A predictive model study of the visual contribution to canine postural control.

Dogs were trained to stand on a movable table and their quiet stance was perturbed by osccillation of the table during normal sighted condition and during blindfolded condition. The data formed the frequency response characteristic (describing function) for postural control with and without visual input. A feedback model was tested to assess the effect of visual input during the perturbation of quiet stance. The results of the tests of the model indicate that the effect of a visual input depends on the context of the multiple sensory factors influencing postural control. The effectiveness of the visual input increases if there is a conflict between the visually derived body position cues and the other cues that indicate the orientation of the body.