Diehl, M. Dyer PhD, PT; Pidcoe, Peter E. PhD, PT
Falls and fall-related injuries are among the costliest health problems in the United States.1 More than 80% of all nonspine fractures and more than 90% of fractures to the hip, distal forearm, and proximal humerus stem from falls among people 65 years or older,2 and fall-related fractures lead to more than $15 billion in health care costs.3 The efficacy of postural control has been shown to decrease markedly in adults past the age of 65 years.4 Declining balance in older adults can lead to decreases in balance confidence and increases in frailty and fear of future falls and loss of independence.4-6
Increased fall frequency among older adults can be related to many factors. Polypharmacy, or the misuse of prescription medication, can predispose some to delayed balance reactions by impairing nerve conduction velocities and motor performance.7 Many older adults demonstrate delayed postural responses to perturbations,8 decreased joint torque production,9 decreased number of type II muscle fibers,10 diminished peripheral sensation,11 and diminished visual acuity.12 Investigations have shown quantitative and morphologic changes in the vestibular system associated with aging13,14 as well as delayed nerve conduction velocities.15 Decreased range of motion and soft-tissue extensibility are also commonly observed among aging adults.16,17 For these reasons, clinicians need to better understand postural control and balance reactions in older adults in an effort to devise appropriate clinical interventions that might decrease falls.
Numerous investigations have quantified kinematic differences between young and old subjects during various balance-recovery and mobility tasks. Wu18 studied head and trunk responses in subjects who stood on a translating platform and found that older adult subjects had a longer latency of trunk and head movements, took longer to reach peak movements of the head and trunk, and responded to perturbations by moving about the ankles with greater trunk-in-space and head-in-space movements. Similar findings were made by Alexander and colleagues,19 who showed variability in segmental excursions among older adult subjects who were perturbed by a moving support surface. Gu and colleagues20 found that older adult subjects exhibited larger horizontal excursions of their whole-body center of mass and took longer to arrest their angular momentum during recovery from a suddenly translated support surface when compared with younger adults. In general, these findings suggest that younger subjects were able to maximize head stability by disassociating the head and trunk segments. Older adult subjects were more likely to rigidly link the head to the trunk and the trunk to the pelvis when responding to destabilizing perturbations. These movement patterns render the head unstable in space and can be a reason why older adults are not able to effectively respond to postural perturbations.
Head-in-space stability is essential for optimum balance recovery.19,21,22 Two of the 3 major systems that contribute to balance, the visual and vestibular systems, are housed in the head, and the ability to minimize extraneous head movements during functional activities and balance corrections are essential to preventing falls. The visual and vestibular systems are linked neuroanatomically and each exerts some influence on the other. Together, these 2 systems assist with head-in-space stability and overall postural equilibrium.23,24
The utilization of visual cues has long been shown to improve performance in balance testing. Nardone and colleagues25 showed that older subjects performed as well as young subjects in stabilizing the head during postural perturbations when testing was performed with eyes open. When both subject groups underwent perturbations with eyes closed, younger subjects demonstrated better head and trunk stability than older adults while each group demonstrated poorer results than with eyes open. Corna and colleagues26 showed that young subjects demonstrated improved head stability during anterior and posterior translations of support surface in eyes open versus eyes closed conditions.
Although it has been well documented that visual information improves balance performance, it is not well understood how visual information is utilized by the central nervous system (CNS) to help with balance and posture. It has been inferred that the presence of a near, earth-fixed (EF) target serves as a reference from which further postural corrections can be made. Simeonov and Hsiao27 reported that anterior-posterior sway, medial-lateral sway, area of sway, and sway velocity were all greater in subjects who stood with eyes open but were without any near, EF visual references. Simoneau et al28 studied postural sway variables in young and older adult subjects while standing in a simulated elevator environment. In both groups, when the elevator doors were closed, thus providing a near visual reference, sway parameters were unremarkable. However, opening of the doors necessitated a gaze shift from a near, EF reference to one that was further away. The older adult subjects demonstrated significant increases in anterior-posterior and medial-lateral sway. In a similar investigation, Dobie and colleagues29 devised a simulated boat-at-sea experiment. Sway parameters in the absence of EF visual references were significantly larger than during test conditions where an EF reference was available.
