642 ORIGINAL ARTICLE Subjective Visual Vertical Perception Relates to Balance in Acute Stroke Isabelle V. Bonan, MD, Emilie Guettard, Marie C. Leman, MD, Florence M. Colle, MD, Alain P. Yelnik, MD ABSTRACT. Bonan IV, Guettard E, Leman MC, Colle FM, Yelnik AP. Subjective visual vertical perception relates to balance in acute stroke. Arch Phys Med Rehabil 2006;87: 642-6. Objective: To determine whether misperception of the subjective visual vertical (SVV) underlies balance difficulties in hemiplegic patients. Design: Descriptive study, using a convenience sample. Setting: Department of physical medicine of a university hospital. Participants: Thirty inpatients with hemiplegia after a hemispheric stroke during the 3 previous months. Interventions: Not applicable. Main Outcome Measures: The SVV was tested while subjects sat in a dark room and were asked to adjust a luminous line to the vertical position. Mean SVV deviation and uncertainty, defined as the standard deviation, were calculated for 8 trials. Balance was assessed by the Postural Assessment Scale for Stroke (PASS) and while patients sat on a laterally rocking platform placed on a Satel force platform. The mean body position and the instability score (Lx), calculated as the length of the course of the center of pressure, were recorded. Functional outcome was also evaluated by the FIM instrument. Results: An abnormal SVV was recorded for 20 of 30 patients. Balance (ie, PASS, Lx) and FIM correlated significantly with SVV tilt (P⬍.001, P⫽.01, and P⬍.001, respectively) and with uncertainty (PASS, P⫽.006; FIM, P⫽.003). Conclusions: Verticality misperception was related to poor balance and might be an important element in the assessment of contributing factors to balance disorders after stroke. It should probably be taken into account when establishing balance rehabilitation programs for patients with hemiplegia. Key Words: Balance; Hemiplegia; Rehabilitation; Stroke; Visual perception. © 2006 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation FTER A STROKE, balance can be affected by various and mixed components, including mechanical components such A as weakness, joint motion limitation, modification of muscle tone, and sensory impairment. Misperception of reference frames such as gravitational cues may constitute another component of imbalance. Contralateral tilt of the subjective visual vertical From the Department of Physical Medicine and Rehabilitation, Groupe Hospitalier Lariboisiere–F. Widal, Paris, France. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Isabelle V. Bonan, MD, Physical Medicine and Rehabilitation Dept, G.H. Lariboisiere–F. Widal, 200 Rue du Faubourg St Denis, 75010 Paris, France, e-mail: isabelle.bonan@lrb.ap-hop-paris.fr. 0003-9993/06/8705-10416$32.00/0 doi:10.1016/j.apmr.2006.01.019 Arch Phys Med Rehabil Vol 87, May 2006 (SVV), tested by asking subjects to adjust a luminous line to the vertical position in a dark room, has been reported after hemispheric lesions. This misperception is frequent after righthemispheric lesion and is closely related to visuospatial neglect.1-3 More than 50% of patients have SVV misperception after recent cerebral hemispheric stroke.1 In addition to the tilt of the SVV, attention has been drawn to the abnormal range of uncertainty in the perception of the SVV, which showed the difficulty that some patients experienced in assessing the vertical.1 The present study deals with the next stage, in which we sought to establish whether SVV misperceptions are related to balance difficulties in patients with hemiplegia. To our knowledge, no studies have been conducted that have determined this relation. Poor ambulation has been related to SVV perturbations, but in that study3 ambulation was scored on a very rough 3-step scale. (Patients were classified as fully mobile, walking up to 50m, or requiring a wheelchair.) The perception of visual verticality after stroke has been investigated in patients with an uncommon specific disturbance of balance called the pusher syndrome.4,5 Discordant results were recorded in these particular patients: no SVV tilt in the study of Karnath et al4 and an unusual tilt, ipsilateral to the lesion, in the study of