Search

Lesion of the "vestibular cortex" in the human posterior insula leads to a tilted perception of visual vertical but not to tilted body posture and loss of lateral balance. However, some stroke patients show the reverse pattern. Although their processing of visual and vestibular inputs for orientation perception of the visual world is undisturbed, they push away actively from the ipsilesional side (the side of lesion location), leading to a contraversive tilt of the body (tilt toward the side opposite to the lesion) and falling to that side. Recently, the origin of contraversive pushing was identified as an altered perception of the body's orientation in relation to gravity. These patients experience their body as oriented "upright" when actually tilted enormously to the ipsilesional side (18 degrees on average). The findings argued for a separate pathway in humans for sensing body orientation in relation to gravity apart from the one projecting to the vestibular cortex. The present study aimed at identifying this brain area. The infarcted brain regions of 23 consecutively admitted patients with severe contraversive pushing were projected onto a template MRI scan, which had been normalized to Talairach space. The overlapping area of these infarctions centered on the posterolateral thalamus. Our finding necessitates reinterpretation of this area as being only a "relay structure" of the vestibular pathway on its way from the brainstem to the vestibular cortex. The ventral posterior and lateral posterior nuclei of the posterolateral thalamus (and probably its cortical projections) rather seem to be fundamentally involved in the neural representation of a second graviceptive system in humans decisive for our control of upright body posture.

The semicircular canals selectively transduce angular velocity and are normally insensitive to gravity and linear acceleration. In acute alcohol intoxication, however, the cupula becomes lighter than the endolymph, rendering it sensitive to gravity (buoyancy hypothesis). This results in positional alcohol nystagmus (PAN) and rotatory vertigo. We evaluated PAN in 8 normal subjects by means of three-dimensional eye-movement analysis in an attempt to clarify if the buoyancy mechanism is sufficient to explain PAN. Forty minutes after intake of 0.8 g of alcohol/kg of body weight, the subjects were positioned such that the lateral canals were earth vertical. They were then rotated in the plane of the lateral canals about an earth-horizontal axis to either 45 degrees or 90 degrees , right or left ear down, and eye movements were recorded for 40 seconds in each position. The spatial analysis of the responses showed that in addition to the nystagmus induced by the buoyancy of all six cupulae, alcohol intoxication also causes a vertical velocity offset (in all subjects, slow phase down) that is independent of the orientation of the subject in space. The offset may represent a toxic effect on central vestibular pathways, producing a tone imbalance of the vertical vestibulo-ocular reflex.

Estimates of the subjective visual and postural vertical were obtained from five patients with acute peripheral vestibular lesions and 20 normal subjects. The visual vertical was assessed by asking the subjects to align a target line to earth vertical by means of remote control. Postural vertical judgments were obtained by exposing them to rotational displacements in the roll plane while sitting on a motor-driven chair and requiring them to align their body to vertical using a joystick control. While the patients showed strong deviations of the visual vertical towards the lesion side, their postural vertical judgments remained veridical. We conclude that the above perceptions are not processed identically and that the participating sensory systems are differently weighted during these tasks.

This paper reviews the features and topodiagnostical meaning of pathological body posture in cerebellar disorders and presents knowledge as to the physiology and pathophysiology of postural disturbances attributed to the cerebellum. Lesions of the spinocerebellar (upper vermal and intermediate) part of the anterior lobe lead to spontaneous anterior-posterior body sway, with a frequency of about 3 Hz with mild or absent ataxia of the upper limbs, but prominent ataxia of the lower limbs. The tremor is provoked by eye closure and by body displacement. Increased gain and poor cerebellar control of the duration and gain of stabilizing reflexes may be the mechanisms of this oscillation. Lesions of the lower (vestibulocerebellar) vermis cause postural ataxia of the head and trunk while sitting, standing, and walking. Postural ataxia is omnidirectional, sometimes of excessive amplitude and contains frequency components below 1 Hz. Dysmetria of the upper and lower limbs is not prominent. Visual stabilization is less than in the other groups of cerebellar patients. Lesions of the cerebellar hemispheres alone, do not cause an increase of postural sway. Lesions of spinocerebellar afferents (Friedreich's disease) lead to a low frequency large amplitude lateral sway with the most of its power below 1.1 Hz. Visual stabilization is preserved. Latencies of early and late EMG responses to sudden displacements are normal in patients with cerebellar disorders. The cerebellum is, therefore, unlikely to be the primary generator of these reflexive responses. On the other hand, the temporal composition of the orderly time sequence of a complex motor program--be it a voluntary act or its postural balancing--is deranged in cerebellar lesions. It is hypothesized that the cerebellum helps to coordinate the timing not only within but also between the single components involved in each subunit of a complex movement in three-dimensional space and that is scales the size and duration of each muscular action. The cerebellum possibly specifies the cortical movement command and sends it back to the motor cortex. The basic structure of a motor program, however, does not seem to be generated within the cerebellum.

