The effect of voluntary lateral trunk bending on balance recovery following multi-directional stance perturbations.

Stabilising shifts of the centre of mass (COM) are observed during balance recovery when subjects simultaneously execute voluntary unilateral knee flexion or unilateral arm raising. Here, we examined whether voluntary lateral trunk bending provided more beneficial stabilising effects, and how motor programs of balance corrections are combined with those of the focal voluntary action. The upright balance of 24 healthy young subjects (19-33 years of age) was perturbed using multi-directional rotations of the support-surface. The perturbations consisted of combined pitch and roll rotations (7.5 degrees and 60 degrees/s) presented randomly in six different directions. Three conditions were tested: perturbation of stance only (PO); combined balance perturbation and cued uphill bending of the trunk (CONT); and combined perturbation and cued downhill bending of the trunk (IPS). For comparison, subjects were required to perform trunk bending alone (TO). Outcome measures were biomechanical responses and surface EMG activity of several muscles. Calculated predicted outcomes (PO + TO) were compared with combined measures (CONT or IPS). CONT trunk bending uphill showed two phases of benefit in balance recovery for laterally but, in contrast to voluntary knee bending, not for posterior directed components of the perturbations. IPS trunk bending had negative effects on balance. Early balance correcting muscle responses were marginally greater than PO responses. Prominent secondary balance correcting responses, having a similar timing as voluntary responses observed under TO conditions, were seen under CONT only in trunk muscles. These, and later stabilising, responses had amplitudes as expected from PO + TO conditions being significantly greater than PO responses. The ability with which different muscle synergies for balance corrections and voluntary trunk bending were integrated into one indicates a flexible adjustment of the CNS programs to the demands of both tasks.

Changes in Vocal Fold Morphology During Singing Over Two Octaves.

OBJECTIVE: Vocal folds are widely assumed to only elongate to raise vocal pitch. However, the mechanisms seem to be more complex and involve both elongation and tensioning of the vocal folds in series. The aim of the present study was to show that changes in vocal fold morphology depend on vocal fold elongation and tensioning during singing. STUDY DESIGN: This was a prospective study. METHODS: Forty-nine professional female singers (25 sopranos, 24 altos) were recruited and three-dimensional laryngeal images analyzed in a coronal view derived from high-resolution computed tomography scans obtained at the mean speaking fundamental frequency (ƒ0) and one (2ƒ0) and two octaves (4ƒ0) above ƒ0. RESULTS: The vocal fold angle, defined by a tangent above and below the vocal folds, was 58° at ƒ0, 47° at 2ƒ0, and 59° at 4ƒ0. CONCLUSION: The decreased caudomedial angle of the vocal fold from ƒ0 to 2ƒ0 (change in muscle belly from ";fat" to "thin") and increased angle from 2ƒ0 to 4ƒ0 (from "thin" to "fat") strongly supports the hypothesis that the vocal folds elongate and then tension when singing from ƒ0 to 4ƒ0. This is the first study to show this relationship in vivo.

Postural instability in cerebellar ataxia: correlations of knee, arm and trunk movements to center of mass velocity.

The aim of this study was to investigate the correlations between body segment movements and center of mass (COM) velocity during pathological balance corrections of spinocerebellar ataxia (SCA) patients compared with controls, and to relate correlations indicating instability to EMG activity differences. Eighteen SCA patients and 21 age-matched controls were tested. Upright standing was perturbed using rotations of the support surface. We recorded body motion and surface EMG. For lateral perturbations peaks in COM lateral velocity were larger in SCA patients than controls. These peaks were correlated with increased ("hypermetric") trunk roll downhill and reduced uphill knee flexion velocity. Subsequent arm abduction partially corrected the lateral instability. Early balance correcting responses in knee and paraspinal muscles showed reduced amplitudes compared with normal responses. Later responses were consistent with compensation mechanisms for the lateral instability created by the stiffened knee and pelvis. We conclude that truncal hypermetria coupled with insufficient uphill knee flexion is the primary cause of lateral instability in SCA patients. Holding the knees and pelvis more rigid possibly permits a reduction in the controlled degrees of freedom and concentration on arm abduction to improve lateral instability. For backwards perturbations excessive posterior COM velocity coincided with marked trunk hypermetric flexion forwards. We concluded that this flexion and the ensuing backwards shift of the pelvis result from rigidity which jeopardizes posterior stability. Timing considerations and the lack of confirmatory changes in amplitudes of EMG activity suggest that lateral and posterior instability in SCA is primarily a biomechanical response to pelvis and knee rigidity resulting from increased muscle background activity rather than changed evoked responses.