These authors postulate that a near, EF target provides the CNS with an absolute reference point that is utilized, in conjunction with somatosensory input as well as vestibular input, to assess the body's position and orientation relative to gravity. In the absence of a near, EF point of reference or in the presence of inappropriate or insufficient visual cues, postural stability can be compromised.29
These studies describe the importance of head-in-space stability and the influence of reliable, EF visual references on postural sway parameters during quiet standing. To date, there has been no investigation that has attempted to describe the influence of visual fixation of an EF target on stepping responses following a destabilizing perturbation. Therefore, the purposes of this investigation were to (1) determine whether gaze stabilization of an EF visual reference is possible during a rapid anteriorly directed postural perturbation, (2) check whether successful gaze stabilization would influence stepping responses, and (3) compare step latency (SL) and foveal fixation responses between the 2 groups.
This study was approved by Virginia Commonwealth University's institutional review board. Two subject groups were established on the basis of age. Ten young subjects (mean age [SD] = 27.39 [5.1] years; 4 men and 6 women) and 10 older subjects (mean age [SD] = 71.9 [8.9] years; 4 men and 6 women) volunteered to participate. The young subjects included students at Virginia Commonwealth University and the older adult subjects were recruited following an educational presentation on balance and aging given to residents of area retirement communities. Each subject completed a brief medical history questionnaire and screening, administered by a licensed physical therapist, to identify characteristics that would exclude them from participating. These exclusion criteria are grouped into the following 7 categories: (1) history of visual disorders including but not limited to amblyopia, diplopia, palsy of cranial nerves III, IV, or VI; (2) history of seizure, stroke, brain injury, multiple sclerosis, vestibular dysfunction, or other CNS disease or injury; (3) history of back pain or other orthopedic condition that requires the subject to depend on an assistive device for basic mobility; (4) evidence of standing balance deficits as identified by a score of less than 45 on the Berg balance test; (5) inability to comprehend testing procedures or follow directions as identified by a score of less than 24 on the Folstein Mini-Mental State Examination; (6) use of antianxiety medications or medications for dizziness; and (7) history of a fall within the past year.
After an informed consent was obtained, the subject's dominant eye was identified.30 Briefly, with both eyes open, the subject was asked to look at a light switch located across the room and then to point at the switch. The subject was then asked to maintain gaze upon the switch while alternately opening and closing one eye. The dominant eye was defined as the eye that did not cause a relative visual shift between the switch and the extended finger. Next, each subject was fitted with a camera-based infrared dark pupil tracker (EyeLink II, SR Research, Ltd, Mississauga, Ontario, Canada).31,32 With proper alignment, the eye tracker was capable of measuring eye movements of 30º in either direction horizontally, 18º in either direction vertically, with an accuracy of 0.5º. The sampling rate of the eye tracker was 250 Hz.
An electromagnetic (EM) motion sensor (Motion Monitor, Innovative Sports Training, Chicago, Illinois) was linked to the eye tracker, which allowed recording, in 3-dimensional space, of all head movements. A second sensor, mounted to the posterior trunk at the fourth thoracic vertebrae, provided a reference for head movements. The sampling rate of the kinematic system was 100 Hz with a resolution of 0.5 mm in translation and 0.1º in rotation. The Eyelink computer was slaved to the EM computer via Ethernet link. Using this embedded control, Eyelink data were time-stamped and streamed to the EM computer in real time during data collection. These data were stored for offline processing. During data export (at 250 Hz), a cubic spine function was used to resample the EM data, providing synchronous samples.
With the eye tracker and EM sensors secured, the head and trunk segments as well as the location of the C7/T1 intervertebral space and the T12/L1 intervertebral space were defined in relation to the world origin that was located on the floor to the subject's right. A right-hand coordinate system was utilized with the subject facing the positive y-axis. The positive x- and positive z-axes were at right angles to the y-axis and pointing to the subject's left and down, respectively.
If subjects required corrective lenses during normal activity, then they were tested with their corrective lens on. Eye tracker calibration began with digitizing of the subject's eyes, thereby providing the location of the eyes in world space. Calibration continued by having the subject visually identify 9 targets positioned centrally and at visual angles of 15º in the horizontal and vertical directions. Each target was a blue circle with a cross in the center and subtended a visual angle of 1.39º. The 9 targets were located on a foam whiteboard placed 0.7 m in front of the subject. The targets were presented to the subject in an order determined by the eye-tracking software. A validation process, identical to the calibration process except for the order in which targets were presented, was used to ensure an average eye angle error of less than 0.5º prior to each experiment. Following the calibration/validation process, if the average eye angle error in either the vertical or the horizontal direction exceeded 0.5º, adjustments were made to the eye tracker and calibration/validation was repeated. Subjects were instructed to minimize head movements during calibration to maximize the visual calibration area. This was done by providing the following verbal instructions: “Please keep your head as still as possible. When I tell you which target to look at, I want you to shift your eyes to that target and keep them fixed on the target.”