Postural adjustments associated with the task of raising oneself on tiptoes were investigated in a reaction time paradigm in six normal subjects and six patients with hemiparesis due to stroke. Body and ankle position in space were measured by means of a movement analysis system (ELITE). The findings indicate that the task of going up on tiptoes is performed in two steps. First, the centre of gravity is shifted forward to a position perpendicular to the forefoot. This movement is initiated by a phasic burst of EMG activity in the tibialis anterior (TA). The activity of the quadriceps femoris (QUA) aids the forward shift and together with the biceps femoris (BF) stabilizes the knee. Following these postural adjustments, the action of going up on tip-toes is performed mainly by the gastrocnemius medialis (MG). The basic pattern of preparatory (TA, QUA, BF) and focal (MG) activity was disturbed in its temporal sequence in patients with hemiparesis. The analysis of the biomechanical data showed smaller movement velocities for leaning forward and going up on tiptoes in patients, with increased movement amplitude on the paretic side. In addition, the correlation between the start of horizontal (leaning forward) and vertical (going up on tiptoes) hip movement was lost in patients. The preserved correlation between the latency of MG activity and the onset of the vertical hip movement on the paretic side in patients and the loss of correlation on the non-paretic side indicates that the EMG activity on the healthy side is adapted to the functional requirements of the affected side.(ABSTRACT TRUNCATED AT 250 WORDS)

Postural adjustments associated with the task of rising on tiptoes were investigated in a reaction time paradigm in 10 normal subjects and 18 patients with cerebellar disorders. Cerebellar dysfunction was due to either degenerative cerebellar disease, tumor, or ischemia. Displacements of the center of foot pressure (CFP) were recorded. The task, accomplished by the triceps surae muscle (executional activity, mean latency of 411 ms), is mechanically effective only if the center of gravity has been shifted forward in advance. To this effect, a phasic burst of preparatory EMG activity in the tibialis anterior normally occurs at a mean latency of 163 ms, shifting the center of gravity forward. Shortly thereafter, activity of the quadriceps femoris (175 ms) extends the knee and aids the forward shift of the center of gravity. Different aspects of this motor sequence were disturbed in individual patients: Latencies of preparatory and executional activity were uncorrelated in 15 of the 18 patients. Executional (n = 16) or preparatory (n = 13) EMG activity was tonic instead of phasic. Latencies of either preparatory or executional EMG activities or both were prolonged (n = 10). The time interval between motor preparation and execution was increased (n = 9). The trial-to-trial variability of biomechanical parameters and EMG latency was increased. Preparatory EMG activity in the quadriceps was entirely missing (n = 9), resulting in knee bending at the unsuccessful attempt to rise on tiptoes. Patients who were most severely affected had no preparatory activity at all (n = 2), and therefore were unable to perform the task.(ABSTRACT TRUNCATED AT 250 WORDS)

In standing subjects the sudden displacement of a platform toe-up evokes a stabilizing postural response of the anterior tibial muscle (long-latency reflex). A preparatory signal (acoustic warning stimulus) preceding the postural perturbation by 1 s results in a significant decrease in latency of this long-latency response. There is no parallel modulation of the afferent cortical input in terms of evoked cerebral potentials following platform tilts toe-up nor of efferent cortico-spinal functions as evaluated by means of transcranial magnetic stimulation. Conceivable underlying mechanisms include faster intraspinal and/or intracortical processing of afferent input or switching to a shorter spinal reflex pathway.