Control of roll and pitch motion during multi-directional balance perturbations.

Does the central nervous system (CNS) independently control roll and pitch movements of the human body during balance corrections? To help provide an answer to this question, we perturbed the balance of 16 young healthy subjects using multi-directional rotations of the support surface. All rotations had pitch and roll components, for which either the roll (DR) or the pitch (DP) component were delayed by 150 ms or not at all (ND). The outcome measures were the biomechanical responses of the body and surface EMG activity of several muscles. Across all perturbation directions, DR caused equally delayed shifts (150 ms) in peak lateral centre of mass (COM) velocity. Across directions, DP did not cause equally delayed shifts in anterior-posterior COM velocity. After 300 ms however, the vector direction of COM velocity was similar to the ND directions. Trunk, arm and knee joint rotations followed this roll compared to pitch pattern, but were different from ND rotation synergies after 300 ms, suggesting an intersegmental compensation for the delay effects. Balance correcting responses of muscles demonstrated both roll and pitch directed components regardless of axial alignment. We categorised muscles into three groups: pitch oriented, roll oriented and mixed based on their responses to DR and DP. Lower leg muscles were pitch oriented, trunk muscles were roll oriented, and knee and arm muscles were mixed. The results of this study suggest that roll, but not pitch components, of balance correcting movement strategies and muscle synergies are separately programmed by the CNS. Reliance on differentially activated arm and knee muscles to correct roll perturbations reveals a dependence of the pitch response on that of roll, possibly due to biomechanical constraints, and accounts for the failure of DP to be transmitted equally in time across all limbs segments. Thus it appears the CNS preferentially programs the roll response of the body and then adjusts the pitch response accordingly.

Effect of subthalamic nucleus deep brain stimulation on axial motor control and protective arm responses in Parkinson's disease.

Stereotactic surgical interventions for Parkinson's disease (PD) can considerably improve appendicular motor signs, but their effect on axial motor signs--especially balance control under optimal drug therapy--remains unclear. Here, we investigated the effect of bilateral subthalamic nucleus (STN) stimulation on levodopa-resistant axial and appendicular postural impairment in PD. Fourteen patients (11 with young-onset PD) and 18 age-matched controls were included. Patients were tested after intake of a suprathreshold levodopa dose, ensuring optimal response to drug therapy, and with stimulators both turned on and off. Balance control was assessed using multidirectional dynamic posturography. Outcome measures included full body kinematics and surface electromyography of paraspinal and deltoid muscles. Patients with stimulators turned off showed early decreased trunk roll with a loss of directional dependency, followed by increased and abnormally directed--i.e. destabilizing--trunk roll. Pelvis pitch motion showed decreased directional dependency in these patients. The abnormal trunk motion was not corrected by STN stimulation, but directional dependency of both trunk and pelvis motion partially improved, along with a general decrease in muscle activity. Even with stimulators off, protective arm movements were similar in the optimally treated patients and controls, indicating that these appendicular signs respond better to dopaminergic treatment than axial motor control. Our findings indicate that instability in PD results from a reduced flexibility of the trunk and pelvis that is largely resistant to STN stimulation combined with optimal drug treatment. These postural abnormalities are therefore likely associated with non-dopaminergic pathology. In contrast, protective arm movements did appear to be levodopa-responsive. Future studies should focus on identifying subgroups of optimal responders, particularly patients with levodopa-induced dyskinesias.

Identifying deficits in balance control following vestibular or proprioceptive loss using posturographic analysis of stance tasks.