Following calibration, the subject was asked to stand on 2 force plates, 1 under each foot (Bertec Corp, Columbus, Ohio). These force plates were able to record the vertical forces applied under each of the subject's feet during balance recovery, and these data were used to quantify the onset of stepping. The force plates sampled at a rate of 1000 Hz. To avoid a trip or fall when stepping off the plates, the plates were sunk into the floor so that the top surface of each plate was level with the surrounding surface. A braided cable was attached to the subject via a belt that secured the cable at the midsternal level. The opposite end of the cable was looped through a pulley and suspended from it was a weight equal to 10% of the subject's BW33 (Figure 1). A second cable was looped through a second pulley and connected the weight to a locking mechanism that kept the weight suspended. The subjects were not able to detect the load when the locking mechanism was engaged. By pulling the trigger of the lock, the weight would fall a vertical distance of approximately 20 cm, thereby creating an anteriorly directed postural perturbation through the subject's midthoracic region.28
A load cell (Interface Inc, Scottsdale, Arizona) was connected, in series, between the subject and the weight. The load cell was used to determine the onset of the postural perturbation. Attempts were made to control the slack in the cable connecting the subject and the weight. This was done in an effort to ensure consistency both within and among subjects. Once subjects were fitted with the cable, they were instructed to step backward until the cable was pulled taut. The experimenter then helped the subject move each foot forward 2 in. A tape marker was placed on the floor at the tip of each shoe to indicate future foot placement for that subject.
The subject testing area was draped with black fabric on all sides to minimize extraneous visual cues. The subject was also unable to see the distal end of the cable or the experimenter who triggered the weight to fall. Once the subject's feet were properly placed, a perturbation was delivered. The time interval between final foot placement and perturbation was randomly varied and the subject was given no verbal cueing or warning immediately preceding the perturbation. In an effort to minimize the risk of fall in the older subject population, a second experimenter stood immediately behind the subject to guard against falling.
Each subject was tested under 2 conditions. In the control condition, subjects were given no specific instructions regarding visual cues other than to look straight ahead. With the test area draped in dark fabric on all sides, there was no discrete visual reference to fixate. Subjects were also instructed to “Try to respond naturally. You may take a step forward if you need to.” In the EF condition, a small foam ball subtending a visual angle of 1.4º was placed directly in front of the subject. The location of the EF target ranged from 1.38 to 1.41 m in front of the subject's eyes at rest depending on their initial stance location. Subjects were instructed to maintain gaze on the target at all times. This condition was designed to maximize visual-vestibular interactions.34 For each condition, the subjects were asked to perform mentally alerting activities such as naming state capitals or performing simple mathematical problems in an effort to minimize anticipation of the perturbation.
The 2 different visual conditions were applied to the subjects in a random order determined prior to data collection. Each subject was pulled 3 times in each condition. To minimize the reporting of erroneous eye-in-head data that might occur with slipping of the eye tracker, the eye tracker was recalibrated as described previously halfway through the experiment.
Onset of stepping response
To quantify the step onset, data from the force plates were analyzed. The force plates were able to provide temporal data regarding the vertical forces applied by the subject as the subject responded to the postural perturbation. The stepping foot was identified as the foot that left the ground first.35 The onset of perturbation was known from the load cell, and the temporal difference between onset and initiation of a step was calculated and considered the SL. The hypothesis was that, for each condition, the young adult would step more rapidly than the older adults and that the step response would be faster for each group during the EF condition when compared with the control condition.
Percent foveal fixation
It was of interest to report on the ability of the vestibuloocular reflex to maintain target image fixation on the fovea during head movements associated with balance recovery, irrespective of head translations. The only visual condition in which this was possible was the EF condition as it was the only condition during which there was a discrete visual target whose Cartesian coordinates were known.
The retinal location of the target was quantified in the following manner. The angle of the eye-in-space (EIS) was determined from the following equation:
where HIS is the head-in-space angular position in the sagittal plane. This was measured by the kinematic sensor located on the head. The EIS value represents the gaze angle relationship to world coordinate space. An EIS value of 0º would represent horizontal gaze.