Compound muscle action potentials (CMAPs) were recorded from anterior tibial (TA) and soleus (SOL) muscles following transcranial magnetic stimulation of motor cortex in 10 healthy subjects: (1) while standing upright without support, (2) while sitting, and (3) while lying supine. The results of this study demonstrate a significant influence of posture upon the amplitudes of CMAPs. Postural facilitation presented itself, firstly, in terms of a higher incidence of bilateral activation of distal leg muscles during stance and, secondly, as significant enhancement of the amplitude of CMAPs while standing as compared to lying supine. The onset latency, however, did not disclose a significant shortening during stance. To assess the effects of preinnervation subjects voluntarily adjusted the level of TA activity to 5%, 10% and 20% of maximum isometric force respectively before cortex stimulation. Voluntary background contraction resulted in a significant increase of amplitude of CMAPs but, in contrast to postural facilitation, concomitant with a significant decrease in onset latency. These results point to a somewhat different mechanism of facilitation during stance as compared to voluntary preinnervation. But it cannot be decided whether cortical mechanisms, different descending systems, the spinal circuitry or a combination of these factors is responsible for the observed effects.

Automatic postural responses of leg muscles to the sudden displacement of standing support were investigated under four different conditions of information given to subjects in advance. Results from three groups of subjects were compared: 6 normal subjects, 10 patients with cerebellar disease, and 9 patients with Parkinson's disease. Specifically, each subject was provided with visual information about the direction and/or the amplitude of an upcoming platform tilt. For the control situation no advance information on the characteristics of platform tilt was provided. Neither the latencies nor the integrals of postural EMG-responses showed alterations with advance information. In contrast, in a control experiment in which 3 normal subjects had to perform large or small forward or backward voluntary movements of the body around the ankle joint, shorter onset-latencies of leg muscle EMG responses were observed with increasing complexity of the advance information. These results suggest that, unlike voluntary movements, postural responses to rapid surface tilts do not benefit from advance visual information on direction or amplitude of a postural disturbance.

Sequential postural adjustments associated with the task of rising on tip-toes were investigated in a reaction time paradigm in 30 normal subjects, 10 patients with Parkinson's disease and 2 patients with cerebellar lesions. The typical motor pattern observed in normal subjects consists of preparatory activity in tibialis anterior (TA, shifting the body forward) and quadriceps femoris (QUA, stabilizing the knee joint) followed by executional activity in triceps surae (TS) resulting in the movement itself. Shortening of TS without preparation would result in shifting the body backward and in flexion of the knee. The coordinated pattern of motor activities is obviously adapted to functional demands. Leaning forward or backward prior to the movement results in additional adaptive changes in the timing of preparatory versus executional EMG activity, and in the amount of preparatory muscle force. The basic pattern of preparatory (TA and QUA) and executional (TS) activity was preserved in most patients with Parkinson's disease. Reaction times were normal or only moderately delayed. Time intervals between postural preparation and execution were also normal. But trial to trial variation of reaction times was increased. In the two patients with cerebellar disorders, motor preparation in QUA was delayed beyond TS onset. EMG activity was tonic and cocontraction of antagonistic muscles occurred. The basic pattern of motor preparation and execution was again preserved. It must therefore be generated outside the basal ganglia and the cerebellum. The temporal coordination of muscular activity within each and between the two components and scaling of muscle force are under cerebellar control.

Postural control during stance was investigated using the EQUITEST system in 10 patients during recovery after an acute unilateral vestibular lesion and was compared to the time course of recovery of the static and dynamic vestibulo-ocular imbalance. During the acute phase the patients showed a characteristic pattern with normal upright stance as long as at least one accurate sensory input (visual or somatosensory) was provided and severe postural disturbances when they had to rely primarily on vestibular afferences. Both static vestibulo-ocular and vestibulo-spinal balance recovered very fast, showing basically normal results on postural testing within about 2 weeks after the lesion. Thereafter, no pathological pattern was detectable during postural testing even in patients with persistent complete unilateral vestibular lesions. Reflexive postural responses to unexpected rapid displacements of the support surface seemed not to be influenced by vestibular imbalance even in the acute phase of the lesion.