OBJECTIVE: To distinguish between normal and deficient balance control due to vestibular loss (VL) or proprioceptive loss (PL) using pelvis and shoulder sway measures. METHODS: Body-worn gyroscopes measured pelvis and shoulder sway in pitch (anterior-posterior) and roll (side-to-side) directions in 6 VL, 6 PL and 26 control subjects during 4 stance tasks. Sway amplitudes were compared between groups, and were used to select optimal measures that could distinguish between these groups. RESULTS: VL and PL patients had greater sway amplitudes than controls when standing on foam with eyes closed. PL patients also swayed more when standing with eyes closed on firm support and eyes open on foam. Standard sensory analysis techniques only differentiated VL patients from controls. Stepwise discriminate analysis showed that differentiation required pitch measures for VL patients, roll measures for PL patients, and both measures for all three groups. Pelvis measures yielded better discrimination than shoulder measures. CONCLUSIONS: Distinguishing between normal and deficient balance control due to VL or PL required pitch and roll pelvis sway measures. SIGNIFICANCE: Accurate identification of balance deficits due to VL or PL may be useful in clinical practice as a functional diagnostic tool or to monitor balance improvements in VL or PL patients.

Mental body transformation deficits in patients with chronic balance disorders.

BACKGROUND: Movements may be generated consistent with imagining one's own body transformed or "disembodied" to a new position. Based on this concept we hypothesized that patients with objective balance deficits (obj-BD) would have altered neural transformation processes executing own body transformation (OBT) with functional consequences on balance control. Also we examined whether feeling unstable due to dizziness only (DO), without an obj-BD, also lead to an impaired OBT. METHODS: 32 patients with chronic dizziness were tested: 16 patients with obj-BD as determined by balance control during a sequence of stance and gait tasks, 16 patients with dizziness only (DO). Patients and 9 healthy controls (HCs) were asked to replicate roll trunk movements of an instructor in a life size video: first, with spontaneously copied (SPO) or "embodied" egocentric movements (lean when the instructor leans); second, with "disembodied" or "transformed" movements (OBT) with exact replication - lean left when the instructor leans left. Onset latency of trunk roll, rise time to peak roll angle (interval), roll velocity, and amplitude were measured. RESULTS: SPO movements were always mirror-imaged. OBT task latencies were significantly longer and intervals shorter than for SPO tasks (p < 0.03) for all groups. Obj-BD but not DO patients had more errors for the OBT task and, compared to HCs, had longer onset latencies (p < 0.05) and smaller velocities (p < 0.003) and amplitudes (p < 0.001) in both the SPO and OBT tasks. Measures of DO patients were not significantly different from those of HCs. CONCLUSIONS: Mental transformation (OBT) and SPO copying abilities are impaired in subjects with obj-BD and dizziness, but not with dizziness only. We conclude that processing the neuropsychological representation of the human body (body schema) slows when balance control is deficient.

The balance control of bilateral peripheral vestibular loss subjects and its improvement with auditory prosthetic feedback.

OBJECTIVES: We investigated whether long-term bilateral vestibular loss subjects could combine auditory biofeedback of trunk sway with their remaining natural sensory inputs on balance to provide an improved control of trunk sway. A successful integration of natural and artificial signals would provide a basis for a balance prosthesis. METHODS: Trunk sway of 6 bilateral peripheral vestibular loss subjects (BVL) was recorded using either angular position- or velocity-based auditory feedback or no feedback during stance and gait tasks. Roll and pitch trunk movements were recorded with angular velocity transducers mounted just above the waist and feedback without a delay to 4 loudspeakers placed at the left, right, front and rear borders of the 5 m long by 4 m wide test environment. The two types of auditory feedback or no feedback were provided to the subjects in random order. In the feedback modes, sway greater than a preset angle (ca. 0.5 deg) or velocity (ca. 3 deg/s) thresholds caused a tone to be emitted from the speaker towards which the subject moved. The tone volume increased with increasing angle or angular velocity amplitude. RESULTS: For all stance tasks BVL subjects without auditory feedback had a significantly different balance control with respect to that of normal controls. BVL sway values eyes open on a normal surface were reduced with auditory feedback with the greatest reductions in the roll plane. Specifically for the task of standing on 1 leg eyes open with position-auditory- feedback, amplitudes of pitch and roll angles and angular velocities were indistinguishable from those of normal controls. Sway during stance tasks on foam with eyes closed showed no improvement with feedback, remaining greater than normal. For some gait tasks there was a decrease in trunk sway with velocity feedback. CONCLUSION: These initial results indicate that subjects with vestibular loss could incorporate the auditory prosthetic sensory information into their balance commands, particularly in the roll plane if the balance task is performed with eyes open. Position information appears more useful than velocity information in reducing trunk sway during stance tasks. Future work will need to determine the effect of a training time on the improvement in balance control using such a prosthetic device and the ideal position and velocity auditory feedback combination.