Saj et al.5 Spatial abilities such as the perception of verticality can be assumed to have a significant role in maintaining balance. We postulated that postural imbalance in patients with hemiplegia may be caused by a misperception of verticality. Therefore, we conducted the present investigation to determine whether SVV misperception correlated with balance difficulties in hemiplegic patients after recent stroke. This might be an important element in assessing factors that contribute to balance disorders after stroke. If a correlation exists, rehabilitation programs of postural control for patients with hemiplegia after stroke should take into account the possible impairment of verticality perception and include new exercises designed to improve the sense of verticality. METHODS Subjects were recruited in our physical and medicine rehabilitation department. All subjects gave informed consent to participate in the study before testing. The study received the approval of the Clinical Investigation Unit of the Lariboisiere University Hospital. Patients had recently experienced their first and only cerebral hemispheric strokes, resulting at least in initial motor and balance impairment. They were all right handed. The time since stroke was never longer than 3 months. Patients were excluded if they had perturbed vigilance, a history of neurologic disturbances, vertigo or vestibular dysfunction, amblyopia or diplopia, or severe aphasia. Only patients who were able to understand the instructions for adjustments of the SVV were included. Patients who could not keep their balance while sitting on the laterally rocking platform were not included. Patients with pusher syndrome, then, were not included. All patients were explored by computed tomography scans or magnetic resonance imaging of the brain and brainstem. An ear, nose, and throat examination was conducted in patients with poor balance. No peripheral vestibular disorder 643 VERTICAL PERCEPTION AND BALANCE AFTER STROKE, Bonan was diagnosed. Before testing, we performed a complete neurologic examination in which the following were evaluated: motricity using the Collin and Wade6 scale (range, 0 –200); visual field, assessed at the bedside; and visuospatial neglect, using the Bells Test7 and bisection test.8 Visuospatial neglect was considered to be present when the difference between the bells omitted on the left and right sides in the Bells Test exceeded 3 and when the bias in line bisection exceeded 0.6cm. Joint position sense was assessed for the big toe, ankle, knee, and hip. Patients were then classified into 3 groups for joint position sense: normal joint position sense (2), hyposensitivity (1), and anesthesia (0). rocking platform with respect to the horizontal. Two parameters were analyzed: xm (in millimeters) and length of the course (ie, Lx) (in millimeters). A control group of 25 healthy subjects (13 men, 12 women; median age, 50y; IQR⫽14y) was tested on the platform. For them, the median absolute deviation xm was 6.5mm (IQR⫽7.8mm) and the median Lx was 40.0mm (IQR⫽13.2mm). Procedure SVV assessment. We conducted SVV testing in a dark room to eliminate any visual reference cues, with subjects sitting in a wheelchair or normal chair. The experimenter oriented an oblique white line on a dark background, placed directly in front of each subject at a distance of 2m, until the subject perceived it as exactly vertical; there was no time limit. The head was maintained in a fixed position by a supporting chinrest. Eight trials were made in random order: for 4 of them, the line was initially tilted 40° to the left of the vertical objective, and patients had to adjust it with a clockwise movement; for the other 4, the rod was tilted 40° to the right of the vertical objective for a counterclockwise adjustment. For each trial, the deviation was evaluated in degrees (⫹, ipsilateral deviation in relation to the lesion; ⫺, contralateral deviation in relation to the lesion). The following parameters were analyzed: (1) the mean SVV deviation (V) value expressed as the mean absolute deviation for the 8 SVV measurements and (2) the uncertainty in the perception of the SVV (U), expressed as the standard deviation (SD) for the 8 SVV measurements. Normative control data were supplied by 20 healthy subjects, who were age-matched with the patients (median age, 60y; interquartile range [IQR], 19y) and underwent the same tests. For the controls, the median V was 0.8° (IQR⫽1.4°) and the maximum individual V was 2.8°. Consequently, the results for each patient were considered different from control values when V exceeded 3°, which is clearly larger than the deviation for the controls and was similar to the normative range previously reported.9 The median U was 1.2° (IQR⫽0.7°) and the maximum interindividual difference was 2.5°. Therefore, for U, 3° was considered significantly abnormal. Postural assessment. For postural assessment we used a specific validated ordinal scale ranging from 0 to 36 called the Postural Assessment Scale for Stroke (PASS)10 and a rocking chair paradigm derived from that proposed by Perennou et al.11-13 Subjects sat in the center of a laterally rocking platform placed over a force platform (Satela). Falls were prevented by the experimenter. The height of the seat was adjusted so that each subject’s legs could hang freely. Subjects were asked to keep their balance while looking straight ahead and fixing on a vertical target positioned on the visual pattern. Subjects themselves were responsible for their imbalance and active correction. The force platform consisted of an upper stable support (480⫻480mm2) fixed on 3 strain-gauge force transducers that recorded the course of each subject’s center of pressure (COP). The platform was connected to a personal computer with a serial connection. Dynamic balance was assessed by placing a laterally moveable rocking platform with a radius of 55cm on the computerized force platform. First, the course of each subject’s COP in the lateral plane was recorded for 25 seconds, and then the mean COP deviation (xm) and the length of the course (Lx) were recorded. The mean COP deviation had a linear relation to the mean orientation of the Statistical Analysis Data were expressed as medians and IQRs, because the parameters were not Gaussian. The relation between each patient’s characteristics (age, motricity), balance (PASS score), mean body orientation (ie, xm), body instability (ie, Lx), FIM score, and the variables V and U describing the SVV were tested by the Kendall correlation test. The visual neglect and nonvisual neglect groups and then the right (RHL) and left hemispheric lesion (LHL) groups were compared for V and U using the Mann-Whitney U test; the 3 sensitivity groups were then compared using the Kruskal-Wallis test. To estimate the role of motricity in the relation between SVV variables and balance, partial coefficients of Kendall were calculated. To estimate the role of joint position sense in the relation between SVV variables and balance, the relation between V and balance and between U and balance were tested among each group of joint position sense with the Kendall correlation test. For all tests, the significance level was fixed at .05 and was analyzed using StatView software.b Functional Outcome The functional independence was evaluated using the FIM instrument (range, 0 –126), used as a global indicator of stability during the activities of daily living.14 RESULTS Patient Characteristics Thirty patients were studied (table 1). Their median age was 59 years (IQR⫽21y), and the median time since stroke was 39.5 days (IQR⫽37d). Thirteen patients had an LHL and 17 had an RHL. Stroke was hemorrhagic in 10 patients and ischemic in 20. Among the ischemic lesions, posterior cerebral artery territory was never involved. Anterior cerebral artery territory was affected in 3 patients and middle cerebral artery territory in 19. The entire territory of middle cerebral artery was involved in 16 patients, deep territory in 4, and superficial territory in 5. A visual field defect was observed in 3 patients. Visuospatial neglect was noted in 15 patients (14 RHL, 1 LHL). Median Motricity Index score was 98 (IQR⫽158). Eight patients exhibited normal joint position sense, 12 had hyposensitivity, and 10 had anesthesia. Table 1: Baseline Characteristics of 30 Patients With Hemiplegia Characteristic Median age (y) Median