In order to further investigate the role of proprioceptive input from the legs for the maintenance of upright human posture ischemia was bilaterally applied at the level of the ankle or the thigh. Preservation of efferent innervation was ascertained by measurements of the maximal force of voluntary dorsi- and plantarflexion. Visual stabilization was excluded by eye closure. To test different frequency domains of postural stabilization, subjects were exposed to sudden ramp tilts or to sinusoidal low frequency (0.3 Hz) anteriorposterior displacements of the supporting platform. The results indicate that proprioceptive input from skin, pressure and joint receptors of the foot (ischemia at the ankle) is of minor importance for the compensation of rapid displacements, but plays a significant role when the platform moves at low frequencies. M1 and M2 responses in the stretched triceps surae and the late antagonistic response of the anterior tibial muscle (M3) are preserved with ischemia of the foot. Reversible ischemic paralysis or other (yet unidentified) mechanisms of destabilization must be responsible for the severe decrease of postural stability observed with ischemia at the level of the thigh and sinusoidal platform movements. Complete loss of the proprioceptive input from the legs leads to a pathognomonic 1Hz body tremor both under static and dynamic conditions.

(1) Methods have been developed to measure, analyse and document postural sway. The main parameters studied were: average sway amplitude; length of the sway path per unit time; sway direction and sway position histograms; and the frequency spectra of the antero-posterior and lateral sway components. (2) Postural ataxia was quantitatively studied in three groups of patients with cerebellar lesions: (a) late atrophy of the anterior lobe, (b) tumours of the cerebellar hemisphere and (c) tumours within the vestibulo-cerebellum. Characteristic differences were found. (3) Patients with anterior lobe lesion show a specific 3 Hz postural tremor in the antero-posterior direction. The tremor can be evoked by sudden destabilization in incipient cases. Less characteristic and smaller in amplitude is a mainly lateral sway component with an average frequency of 0.5 Hz. This is also seen in cases with spinal ataxia. Visual stabilization of posture is frequently preserved. Its amount does not correlate with general instability of posture. Tremor and characteristically exaggerated intersegmental responses between head, trunk, hips and legs are interpreted as the consequence of hyper-excitability of postural reflexes in these patients. (4) Patients with lesions of the hemispheres show only slight postural instability without directional preference. Their sway parameters with eyes open are within the 2 sigma range of normals and there is no significant difference from normals in these parameters even when the eyes are closed. Therefore these patients cannot be distinguished from normals by means of their platform recordings. (5) Two patients with posterior vermal and flocculo-nodular lesions were very unstable without preferred axis or frequency of instability. In contrast to the anterior lobe group the instability was characterized by the absence of intersegmental movements.

Displacements of the center of foot pressure, the hip and the head were recorded in six subjects standing on a platform, sinusoidally tilting in pitch (anterior-posterior). Stimulus frequencies ranged between 0.01 and 1 Hz. Stimulus amplitudes were 2, 4 and 6 degrees. With eyes open the displacements were minimal at 0.3 Hz. With eyes closed, however, induced sway was maximal at this frequency. The apparent lack of visual stabilization at the lowest frequency (0.01 Hz) might be attributed to a subthreshold velocity of the retinal image motion induced by the swaying body. A similar absence of visual stabilization of 1 Hz is assumed to indicate the limit of the working range of visual stabilization of posture. Independent of stimulus amplitude a phase lead of about 90 degrees was found at 0.01 Hz. This decreased with increasing frequency up to a phase lag of 100 degrees at the highest frequency (1 Hz). Head stabilization was generally more effective than hip stabilization. EMG recordings from the leg muscles suggest that with eyes closed the center of force is mainly stabilized by leg muscle activity, while with eyes open this stabilization is best, when vision allows for stabilization of body posture by intersegmental movements between head, trunk and legs.

Short (SL), medium (ML), and long (LL) latency EMG responses of leg muscles were recorded after perturbation of stance by means of a sudden toe-up tilt of a movable platform. 56 healthy children varying in age between 14 months and 15 years were investigated. All three responses were present when children were able to stand on the recording platform. The SL-response in the triceps surae muscle, which corresponds to the mono- and oligo-synaptic spinal stretch reflex, showed a decreasing latency up to the age of 5 years. This reflects the increasing peripheral nerve conduction velocity. The ML-response in the triceps surae muscle, which as the SL-response has no stabilizing effect in this experiment, showed somewhat delayed maturational changes. The LL-response in the relaxed anterior tibial muscle helps to restore upright posture even in the youngest children. Its maturational changes in terms of latency by far exceed the range that can be explained by the increase of peripheral and spinal conduction velocities. Its mechanisms of maturation, besides the biophysical optimalization of a polysynaptic network, may include learning in terms of selecting the shortest pathways by way of synaptic potentiation within structures involved in the supposedly transcortical pathway of the LL-response. Qualitative observations made during the trials showed that the pattern of postural adaptation changed with age, suggesting the development of additional intersegmental mechanisms.

