207 Original Contribution The inﬂuence of an immersive virtual environment on the segmental organization of postural stabilizing responses E.A. Keshner a,∗ and R.V. Kenyon b a Sensory Motor Performance Program, Rehabilitation Institute of Chicago, Chicago, IL 60611, USA b Department of Electrical Engineering and Computer Science, The University of Illinois at Chicago, Chicago, IL 60607, USA Received 25 July 2000 Accepted 19 October 2000 We examined the effect of a 3-dimensional stereoscopic scene on segmental stabilization. Eight subjects participated in static sway and locomotion experiments with a visual scene that moved sinusoidally or at constant velocity about the pitch or roll axes. Segmental displacements, Fast Fourier Trans- forms, and Root Mean Square values were calculated. In both pitch and roll, subjects exhibited greater magnitudes of motion in head and trunk than ankle. Smaller amplitudes and frequent phase reversals suggested control of the ankle by segmental proprioceptive inputs and ground reaction forces rather than by the visual-vestibular signals. Postural con- trollers may set limits of motion at each body segment rather than be governed solely by a perception of the visual verti- cal. Two locomotor strategies were also exhibited, implying that some subjects could override the effect of the roll axis optic ﬂow ﬁeld. Our results demonstrate task dependent dif- ferences that argue against using static postural responses to moving visual ﬁelds when assessing more dynamic tasks. Keywords: Sway, stabilization, vision, stereoscopic ∗ Corresponding author: Emily A. Keshner, Sensory Motor Performance Program, Rehabilitation Institute of Chicago, Room 1406 – E809, 345 East Superior Street, Chicago, IL 60611, USA. Tel.: +1 312 238 2228; Fax: +1 312 238 2208; E-mail: eak@northwestern.edu. 1. Introduction Studies of postural control have often deﬁned the inﬂuence of vision by its presence (eyes open) or ab- sence (eyes closed). Generally, visual information was thought to be redundant unless vestibular or somatosen- sory inputs were lost as well [10] because removing vision by closing the eyes during platform disturbances does not signiﬁcantly alter the postural response [21,33, 39]. But studies of dynamic visual inputs during quiet stance indicated that motion of the peripheral visual ﬁeld induced body motion [35]. When a moving visual surround was ﬁxed with respect to the subject’s motion during quiet sway [1,2,33], greater sway amplitudes appeared. When spatial orientation and static postural sway were examined in the presence of a looming visual environment [5,18,29], visual control was dominant in the lower (0.1 Hz) frequency range [12]. Changes in the velocity and frequency of visual ﬁeld motion have been shown to induce postural readjustments [13,29– 31,37], with the most robust postural changes elicited in the roll and pitch planes [14,34]. A multivariate model of sway [25] indicated that both somatosensory and visual information make a unique contribution to the control of whole body posture. Actual postural readjustments relative to visual dis- turbances remain poorly described. In most studies, postural instability has been deﬁned as an increase in static sway obtained through center of pressure mea- sures at the feet [3]. Buchanan and Horak [6] examined segmental organization of postural responses, but relied only upon eyes open and closed as the visual inﬂuence and related segmental organization to the frequency of platform translation. Even so they reported differential controls of the head and trunk and of the lower limbs as did Keshner et al. in an earlier study [23]. Only Kuno et al. [24], when examining the relation between body Journal of Vestibular Research 10 (2000) 207–219 ISSN 0957-4271 / $8.00  2000, IOS Press. All rights reserved