Synergies and strategies underlying normal and vestibulary deficient control of balance: implication for neuroprosthetic control.

Future developments of neuroprosthetic control will probably permit locomotion and posture to be maintained without the aid of crutches and will therefore require some form of balance control. Three fundamental questions will arise. First, the question of the location of imbalance-sensing transducers must be assessed. Secondly, the synergy, which is the relative amplitude and timing of muscle activity, and/or the strategy of joint torques required to re-establish a stable posture for different types of balance disturbances must be addressed. Thirdly, the control laws that map either trunk muscle activity or imbalance-sensing transducer outputs into multi-joint postural control of standing by paraplegic individuals must be generated. The most appropriate means of gathering the relevant information applicable to neuroprosthetic control systems is through the detailed analysis of normal and non-normal human models. In order to gain such detailed insights into normal balance control and its dependence on head angular and linear accelerations, the synergy and strategy of balance corrections in normal subjects or patients with vestibular deficits were investigated for two types of support surface perturbation, a dorsiflexion rotation (ROT) and a rearward translation (TRANS). These experimentally induced perturbations to upright stance were adjusted to cause equal amplitudes of ankle dorsiflexion, thus providing additional information about the role of lower leg proprioception on balance control. Synergies defined on the basis of peak cross-correlations of each recorded muscle's EMG to that of the largest muscle response were significantly different for TRANS and ROT. Translation synergies consisted of a sequential coactivation at several levels (soleus and abdominals some 30 msec before hamstrings, and trapezius some 15 msec before paraspinals), whereas the sequential activation of paraspinals and tibialis anterior dominated the balance synergy to ROT. Likewise, response strategies, defined using cross-correlations of joint torques, differed. That for TRANS was organised as a multi-link strategy with neck torques leading those of all other joints by 40 msec or more; hip joint lead ankle torques by 30 msec. That for ROT was organised around hip and ankle torques without a major correlation to neck torques. Vestibulary deficient subjects developed weaker synergies with respect to subjects with normal balance systems under eyes-open conditions and there was no clear synergy with eyes closed. Consequently, hip torques were delayed some 180 msec with respect to ankle torques, and correlations to neck torques were completely out of phase under eyes-closed conditions. Fundamental changes in TRANS synergies and strategies also occurred in vestibulary deficient subjects for eyes-open and eyes-closed conditions.(ABSTRACT TRUNCATED AT 400 WORDS)

The role of stretch and vestibulo-spinal reflexes in the generation of human equilibrating reactions.

Equilibrating reactions in standing humans were examined for evidence that either vestibulo-spinal or proprioceptive long loop stretch reflexes from ankle muscles, or both, are responsible for the control and organization of rapid postural responses. Specifically, the hypothesis was tested that the same postural response could be evoked by rotation of the support surface that mimics the ankle rotation occurring during support surface translations. Rotation perturbations evoked postural responses in leg and trunk muscles that were different in strategy, synergy and coactivation from translation responses, even though the short-latency response in the stretched triceps surae muscles was equal in latency and size. Movement patterns consisted of a stiffening strategy and hardly any compensating ankle rotation for rotation stimuli, and a multi-link strategy with motion focused about the neck, hip and ankle joints for translation stimuli. Dorsiflexion rotations caused earlier and stronger responses in tibialis anterior and quadriceps muscles just post to the onset of paraspinal muscles, whereas rearward translation activated soleus and abdominals strongest, both just prior to hamstring muscles. Correlated activation strengths of agonist and antagonist activity was a common feature for both types of perturbation, albeit, only in the ankle muscles for rotations and only in the trunk muscles for translations. These data suggest that sensory inputs, other than those generated in the lower leg predominate, in the triggering and modulation of equilibrating reactions. Possible candidates are those of the vestibular system or proprioceptive inputs from the trunk.

Trunk sway measures of postural stability during clinical balance tests: effects of age.