time since stroke (d) Sex (men/women) Side of lesion (left/right) Type of lesion Ischemic/hemorrhagic Median Motricity Index score Joint position sense (group 0/group 1/group 2) Value 59 (21) 39.5 (37) 16/14 13/17 20/10 98 (158) 10/12/8 NOTE. Values are median (IQR), n, or as otherwise indicated. Arch Phys Med Rehabil Vol 87, May 2006 644 VERTICAL PERCEPTION AND BALANCE AFTER STROKE, Bonan Table 2: SVV Values (V, U) and Scores for Balance (xm, Lx) in Healthy Controls and Patients With LHL or RHL Subjects Controls Mean ⫾ SD Median (IQR) Total patients Mean ⫾ SD Median (IQR) LHL patients Mean ⫾ SD Median (IQR) RHL patients Mean ⫾ SD Median (IQR) V (deg) U (deg) xm (mm) Lx (mm) 1.1⫾1.4 0.8 (1.4) 1.4⫾0.9 1.2 (0.7) 7.8⫾6.1 6.5 (7.8) 40.4⫾9.1 40 (13.2) 3.6⫾3.4* 2.6 (3.6) 4.1⫾4.2* 2.6 (3.7) 16.2⫾9.9* 15.0 (17.0) 64.6⫾37.5* 55.0 (32.5) 3.4⫾4.0 1.6 (3.9) 1.8⫾1.0 1.2 (1.6) 13.5⫾10.8 9.3 (19.2) 55.3⫾20.9 55.0 (33.6) 3.8⫾3.1 2.8 (3.3) 5.9⫾4.9† 3.9 (5.7) 18.6⫾8.8 16.7 (13) 73.2⫾47.3 54.5 (36.0) *Significantly different from controls. Significantly different between LHL and RHL. † SVV Values The median absolute SVV deviation (ie, V) was 2.6° (IQR⫽3.6°), and the median uncertainty (ie, U) was 2.6° (IQR⫽3.7°). Both V and U were higher in the patient group than in the control group (V, P⬍.001; U, P⫽.002) (table 2). The SVV was tilted in 12 patients, and the tilt was always contralateral to the lesion. U was abnormal in 12 patients. Values outside the normative range were recorded for both V and U in 8 patients. Both correlated with the Motricity Index score (P⫽.002 and P⫽.005, respectively), and V also correlated with joint position sense (P⫽.01). U was greater in RHL than LHL (P⫽.002) and was greater in the neglect than in the nonneglect group (P⬍.001). Postural and Functional Values The median score for the PASS was 19 out of 36 (IQR⫽19). In the patient group, median xm was 15mm (IQR⫽17mm) and median Lx was 55mm (IQR⫽32.5mm); both were higher than the control values (xm, P⫽.001; Lx, P⬍.001) (see table 2). The median FIM score was 76 out of 126 (IQR⫽48). PASS score, Lx, and FIM score were related to Motricity Index score (P⬍.001, P⫽.03, and P⬍.001, respectively), and PASS and FIM scores were related to joint position sense (P⫽.003). No relation was found between balance and age, side of lesion, or neglect. Correlation Between the SVV and Balance The scores for PASS, Lx, and FIM were linked to those for V (P⬍.001, ⫽⫺.47; P⫽.01, ⫽.32; P⬍.001, ⫽⫺.49, respectively) and the scores for PASS and FIM were also linked to U (P⫽.006, ⫽⫺.35; P⫽.005, ⫽⫺.35, respectively) (table 3). Partial coefficient scores describing the effect of motricity on the relation between balance and V were as follows: for the PASS, ⫺0.3; for Lx, 0.2; and for the FIM, ⫺0.31. Partial coefficient scores describing the effect of motricity on the relation between balance and U were as follows: for the PASS, ⫺0.1 and for the FIM, ⫺0.1. Among the group of patients with normative joint position sense (group 2), the relations between balance (PASS and FIM) and V were significant (P⫽.03 and P⫽.009, respectively). Among the group of patients with intermediate joint position sense (group 1), the relation between PASS and V was significant (P⫽.03). Among the group of patients with anesthesia (group 0), no relation was found between balance and SVV variables. DISCUSSION Thirty hemiplegic patients who had recently experienced their only stroke were studied. To evaluate balance, we chose to use a clinical tool (PASS), fully validated for recent stroke, and a force platform. On the platform, the sitting position was used—first, to obtain the quickest assessment possible and second, because sitting balance has been reported to correlate with mobility and functional outcomes after stroke.15-17 Another advantage of this position was that it minimized the effects of weakness and spasticity on balance. In the rocking chair paradigm used here, as used by Perennou et al,11-13 2 complementary balance parameters were studied (COP deviation, body instability). All our patients had hemiplegia, with various Motricity Index scores, functional independence, and clinical balance. The time since stroke was homogeneous and short. We found