The goal of the present study was to determine how the multiple, interdependent full-body sensorimotor subsystems respond to a change in gaze stabilization task constraints during locomotion. Nine subjects performed two gaze stabilization tasks while walking at 6.4 km/hr on a motorized treadmill: 1) focusing on a central point target; 2) reading numeral characters; both presented at 2 m in front at the level of their eyes. While subjects performed the tasks we measured: temporal parameters of gait, full body sagittal plane segmental kinematics of the head, trunk, thigh, tibia and foot, accelerations along the vertical axis at the head and the tibia, and the vertical forces acting on the support surface. We tested the hypothesis that with the increased demands placed on visual acuity during the number recognition task, subjects would modify full-body segmental kinematics in order to reduce perturbations to the head in order to successfully perform the task. We found that while reading numeral characters as compared to the central point target: 1) compensatory head pitch movement was on average 22% greater despite the fact that the trunk pitch and trunk vertical translation movement control were not significantly changed; 2) coordination patterns between head and trunk as reflected by the peak cross correlation between the head pitch and trunk pitch motion as well as the peak cross correlation between the head pitch and vertical trunk translation motion were not significantly changed; 3) knee joint total movement was on average 11% greater during the period from the heel strike event to the peak knee flexion event in stance phase of the gait cycle; 4) peak acceleration measured at the head was significantly reduced by an average of 13% in four of the six subjects. This was so even when the peak acceleration at the tibia and the transmission of the shock wave at heel strike (measured by the peak acceleration ratio of the head/tibia and the time lag between the tibial and head peak accelerations) remained unchanged. Taken together these results provide further evidence that the full body contributes to gaze stabilization during locomotion, and that its different functional elements can be modified online to contribute to gaze stabilization for different visual task constraints.