BACKGROUND: The major disadvantage of current clinical tests that screen for balance disorders is a reliance on an examiner's subjective assessment of equilibrium control. To overcome this disadvantage we investigated, using quantified measures of trunk sway, age-related differences of normal subjects for commonly used clinical balance tests. METHODS: Three age groups were tested: young (15-25 years; n = 48), middle-aged (45-55 years; n = 50) and elderly (65-75 years; n = 49). Each subject performed a series of fourteen tasks similar to those included in the Tinetti and Clinical Test of Sensory Interaction in Balance protocols. The test battery comprised stance and gait tasks performed under normal, altered visual (eyes closed), and altered proprioceptive (foam support surface) conditions. Quantification of trunk sway was performed using a system that measured trunk angular velocity and position in the roll (lateral) and pitch (fore-aft) planes at the level of the lower back. Ranges of sway amplitude and velocity were examined for age-differences with ANOVA techniques. RESULTS: A comparison between age groups showed several differences. Elderly subjects were distinguished from both middle-aged and young subjects by the range of trunk angular sway and angular velocity because both were greater in roll and pitch planes for stance and stance-related tasks (tandem walking). The most significant age group differences (F = 30, p <.0001) were found for standing on one leg on a normal floor or on a foam support surface with eyes open. Next in significance was walking eight tandem steps on a normal floor (F = 13, p <.0001). For gait tasks, such as walking five steps while rotating or pitching the head or with eyes closed, pitch and roll velocity ranges were influenced by age with middle-aged subjects showing the smallest ranges followed by elderly subjects and then young subjects (F = 12, p <.0001). Walking over a set of low barriers also yielded significant differences between age groups for duration and angular sway. In contrast, task duration was the only variable significantly influenced when walking up and down a set of stairs. An interesting finding for all tasks was the different spread of values for each population. Population distributions were skewed for all ages and broadened with age. CONCLUSIONS: Accurate measurement of trunk angular sway during stance and gait tasks provides a simple way of reliably measuring changes in balance stability with age and could prove useful when screening for balance disorders of those prone to fall.

Estimating net joint torques from kinesiological data using optimal linear system theory.

Net joint torques (NJT) are frequently computed to provide insights into the motor control of dynamic biomechanical systems. An inverse dynamics approach is almost always used, whereby the NJT are computed from 1) kinematic measurements (e.g., position of the segments), 2) kinetic measurements (e.g., ground reaction forces) that are, in effect, constraints defining unmeasured kinematic quantities based on a dynamic segmental model, and 3) numerical differentiation of the measured kinematics to estimate velocities and accelerations that are, in effect, additional constraints. Due to errors in the measurements, the segmental model, and the differentiation process, estimated NJT rarely produce the observed movement in a forward simulation when the dynamics of the segmental system are inherently unstable (e.g., human walking). Forward dynamic simulations are, however, essential to studies of muscle coordination. We have developed an alternative approach, using the linear quadratic follower (LQF) algorithm, which computes the NJT such that a stable simulation of the observed movement is produced and the measurements are replicated as well as possible. The LQF algorithm does not employ constraints depending on explicit differentiation of the kinematic data, but rather employs those depending on specification of a cost function, based on quantitative assumptions about data confidence. We illustrate the usefulness of the LQF approach by using it to estimate NJT exerted by standing humans perturbed by support-surface movements. We show that unless the number of kinematic and force variables recorded is sufficiently high, the confidence that can be placed in the estimates of the NJT, obtained by any method (e.g., LQF, or the inverse dynamics approach), may be unsatisfactorily low.

Differential diagnosis of proprioceptive and vestibular deficits using dynamic support-surface posturography.