that more than 40% of patients had an SVV tilt and that 40% had abnormal uncertainty in the perception of verticality. The severity of our patients’ impairments probably explains why the SVV values were more perturbed than in some preceding studies,2,3 but they were nevertheless similar to the SVV values recorded in our previous study1 of patients with similarly severe impairments. SVV tilt after stroke has been shown to be a consequence of lesions involving several regions. These include the central vestibular pathways (brainstem, thalamus, cortex) on either side,2,9,18 sensory pathways (thalamus, sensory cortex),19,20 and lesions in regions concerned with visuospatial analysis such as parietal lesions, and, more specifically, right parietal lesions.3 In the right parietal region, the role of the temporoparietal junction is of very great interest—first, it is a polymodal sensory area, because it has been shown to respond to either visual or vestibular inputs2,21 and, second, it is known to be of crucial importance to balance.12,22 The notion of an abnormal range of uncertainty is probably of interest. It has already been observed in hemiplegic patients, and—as here—is very closely related to visuospatial neglect.23 Here, uncertainty was often dissociated from tilt perception. Some patients assessed the verticality with a constant tilt, whereas others assessed it with a great variability, as if they had lost the sense of verticality. Therefore, such uncertainty might be induced by a disturbance different from tilt, probably located at the level of polymodal sensory integration, which is also the most common location of visual neglect. Intuitively accurate perception of visual verticality seems important to maintaining an upright stance and to walking. The perception of visual verticality of patients with hemiplegia has Table 3: The Statistical Correlation Between Variables Describing the SVV (V, U) and Balance (PASS, xm, Lx) and Characteristics SVV PASS xm (mm) Lx (mm) FIM Score Motricity Index Score Age (y) V (deg) U (deg) ⫺.47 (⬍.001) ⫺.35 (.006) .11 (.3) .08 (.5) .32 (.01) .18 (.1) ⫺.49 (⬍.001) ⫺.37 (.003) ⫺.39 (.002) ⫺.35 (.005) ⫺.07 (.5) .05 (.6) NOTE. Values are reported as (P value). Arch Phys Med Rehabil Vol 87, May 2006 VERTICAL PERCEPTION AND BALANCE AFTER STROKE, Bonan been investigated in the case of pusher syndrome.4,5 Karnath et al4 found that this syndrome was related to a perturbed perception of the verticality of one’s own body, known as the subjective postural vertical (SPV), but they recorded no bias in perception of the SVV. Such results contrasted with those of Saj et al,5 who found that patients with pusher syndrome had an ipsilateral SVV tilt. However, these studies concerned patients with a very specific and unusual disturbance of balance, who were not included here because of their incapacity to keep their balance on the unstable platform. Except in these particular patients, to our knowledge this is the first systematic study to find a relation between SVV misperceptions and balance difficulties. The relation between SVV values and balance is highly significant, although correlation coefficients are relatively not impressive because of the particular distribution of the statistics used here. Furthermore, the relation seems in great part independent of motricity and sensory disorders. The partial Kendall coefficients, calculated to estimate the role of motricity in the relation between SVV values and balance, are actually still significant when they are compared with the total Kendall coefficients describing the relation between the SVV misperceptions and balance. Furthermore, the link between SVV misperception and balance difficulties is stronger when the joint position sense is not completely impaired. However, the independence from the joint position sense has to be interpreted cautiously because the number of patients in the 3 sensory groups was small and the confidence intervals for these estimates were therefore large. SVV misperception might be a component of the postural disorders occurring after stroke and should be taken into account when establishing rehabilitation programs. Assessment of the SVV in the upright