Equation (Uncited)Image Tools
The EIS was then used to determine the vertical coordinate where the gaze vector intersected the target plane. This was done using a trigonometric relationship that included both the EIS data and the Cartesian coordinate location of the center of the eye. The center of the eye was first estimated by digitizing the center of the closed eye. The center of rotation of the eye was assumed to be offset by an additional 1.3 cm from this digitized point.36 The formula for calculating the location of the gaze vector's intersection with the target plane is presented in the following equation:
where 1.606 is the distance, in meters, along the y-axis and between the x-axis and the location of the target; eye_y and eye_z describe the position of the center of eye rotation relative to the world origin.37
Equation (Uncited)Image Tools
Taking the difference between the target plane intersection and the known target location provided the linear error. A difference of 0 represented image stabilization upon the retina. The linear error data were used to determine the location of the retinal projection.37 Here, the linear error was divided by the distance between the target and the eye. The inverse tangent of this value represented the visual angle of the image projection on the retina relative to the fovea.
It was desired to quantify the ability of the vestibuloocular reflex in maintaining foveal fixation during balance-recovery tasks when an EF target was available. To do this, the visual angle for each sample point during each recording was examined. If the visual angle was within ±2.15°, the image was considered to be foveated at that moment. This value was obtained by considering the visual angle of the fovea to be ±0.5º, plus the system error of the eye tracker (±0.25º) plus the visual angle of the target (1.4º). For each recording, the error points that fell within ±2.15° were totaled and the percent foveal fixation (% FF) was calculated.
All data were normally distributed. To assess the strength of the relationship between SL and % FF, Pearson correlation coefficients were computed. Significant differences within each group were examined using 2 ÷ 2 analysis of variance. For instances when significant differences were observed, Tukey's Honestly Significant Difference test was used for further comparison. The level of statistical significance for all tests was α = .05.
Table 1 presents direct comparisons between young and older adults for each test condition. Comparisons between the 2 groups demonstrated that for each visual condition, the older adults took longer to initiate a step than the young adults. Each group demonstrated a more rapid step when provided with an EF visual reference than when no discrete reference point was available. There were no interaction effects (P > .05).
Percent Foveal Fixation
Young adult subjects were able to fixate the target 78.63% (SD = 4.65%) of the time during the EF conditions, compared with older adults, who maintained fixation 61.44% (SD = 8.86%) of the time. T test comparisons between the 2 groups showed a significant difference (P = .005), with younger adults performing better than older adults.
Figure 2 represents the correlation between % FF during perturbations (x-axis) and the resultant stepping response (y-axis). Significant negative linear relationships can be observed between the 2 variables. The young adult subjects (dark circles, r = −0.76 and P = .001) and the older adult subjects (open circles, r = −0.87 and P < .001) demonstrated quicker steps when % FF was high.
This investigation has yielded several significant findings in the areas of balance reactions and visual fixation among the older adult population. Although numerous investigations have previously shown that specific visual input yields improvements in postural control and balance,27-29 this is the first known investigation to suggest the importance of retinal image stability in balance reactions. Further, this is the first investigation that has attempted to quantify retinal image stability during balance-recovery tasks and to compare gaze stabilization between young and older adult subjects.
Table 1 confirms the work of others who have reported increased postural sway between young and older subjects.19,38,39 In this investigation, young subjects consistently stepped more rapidly than older adult subjects. When provided with a discrete EF target, each group's stepping response time improved when compared with trials in which no visual stimulus was presented.
What is unique to this investigation is the ability to quantify image stabilization upon the retina's fovea. Attempts have been made previously using dynamic visual acuity measures while performing functional tasks,40 but no relationships have been established between successful image stabilization and corrective responses following destabilizing perturbations. The results demonstrate the relative success in image stabilization while trying to visually fixate an EF target during postural perturbations. While each group demonstrated successful image stability, the young adult subjects were far more successful in maintaining the image upon the central portion of the retina. A possible reason for the decreased foveal fixation observed in the older adult subjects may be related to vestibular hyposensitivity. With normal aging, the vestibular organs have been shown to decrease in the number of ciliated hair cells, and this degradation may lead to a greater displacement of the center of mass before such displacement can be appreciated.13,14 This could well render older adults less able to rapidly respond to a destabilizing event. Increased somatosensory threshold has been observed in older adults as well, and this phenomenon could diminish stepping reaction times and explain the results reported in Table 1.41 While no efforts were made to quantify vestibular or somatosensory loss in either subject group, subjects were required to meet a minimum score on the Berg balance test, a combined measure of vestibular, visual, and somatosensory input, as well as motor output. It may be possible that the older subjects displayed small degrees of vestibular, somatosensory, visual, and/or motor deficiencies that are not possible to detect using the Berg balance test. Future investigations should consider a more sensitive and objective assessment of somatosensation and vestibular integrity.