The objective of this study was to evaluate how effective dynamic support-surface posturography could be as a diagnostic tool in patients with balance disorders (proprioceptive or vestibular deficits). Specifically, we studied whether measures of trunk control and simple toe-up rotational perturbations, selected using statistical techniques, could provide a better diagnostic yield than either the analysis of lower-body movements or use of a "nulled" ankle input paradigm. The test subjects were 15 control subjects, five patients with bilateral peripheral vestibular loss (VL) and five patients with selective bilateral, lower-leg proprioceptive loss (PL). Amplitudes and onset latencies of bursts of EMG activity in upper and lower-leg muscles, paraspinals and trapezius muscles, concurrent changes in ankle torque, and peak amplitudes of upper-leg, lower-leg, and trunk angular-velocities were measured. Stimuli included three different types of sudden movements of the support surface, a "nulled" ankle input paradigm, a simple toe-up rotation paradigm, and a combined toe-up rotation and backwards translation of the support surface. All stimuli were tested under eyes-open and eyes-closed conditions. For each type of movement and condition the diagnostic classification accuracy (i.e. the overall sensitivity and specificity) was calculated based on those posturography measures providing the highest diagnostic separation between the three populations. Both patient groups showed increased trunk sway, changed support-surface reaction forces and muscle amplitudes compared with controls for toe-up and "nulled" test conditions. Measures providing the greatest diagnostic utility were the amplitude of trunk-angular velocity (increased in VL subjects, less so in PL), the amplitude of balance-correcting paraspinal responses (increased in VL subjects, decreased in PL subjects), the amplitude of trapezius stabilising responses (increased in both patient groups) for simple toe-up rotations under eyes-closed conditions. We conclude, that diagnosis of balance disorders using dynamic posturography is best achieved using measures of trunk control following pure toe-up rotational perturbations tested under eyes-closed conditions.

Trunk sway measures of postural stability during clinical balance tests: effects of a unilateral vestibular deficit.

This research evaluated whether quantified measures of trunk sway during clinical balance tasks are sensitive enough to identify a balance disorder and possibly specific enough to distinguish between different types of balance disorder. We used a light-weight, easy to attach, body-worn apparatus to measure trunk angular velocities in the roll and pitch planes during a number of stance and gait tasks similar to those of the Tinetti and CTSIB protocols. The tasks included standing on one or two legs both eyes-open and closed on a foam or firm support-surface, walking eight tandem steps, walking five steps while horizontally rotating or pitching the head, walking over low barriers, and up and down stairs. Tasks were sought, which when quantified might provide optimal screening for a balance pathology by comparing the test results of 15 patients with a well defined acute balance deficit (sudden unilateral vestibular loss (UVL)) with those of 26 patients with less severe chronic balance problems caused by a cerebellar-pontine-angle-tumour (CPAT) prior to surgery, and with those of 88 age- and sex-matched healthy subjects. The UVL patients demonstrated significantly greater than normal trunk sway for all two-legged stance tasks especially those performed with eyes closed on a foam support surface. Sway was also greater for walking while rotating or pitching the head, and for walking eight tandem steps on a foam support surface. Interestingly, the patients could perform gait tasks such as walking over barriers almost normally, however took longer. CPAT patients had trunk sway values intermediate between those of UVL patients and normals. A combination of trunk sway amplitude measurements (roll angle and pitch velocity) from the stance tasks of standing on two legs eyes closed on a foam support, standing eyes open on a normal support surface, as well as from the gait tasks of walking five steps while rotating, or pitching the head, and walking eight tandem steps on foam permitted a 97% correct recognition of a normal subject and a 93% correct recognition of an acute vestibular loss patient. Just over 50% of CPAT patients could be classified into a group with intermediate balance deficits, the rest were classified as normal. Our results indicate that measuring trunk sway in the form of roll angle and pitch angular velocity during five simple clinical tests of equilibrium, four of which probe both stance and gait control under more difficult sensory conditions, can reliably and quantitatively distinguish patients with a well defined balance deficit from healthy controls. Further, refinement of these trunk sway measuring techniques may be required if functions such as preliminary diagnosis rather than screening are to be attempted.

Differential control of leg and trunk muscle activity by vestibulo-spinal and proprioceptive signals during human balance corrections.

