Dynamics of inter-modality re-weighting during human postural control

Abstract

Flexible and stable postural control requires adaptation to changing environmental conditions, a process which requires re-weighting of multisensory stimuli. Recent studies, as well as predictions from a computational model, have indicated a reciprocal re-weighting relationship between modalities when a sensory stimulus changes amplitude. As one modality is down-weighted, another is up-weighted to compensate (and vice versa). The purpose of this study was to investigate the dynamics of intra- and inter-modality re-weighting process by examining postural responses to manipulation of proprioception and visual modalities simultaneously. Twenty-two young adults were placed in a visual cave and stood on a variable-pitch platform for thirteen trials of 250 s apiece. The platform was rotated at constant frequency of 0.4 Hz and amplitudes of 0.3 (low) or 1.5 (high) degrees. Platform amplitude was manipulated in two conditions: low-to-high or high-to-low. The visual stimulus was displayed at constant frequency of 0.35 Hz and amplitude of 0.08 degrees. The results showed both fast and slow changes in center of mass (CoM) response to the switch in platform amplitude. on both timescales, CoM response changed in a reciprocal manner relative to platform amplitude. When the platform amplitude increased (low-to-high condition), CoM response decreased relative to the platform and increased relative to the visual stimulus, indicating both intra-modality and inter-modality sensory re-weighting. In the high-to-low condition, however, there was no change in CoM response relative to visual stimulus, indicating that re-weighting may also be dependent on the absolute level of gain. Sway variability at frequencies other than the stimulus frequency also showed a reciprocal relationship with CoM gain relative to platform. Overall, these results indicate that dynamics of multisensory re-weighting is clearly more complicated than the schemes proposed by current adaptive models of human postural control.