position needs vestibular and visual inputs. Somatosensory information is required when the head or the entire body is tilted,1,19,20,24 which was not tested in the present study. With regard to vestibular input, perturbation of the vestibular pathway under the thalamus could not have caused the misperception of the SVV, because our series included no patients with peripheral vestibular or brainstem disorders. With regard to the visual input, visual field defects do not seem to disturb SVV perception, but perturbed visuospatial integration, like neglect, is clearly linked to perturbed SVV.1,3 Nevertheless, neglect does not stem from visual input alone but constitutes impairment of higher sensory integration, because it stems from defects not only of visual but also of vestibular and somatosensory information.13,25-27 Therefore, in our patients with hemispheric cerebral lesions, SVV misperception seemed to be caused by higher visuovestibular integration dysfunction, resulting in perturbed construction of the gravitational referential. Impaired perception of the SPV has also been shown to be related to imbalance after stroke. It may constitute another form of higher sensory perturbation, which affects somatosensory integration more directly.4,28,29 Several researchers29,30 have argued in favor of separate pathways for the perception of the SPV and SVV. SPV perception is mainly based on somatosensory input through sense organs in the trunk30 and projects into the posterolateral thalamus,29 whereas SVV perception is mainly based on visual and vestibular inputs31-33 and projects into the vestibular cortex.2 Discordant perceptions of the SPV and SVV have often been reported in both hemiplegic patients and those with vestibular disorders.4,28,33 Both these perceptions of verticality are probably important in maintaining balance and must be taken into account. 645 CONCLUSIONS SVV misperception seems, then, to be directly related to poor balance in acute stroke. This misperception of verticality might reflect impairment of the higher visuovestibular integration process. The results of our study give insight into balance disorders after stroke and may change clinical practice. SVV misperception may be an important element in the assessment of factors that contribute to balance disorders after hemispheric stroke. SVV assessment might be a very simple tool for appreciating central postural dysfunction. A test of verticality perception could be proposed for patients with hemiplegia, with the aim of adapting the rehabilitation program for each person to the possible impairment of verticality perception. Effective interventions for improving the verticality perception must, therefore, be designed. It might be of interest to test such interventions for their effectiveness in restoring balance in future studies. Acknowledgment: We are grateful to Eric Vicaut, MD, of the Clinical Investigation Unit, GH Lariboisiere–F. Widal, France for assistance with statistics. References 1. Yelnik AP, Lebreton FO, Bonan IV, et al. Perception of verticality after recent cerebral hemispheric stroke. Stroke 2002;33:2247-53. 2. Brandt T, Dieterich M, Danek A. Vestibular cortex lesions affect the perception of verticality. Ann Neurol 1994;35:403-12. 3. Kerkhoff G. Multimodal spatial orientation deficits in left-sided visual neglect. Neuropsychologia 1999;37:1387-405. 4. Karnath HO, Ferber S, Dichgans J. The origin of contraversive pushing. Evidence for a second graviceptive system in humans. Neurology 2000;55:1298-304. 5. Saj A, Honoré J, Coello Y, Rousseaux M. The visual vertical in the pusher syndrome. Influence of hemispace and body position. J Neurol 2005;252:885-91. 6. Collin C, Wade DT. Assessing motor impairment after stroke: a pilot reliability study. J Neurol Neurosurg Psychiatry 1990;53: 576-6. 7. Gauthier L, Dehaut F, Joanette Y. The Bells Test for visual neglect: a quantitative and qualitative test for visual neglect. Int J Neuropsychol 1989;11:49-54. 8. Shenkenberg T, Bradford DC, Ajax ET. Line bisection and unilateral neglect in patients with neurologic impairment. Neurology 1980;25:695-9. 9. Dieterich M, Brandt T. Ocular torsion and tilt of subjective visual vertical are sensitive brainstem signs. Ann Neurol 1993;33:292-9. 