As suggested in Figure 2, when subjects were successful in maintaining foveal fixation, observed SLs were improved. Failure to maintain retinal image stability led to delayed responses. Perhaps a more important finding from Figure 2 was the comparison between young adult and older adult responses. Young adults demonstrated a markedly steeper slope than older adults. To assign meaning to these findings, compare the changes in SL when % FF improved from 60% to 80%, for example. At 60% foveal fixation, young subjects stepped with a latency of approximately 0.41 seconds. As foveal fixation improved to 80%, young subjects demonstrated a 20% improvement in stepping response, suggesting an effective utilization of visual input. When older adults improved target foveation from 60% to 80%, their stepping responses improved by only 4%, suggesting a poorer utilization of visual information. While the visual input provided to each group was identical during the EF condition, it appears that the young subjects' stepping responses were more sensitive to improvements in image stability than their older counterparts. In addition, by examining the distribution of data points about the line of best fit, it can be observed that older adults demonstrated significantly less variability. These findings echo those of previous investigations that studied postural sway and balance in older adults.28 The variability of data seen among the young is less impressive and suggests an ability to step rapidly when gaze stabilization is poor. As suggested by Simoneau and colleagues,28 young subjects may be better able to utilize somatosensory input or vestibular cues during postural and balance corrections.
We believe that the findings presented here have significant clinical importance in the area of balance recovery. While numerous previous investigations have documented improved postural sway and balance recovery in the presence of a visual stimulus, this is the first known investigation to specifically suggest the importance of retinal fixation during balance-recovery tasks. Significant differences were observed in stepping responses and gaze fixation ability, and a relationship has been discovered between retinal image fixation and stepping response. The sum total of these findings leads to questions that could be addressed in future investigations. For instance, do older adults have poorer success in gaze stabilization because of postural recovery responses that render the head less stable in space? If so, can head stability be improved in older adults through clinical interventions and will such efforts lead to improvements in stepping responses following a destabilizing perturbation?
The results of this investigation advance our understanding of head and eye kinematics and coordination during balance-recovery tasks. Further demonstrated in these findings are the differences between young adult and older adult subjects during balance recovery. Among these findings is the description of a link between retinal image stability and improved stepping reactions following a perturbation. Finally, this work suggests the relative importance of head-in-space stability during balance-recovery tasks. Together, the findings reported here have potentially great significance for scientists and clinicians who wish to better understand the motor plans and sensory utilization of the older adult who is attempting to overcome a destabilizing perturbation.
The study was funded by PODS II grant from the Foundation for Physical Therapy (M.D.D.).
1. Baloh RW, Ying SH, Jacobson KM. A longitudinal study of gait and balance dysfunction in normal older people. Arch Neurol
2. Nevitt MC, Cummings SR. Type of fall and risk of hip and wrist fractures: the study of osteoporotic fractures. J Am Geriatr Soc
3. Ray NF, Chan JK, Thamer M, Melton LJ III. Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: report from the National Osteoporosis Foundation. J Bone Miner Res
4. Talbot LA, Musiol RJ, Witham EK, Metter EJ. Falls in young, middle-aged and older community dwelling adults: perceived cause, environmental factors and injury. BMC Public Health
5. Lach HW. Incidence and risk factors for developing fear of falling in older adults. Public Health Nurs
6. Schwartz AV, Nevitt MC, Brown BW Jr, Kelsey JL. Increased falling as a risk factor for fracture among older women: the study of osteoporotic fractures. Am J Epidemiol.
