynamics, reshaping the stability of canonical coordination patterns. During parabolic flight, participants performed bimanual isometric force tasks with Lissajous feedback at 0°, 90°, and 180° relative phase in microgravity (0 g), partial gravity (0.25 g, 0.5 g, 0.75 g), and Earth's gravity (1 g). Behavioral outcomes included indices of bimanual coordination accuracy, bias, and stability, along with unimanual measures of timing and force control. At 1 g, performance followed expected predictions: 0° was most stable, 180° less stable, and 90° least stable. In microgravity, the coordination landscape was destabilized, with increased variability and systematic drifts in constant error during 90° and 180° toward the more stable 0° pattern. Partial gravity (0.25–0.75 g) supported partial recovery of stability for the 90° and 180° task, with higher g-levels generally associated with greater accuracy, reduced variability, and diminished bias. These findings indicate that gravity reshapes the attractor-repeller landscape in a graded and task-dependent manner, supporting its role as a contextual control parameter in coordination dynamics, with implications for motor performance and training in altered-gravity environments.
Weinrich et al. (Wed,) studied this question.