Influence of local sensory afference in the calibration of human balance responses.

Peripheral sensory modulation of balance behavior may require a "calibrated" mechanism which would maintain upright standing by a feedback control of torque at the ankle joint. The calibration of human balance was studied using a systematic presentation of perturbation excursions and velocities in normal freely standing subjects. All perturbations (posterior movements of a force platform) induced a forward body sway and were presented by first increasing and then decreasing the magnitude of perturbation. In preselected conditions the stability of the ankle and hence the accuracy of surface orientation inputs was altered using a foam base placed under the subjects feet. Each subject pressed a hand held response key at the moment a postural disturbance was detected. The automatic neuromuscular response (ANR) was recorded from the gastrocnemius muscles bilaterally and the perturbation detection time (DT) was obtained from the onset of thenar muscle discharge. The major findings in this study were: (1) Conscious DT changed as a function of step variations in perturbation excursion and was disassociated from the ANR latency. The ANR latency remained essentially constant in all conditions and did not have any influence on the kinematics of body sway. (2) Normalized peak body sway decreased during unstable ankle conditions and the reduction of body sway could be attributed to an increase in the gain of the ANR across a 200 ms integration period. The ANR 200 ms amplitude also showed higher correlations with perturbation magnitude during unstable (versus stable) ankle conditions. (3) The 200 ms gastrocnemius amplitude was modulated by excursion and velocity of platform displacement but the amplitude integrated over 100 ms was dependent on only the velocity of perturbation. Our results indicate that balance is controlled by a centrally initiated postural response but regulated in amplitude by local sensory information. These results establish that the gain of the ANR is functional, peripherally driven, and occurs subconsciously to alter the kinematics of body sway.