10. Benaim C, Perennou DA, Villy J, Rousseaux M, Pelissier JY. Validation of a standardized assessment of postural control in stroke patients. Stroke 1999;30:1862-8. 11. Perennou DA, Amblard B, Leblond C, Pelissier J. Biased vertical postural in humans with hemispheric lesions. Neurosci Lett 1998; 252:75-8. 12. Perennou DA, Leblond C, Amblard B, Micallef JP, Rouget E, Pelissier J. The polymodal sensory cortex is crucial for controlling lateral postural stability: evidence from stroke patients. Brain Res 2000;53:359-65. 13. Perennou DA, Leblond C, Amblard B, Micallef JP, Herisson C, Pelissier JY. Transcutaneous electric nerve stimulation reduces neglect-related postural instability after stroke. Arch Phys Med Rehabil 2001;82:440-8. 14. Granger CV, Hamilton BB. The uniform data system for medical rehabilitation report of first admissions for 1990. Am J Phys Med Rehabil 1992;71:33-8. 15. Sandin KJ, Smith BS. The measure of balance in sitting in stroke rehabilitation prognosis. Stroke 1990;21:82-6. Arch Phys Med Rehabil Vol 87, May 2006 646 VERTICAL PERCEPTION AND BALANCE AFTER STROKE, Bonan 16. Loewen SC, Anderson BA. Predictors of stroke outcome using an objective measurement scale. Stroke 1990;21:78-81. 17. Feigin L, Sharon B, Czaczkes B, Rosin AJ. Sitting equilibrium 2 weeks after a stroke can predict the walking ability after six months. Gerontology 1996;42:348-53. 18. Dieterich M, Brandt T. Thalamic infarctions: differential effects on vestibular function in the roll plane. Neurology 1993; 43:1732-40. 19. Anastasopoulos D, Bronstein AM. A case of thalamic syndrome: somatosensory influences on visual orientation. J Neurol Neurosurg Psychiatry 1999;67:390-4. 20. Anastasopoulos D, Bronstein A, Haslwanter T, Fetter M, Dichgans J. The role of somatosensory input for the perception of verticality. Ann N Y Acad Sci 1999;871:379-83. 21. Ro T, Cohen A, Ivry RB, Rafal RD. Response channel activation and the temporoparietal junction. Brain Cogn 1998;37:461-76. 22. Miyai I, Mauricio RL, Reding MJ. Parietal-insular strokes are associated with impaired standing balance as assessed by computerized dynamic posturography. J Neuro Rehabil 1997;11:35-40. 23. Kerkhoff G, Zoelch C. Disorders of visuospatial orientation in the frontal plane in patients with visual neglect following right or parietal lesions. Exp Brain Res 1998;122:108-20. 24. Yardley L. Contribution of somatosensory information to perception of the visual vertical with body tilt and rotating visual field. Percept Psychophys 1990;48:131-4. 25. Rode G, Tiliket C, Charlopain P, Boisson D. Postural asymmetry reduction by vestibular caloric stimulation in left hemiparetic patients. Scand J Rehabil Med 1998;30:9-14. Arch Phys Med Rehabil Vol 87, May 2006 26. Rorsman I, Magnusson M, Johansson B. Reduction of visuospatial neglect with vestibular galvanic stimulation. Scand J Rehabil Med 1999;3:117-24. 27. Vallar G, Rusconi ML, Barozzi S, et al. Improvement of left visuo-spatial neglect by left-sided transcutaneous electrical stimulation. Neuropsychologia 1995;33:73-82. 28. Bisdorff AR, Wolsley CJ, Anastopoulos D, Bronstein AM, Gresty MA. The perception of body verticality (subjective postural vertical) in peripheral and central vestibular disorders. Brain 1996; 119:1523-34. 29. Karnath HO, Ferber S, Dichgans J. The neural representation of postural control in humans. Proc Natl Acad Sci USA 2000;97: 13931-6. 30. Mittelstaedt H. Origin and processing of postural information. Neurosci Biobehav Rev 1998;22:473-8. 31. Böhmer A, Rickenmann J. The subjective visual vertical as a clinical parameter of vestibular function in peripheral vestibular diseases. J Vestib Res 1995;5:35-45. 32. Guerraz M, Poquin D, Ohlmann T. The role of head-centric spatial reference with a static and kinetic visual disturbance. Percept Psychophys 1998;60:287-95. 33. Anastopoulos D, Haslwanter T, Bronstein A, Fetter M, Dichgans J. Dissociation between the perception of body verticality and the visual vertical in acute peripheral disorder in humans. Neurosci Lett 1997;233:151-3. Suppliers a. Satel, 6 rue du Limousin-31700 Blanac, France. b. SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513.