Modification of postural responses and step initiation: Evidence for goal- directed postural interactions

1. In this study, the interaction between anticipatory postural adjustments for step initiation and automatic postural responses to an external perturbation were investigated by having subjects initiate a voluntary forward step while perturbed by a backward surface translation, which caused forward sway of the body. The postural adjustments for step initiation act to move the body center of mass (COM) forward, whereas the automatic postural responses act to move the COM backward to restore stance equilibrium. Because the postural behaviors are in opposition, we asked whether a temporal hierarchy exists in which automatic postural responses are executed to restore equilibrium and followed by stereotypic postural adjustments for step initiation, or whether the interaction between these two postural behaviors is more dynamic. 2. Lower extremity electromyographs (EMGs), ground reaction forces, and kinematics were recorded from 10 subjects during three conditions: to quantify the anticipatory postural adjustments for step initiation, subjects stepped forward as soon as they felt a proprioceptive cue; to quantify the automatic postural responses to perturbation, subjects maintained stance equilibrium in response to a backward surface translation under both feet; and to quantify the interaction between the postural adjustments for the voluntary step and the automatic responses to the perturbation, subjects were exposed to a backward surface translation and instructed to step forward as soon as they felt the platform begin to move. 3. The anticipatory adjustments for step initiation included tibialis activation [stance limb = 163 ± 28 (SE) ms; swing limb = 173 ± 33 ms] and soleus inhibition resulting in center of foot pressure (COP) moving backward and lateral toward the swing limb to propel the COM forward over the stance limb. Subsequently, activation of the swing limb gastrocnemius resulted in heel-off. In contrast, the automatic postural adjustments for maintenance of stance equilibrium during a backward surface translation included activation of soleus and gastrocnemius (104 ± 23 ms and 115 ± 14 ms, respectively) resulting in a symmetrical forward displacement of the COP that moved the COM back to its original position with respect to the feet. 4. When a forward step was initiated in response to the translation, the automatic postural responses were reduced in amplitude bilaterally in soleus and in the stance limb gastrocnemius. When present the postural response occurred at the same latency when the goal was to initiate a step as when the goal was to maintain standing. The amplitude of the swing limb gastrocnemius was less affected, suggesting that reduction of the postural response was specific to the role of the limbs in the impending step. 5. Even though the onset of tibialis for step initiation was unchanged when a step was initiated in response to the translation rather than in response to a proprioceptive cue, the duration of the postural adjustments for step initiation were shortened and the muscle-activation patterns were modified with respect to the passive forward displacement of the body COM. In contrast to the fixed onset of the tibialis, the lateral excursion of the COP began earlier when the step was initiated in response to the translation (150 ± 27 ms) rather than the proprioceptive cue (204 ± 40 ms), suggesting that a more rapid unweighting of the swing limb was required in the perturbed condition because the COM was displaced further forward and faster in the perturbed condition. 6. These results demonstrate that both the automatic postural responses to platform translation and the postural adjustments for step initiation were modified, both by changes in body position and by changes in the intended movement goal. This suggests that the descending central motor program for step initiation and the ascending information associated with the platform translation interacted even though the voluntary reaction time for step initiation was considerably longer than the onset latency of the automatic postural response. These results are not consistent with a temporal hierarchy for postural organization. Rather, there is a dynamic interaction between the automatic postural responses to an external perturbation and anticipatory postural adjustments for goal-directed movements.