In an earlier posturographic investigation (Collins and De Luca 1993) it was proposed that open-loop and closed-loop control mechanisms are involved in the regulation of undisturbed, upright stance. In this study, stabilogram-diffusion analysis was used to examine how visual input affects the operational characteristics of these control mechanisms. Stabilogram-diffusion analysis leads to the extraction of repeatable center-of-pressure (COP) parameters that can be directly related to the resultant steady-state behavior and functional interaction of the neuromuscular mechanisms underlying the maintenance of erect posture. Twenty-five healthy male subjects (aged 19-30 years) were included in the study. An instrumented force platform was used to measure the time-varying displacements of the COP under each subject's feet during quiet standing. The subjects were tested under eyes-open and eyes-closed conditions. The COP trajectories were analyzed as one-dimensional and two-dimensional random walks, according to stabilogram-diffusion analysis. Using this technique, it was found that visual input affects the performance of the postural control system in one of two different ways--either it significantly modifies the steady-state behavior of the open-loop postural control mechanisms, or it significantly alters the characteristics of the other closed-loop feedback mechanisms that are involved in balance control. This result is interpreted as an indication that the visual system is integrated into the postural control system in one of two different ways. The experimental population was roughly evenly divided between these two schemes. For the first group (13 of 25 subjects), visual input principally caused a decrease in the "effective" stochastic activity of the open-loop control mechanisms in both the mediolateral and anteroposterior directions. For the second group (12 of 25 subjects), visual input caused an increase in the effective stochastic activity and uncorrelated behavior of the closed-loop control mechanisms in the anteroposterior direction only. On the basis of these results, it is hypothesized that visual input, in both schemes, serves to decrease the stiffness of the musculoskeletal system. In the former case, this may be accomplished by decreasing the level of muscular activity across the joints of the lower limb, whereas, in the latter case, reduced stiffness may be achieved by reducing the gain(s) of the other postural feedback mechanisms, i.e., the proprioceptive and/or vestibular systems. Using stabilogram-diffusion analysis, it was also found that the two groups of subjects behaved similarly under eyes-closed conditions.(ABSTRACT TRUNCATED AT 400 WORDS)