Knowledge about how proprioceptive signals trigger and modulate human balance corrections has important implications for the rehabilitation of postural and gait disorders, and increases our understanding of normal interactions between these sensory systems. We used combinations of support-surface rotation and rearward translation to examine the triggering effects of ankle and knee movements on balance corrections. By comparing the responses in normal subjects to those in persons with a bilateral peripheral vestibular deficit, we determined the modulating influence of vestibular inputs on balance responses. Differences in normal and vestibular-loss responses under the different proprioceptive conditions revealed four general findings. First, ventral leg muscle responses are strongly modulated by vestibulo-spinal inputs and by proprioceptive inputs from the ankle and knee. Second, triceps surae muscle responses are initially dependent on ankle inputs, and after 100 ms are modulated by knee inputs; they are not altered by vestibular loss. Third, paraspinal responses in vestibular-loss subjects are enhanced because of unstable trunk sway induced by the lack of ventral leg-muscle activity. Fourth, the earliest possible triggering signal for establishing the timing of interlink muscle activity appears to be knee flexion and/or trunk rotation on the pelvis. These results indicate that a confluence of knee and trunk proprioceptive and vestibulo-spinal inputs, rather than either input alone, is involved in establishing the muscle synergy underlying normal balance corrections.

Classification of peripheral and central (pontine infarction) vestibular deficits. Selection of a neuro-otological test battery using discriminant analysis.

The results obtained from a complete neuro-otological test battery were examined statistically in order to select measurement variables which would optimally indicate significant differences between four groups: normal patients, patients with partially compensated unilateral peripheral vestibular deficit, patients with an acoustic neurinoma and patients with central (brainstem) vestibular deficit. A stepwise-discriminant analysis was performed on measurements of slow-phase velocity obtained from each test. The primary measurements selected to assign a subject optimally to one population were the canal paresis (CP) of the caloric test, the eye-tracking gain contralateral to the deficit for a 15 deg/s stimulus, the gain asymmetry for optokinetic nystagmus with a 30 deg/s stimulus, and the level of spontaneous nystagmus. The resulting classifications were 100% correct for normal and central deficit patients. However, the division between peripheral deficit and acoustic neurinoma patients overlapped causing about 30% false classifications of neurinoma patients: some 20% of the peripheral deficit patients were classified as normal. If the CP was not available the discriminant analysis substituted the rotating chair response for 5 deg/s2, in place of CP. This substitution caused a 10 to 20% decrease in classification accuracy.

Indicators of the influence a peripheral vestibular deficit has on vestibulo-spinal reflex responses controlling postural stability.

For a controlled sway stabilization task, the areas underlying EMG responses in ankle and neck muscles, as well as amplitudes of ankle torque responses, were shown to be significantly correlated with the clinically defined extent of a patient's peripheral vestibular deficit. The responses, elicited by ankle dorsiflexion of the support surface on which the subject stood, were statistically examined in order to select those measurements which would best indicate differences between a normal, a patient with a unilateral deficit, or one with a bilateral deficit. For this purpose, a stepwise discriminant analysis was performed on measurements of head and trunk angular accelerations in addition to muscle EMG and ankle torque signals. The primary measurements selected to optimally assign a subject to a population were the periods of ankle torque and neck extensor activity associated with correcting for the imposed body displacement backwards and maintaining upright head position respectively. The resulting division into populations was 100% correct. However, within the population of unilateral deficit patients, the technique failed to correctly identify those with acute from those with compensated deficit. This technique of investigating vestibulo-spinal reflex responses is more specific and sensitive than Romberg tests, because it will quantify and specify the underlying cause of the patient's balance and ambulatory disorder.

A postural model of balance-correcting movement strategies.