7. Turnheim K. Drug therapy in the elderly. Exp Gerontol.
8. Tang PF, Woollacott MH. Inefficient postural responses to unexpected slips during walking in older adults. J Gerontol A Biol Sci Med Sci
9. Thelen DG, Schultz AB, Alexander NB, Ashton-Miller JA. Effects of age on rapid ankle torque development. J Gerontol A Biol Sci Med Sci
10. Lexell J. Human aging, muscle mass, and fiber type composition. J Gerontol A Biol Sci Med Sci.
11. Kokmen E, Bossemeyer RW Jr, Barney J, Williams WJ. Neurological manifestations of aging. J Gerontol
12. Lord SR, Dayhew J. Visual risk factors for falls in older people. J Am Geriatr Soc.
13. Bergstrom B. Morphology of the vestibular nerve. II. The number of myelinated vestibular nerve fibers in man at various ages. Acta Otolaryngol
14. Rosenhall U. Degenerative patterns in the aging human vestibular neuro-epithelia. Acta Otolaryngol
15. LaFratta CW, Zalis A. Age effects on sural nerve conduction velocity. Arch Phys Med Rehabil.
16. Gresham GE, Rathey UK. Osteoarthritis in knees of aged persons. Relationship between roentgenographic and clinical manifestations. JAMA
17. Raab DM, Agre JC, McAdam M, Smith EL. Light resistance and stretching exercise in elderly women: effect upon flexibility. Arch Phys Med Rehabil
18. Wu G. Age-related differences in body segmental movement during perturbed stance in humans. Clin Biomech
19. Alexander NB, Shepard N, Gu MJ, Schultz A. Postural control in young and elderly adults when stance is perturbed: kinematics. J Gerontol Med Sci Bio Sci.
20. Gu MJ, Schultz AB, Shepard NT, Alexander NB. Postural control in young and elderly adults when stance is perturbed: dynamics. J Biomech
21. Wu G. Head movement during sudden base translations as a measure of risks for falls in the elderly. Clin Biomech
22. Di Fabio RP, Emasithi A, Greany JF, Paul S. Suppression of the vertical vestibulo-ocular reflex in older persons at risk of falling. Acta Otolaryngol
23. Bloomberg JJ, Reschke MF, Huebner WP, Peters BT. The effects of target distance on eye and head movement during locomotion. Ann N Y Acad Sci
24. Mulavara AP, Bloomberg JJ. Identifying head-trunk and lower limb contributions to gaze stabilization during locomotion. J Vestib Res
25. Nardone A, Grasso M, Tarantola J, Corna S, Schieppati M. Postural coordination in elderly subjects standing on a periodically moving platform. Arch Phys Med Rehabil
26. Corna S, Tarantola J, Nardone A, Giordano A, Schieppati M. Standing on a continuously moving platform: is body inertia counteracted or exploited? Exp Brain Res
27. Simeonov P, Hsiao H. Height, surface firmness, and visual reference effects on balance control. Inj Prev
28. Simoneau M, Teasdale N, Bourdin C, Bard C, Fleury M, Nougier V. Aging and postural control: postural perturbations caused by changing the visual anchor. J Am Geriatr Soc.
29. Dobie TG, May JG, Flanagan MB. The influence of visual reference on stance and walking on a moving surface. Aviat Space Environ Med
30. Clark MM. Measurement and interpretation of eye-dominance. Nature
31. Green MJ, Waldron JH, Simpson I, Coltheart M. Visual processing of social context during mental state perception in schizophrenia. J Psychiatry Neurosci
32. Jainta S, Hoormann J, Jaschinski W. Accommodation modulates the individual difference between objective and subjective measures of the final convergence step response. Ophthalmic Physiol Opt
33. Luchies CW, Alexander NB, Schultz AB, Ashton-Miller J. Stepping responses of young and old adults to postural disturbances: kinematics. J Am Geriatr Soc
34. Barnes GR. Visual-vestibular interaction in the control of head and eye movement: the role of visual feedback and predictive mechanisms. Prog Neurobiol
35. Schulz BW, Ashton-Miller JA, Alexander NB. Compensatory stepping in response to waist pulls in balance-impaired and unimpaired women. Gait Posture.
36. Medendorp WP, Van Gisbergen JA, Van Pelt S, Gielen CC. Context compensation in the vestibuloocular reflex during active head rotations. J Neurophysiol
37. Demer JL, Crane BT. Ocular compensation due to labyrinthine input during natural motion. Ann N Y Acad Sci
38. Dickin DC, Brown LA, Doan JB. Age-dependent differences in the time course of postural control during sensory perturbations. Aging Clin Exp Res
39. Prado JM, Stoffregen TA, Duarte M. Postural sway during dual tasks in young and elderly adults. Gerontology
40. Grossman GE, Leigh RJ, Bruce EN, Huebner WP, Lanska DJ. Performance of the human vestibuloocular reflex during locomotion. J Neurophysiol
41. Bear-Lehman J, Albert S, Burkhardt A. Cutaneous sensitivity and functional limitation. Top Geriatr Rehabil