The patterns of joint torques and movement strategies underlying human balance corrections were examined using a postural model. Two types of support-surface perturbation, dorsiflexion rotation (ROT) and rearward translation (TRANS), were employed. These two perturbations were adjusted to produce similar profiles of ankle dorsiflexion in order to obtain information on the role of lower leg proprioceptive inputs on triggering balance corrections. In addition, the dependence of balance control on head angular and linear accelerations was investigated by comparing the responses of normal and vestibularly deficient subjects under eyes-closed and eyes-open conditions. Differences in ROT and TRANS movement strategies were examined in three ways First, the amplitude and polarity of active joint torques were analysed. These were obtained by altering joint torques applied to a postural model until movements of the model accurately duplicated those of measured responses. Second, the pattern of body-segment angular movements depicted by stick figures moving in response to the computed joint torques was investigated. Third, the peak amplitude and patterns of crosscorrelations between joint torques were measured. Active ankle, knee, and hip joint torques computed for normal subjects rotated the body forward for ROT. In the case of TRANS, computed active torques in normals were of opposite polarity to those of ROT and reversed the forward motion of the body. Subjects with vestibular deficits had lower amplitude torques for ROT and failed to counter the platform rotation. Hip torques for TRANS in vestibular deficient subjects were of opposite polarity to those of normal subjects and resulted in excessive forward trunk rotation. Normally, neck torques acted to stabilize the head in space when trunk angular velocity peaked. Vestibular deficient subjects displayed head movements in response to ROT similar to those generated when neck torques were absent. For TRANS, these same subjects exhibited overcompensatory neck torques. Stick figures of normal responses indicated a stiffening of the body into a leg and a trunk-head link for ROT and a flexible multilink motion for TRANS. Likewise, normal response strategies, defined by using crosscorrelations of joint torques, differed for ROT and TRANS. All joint torque crosscorrelations were significant for TRANS. Neck torques led those of all other joint torques by 40 ms or more, and hip joint led ankle torques by 30 ms. Joint torque correlations for ROT were organised around hip and ankle torques without a major correlation to neck torques. Fundamental changes in all torque crosscorrelations occurred for vestibularly deficient subjects under both eyes-open and eyes-closed conditions.(ABSTRACT TRUNCATED AT 400 WORDS)

Vestibular and proprioceptive modulation of postural synergies in normal subjects.

One way of investigating different muscle synergies underlying human balance control is to assume that only 1 or 2 centrally preprogrammed synergies are available to reestablish upright stance when it is perturbed. According to this hypothesis, the many, apparently different, synergies elicited by rotation or translation of a support-surface on which test subjects stand, in fact, result from a modulation of muscle responses induced by different amplitudes of afferent inputs. To test this hypothesis, we probed the balance control of 16 normal subjects with 5 combinations of rotation and translation of the support surface. Each combination yielded a constant angle (3 or 4 degrees) and angular velocity (18 and 36 degrees/s, respectively) over the first 120 ms of ankle dorsiflexion but resulted in differing velocities of upper leg, trunk, and head movements. These first 120 ms of link movements and the resulting muscle responses were analysed for amplitude and timing modulation using 3 techniques. First, velocities of initial link movements and areas of muscle EMG activity were examined separately for the minimum number of descriptors, which would optimally describe the linear variation of the interlink amplitude synergy with respect to the amount of support-surface rotation or translation employed to perturb balance. Initial trunk angular velocity, which was highly correlated with head linear acceleration (r = 0.9), provided the first best descriptor of initial link movements. Ankle angular velocity provided the second descriptor because it was not correlated with trunk angular velocity. The amplitude modulation synergy of EMG responses could be characterised by the modulation of tibialis anterior and paraspinal muscles between 160 and 240 ms and by that of soleus between 80 and 120 ms after stimulus onset. The linear combination of these best descriptors of link movements and that for EMG response amplitudes changed continuously in an identical manner with changes in the stimulus combination. Second, multivariate linear correlations between the amplitudes of initial link velocities and muscle EMG response areas best describing the response amplitude synergy were examined. Several significant correlations (r > 0.6) were obtained between leg and trunk muscle activity 120 ms after stimulus onset and trunk, or upper leg angular velocity, or head linear velocity, prior to 120 ms. Finally, crosscorrelations between muscle responses were examined for consistent interlink timing synergies between muscle responses.(ABSTRACT TRUNCATED AT 400 WORDS)

Predictors of less stable postural responses to support surface rotations in healthy human elderly.

Balance corrections elicited in response to a rotation of the support-surface were compared between healthy elderly and young normal subjects using surface EMG records from the soleus, tibialis anterior, and neck extensor muscles, and measurements of trunk angular acceleration and ankle torque. Three differences were observed. First, EMG response latencies were significantly longer in the elderly. Second, the normal linear correlation between stabilizing ankle muscle activity and ankle torque was disturbed. These two differences were presumably responsible for the diminished ankle torque exerted on the support surface by the elderly subjects. Third, the magnitude of neck muscle activation was increased in elderly subjects, indicating an increased compensation at the head for trunk angular acceleration. The findings suggest that there are both neural and mechanical changes that may impact on postural corrections in elderly subjects, and that more than one factor needs to be identified when predicting an individual's risk for falling.