The development of the visual contribution to postural stabilization has been tested using a large visual display rotating around the stationary subject's line of sight. This, in the adult, causes a marked optokinetic postural reaction with a shift of the body center of gravity toward the direction of pattern motion. Scalings of the reactions in children between 6 months and 16 years revealed three phases of development: (1) 6--12-month-old-babies show none or very little optokinetic disturbance of their newly acquired ability to sit. With the development of upright stance and gait, optokinetic influences become increasingly important. (2) Children between the age of 2--5 show a marked dependence of postural stability on vision. In them, the disturbing optokinetic stimulus leads to a marked ipsilateral postural deviation or irresistible fall. (3) From 5 to 15 years of age, visual effects on postural balance slowly decrease to their final strength in adulthood--moderate head and body tilt--in response to the rotating stimulus. It is concluded that the optokinetic loop participates rather late in the multisensory process of postural stabilization. The calibration of the three main loops, visual, vestibular, and proprioceptive, seems to be sequential and mutually interactive. Optimal functioning requires the continuous evaluation of the reafferent sensory consequences of self-generated body movements. Optokinetic destabilization of stance requires the maturity of opto-vestibulo-spinal pathways. Its clinical applicability as a maturity test, however, sofar seems limited.

Posturography is a very useful clinical instrument to quantify, analyze, and document postural sway and to differentiate patterns of instability that are typical for the different types of lesions encountered within the cerebellum and to differentiate cerebellar ataxia from spinal ataxia. An objective method of measurement like this is most valuable for follow-up studies and the surveillance of therapeutic efficacy. The method of recording short, medium, and long latency reflexes in leg muscles after a sudden tilt of a movable platform toe-up is not only totally atraumatic and fast, but more importantly it gives significant hints towards the possible site of a central lesion in the sensory-motor system.

Postural stabilization in altered visual and vestibular conditions was investigated in humans subjected to fast transient disturbances and during sinusoidal movement of the standing support. Visual inputs were varied by applying stroboscopic illumination, stabilizing the visual surround in respect to head movements, inducing apparent body movement in pitch by continuously moving stripe patterns up or down and by eye closure. Static vestibular input was modified by bending the head forwards or backwards, or to the right or left shoulder (eyes closed). Neither biomechanical parameters of standing nor EMG responses of the anterior tibial and triceps surae muscles were modified by the different visual and vestibular conditions during fast transient (80 degrees/s) platform movements 4 degrees toe-up. Continuous regulation of upright stance during sinusoidal movements (1 Hz, 0.3 Hz), however, clearly depended on the different modifications of visual and vestibular inputs. Fast transient disturbances are easily compensated in a reflex-like manner independent of visual and vestibular feedback. Continuous regulation of upright posture during slow disturbances, however, clearly depends on the evaluation of afferent information from the visual, vestibular, and proprioceptive systems.

The influence of an active lateral head tilt on postrotatory nystagmus I and II (PI, PII) was explored in normal humans. During postrotatory nystagmus the head was tilted laterally either towards the direction of the previous rotation (ipsilateral tilt) or towards the opposite direction (contralateral tilt) or the head was kept erect. Both ipsi- and contralateral head tilts led to a substantially weaker and shorter PI and PII as compared with the trial without head tilt. The time constants of the decline of PI were shortened to the ones observed in the peripheral nerve of animals. The reduction of PI and PII suggests a position-dependent tonic inhibition of the vestibular storage mechanism in the vestibular nuclei probably by the otoliths. The stronger reduction of time constant and cumulative amplitude of PI after contralateral as compared with ipsilateral head tilt may be explained by different combinations of asymmetric responses in canal and otolith afferents.

The maintenance of upright stance in humans requires the integrative evaluation of vestibular, visual, and proprioceptive information. Recent experiments prove the existence of fast vestibulo-spinal reactions in humans. Their physiological significance for the compensation of sudden external displacements of the body is minor. For low frequency stabilization of the body, however, vestibulo-spinal afferents as well as vision and proprioception are important. Unilateral loss of vestibular function initially causes a tendency to fall to the ipsilateral side as well as nystagmus and vertigo. After the first days of compensation, the remaining functional disturbance can only be detected by exposing the patient to conflicting sensory reafferents or by reducing proprioceptive and visual inputs simultaneously. The loss of vestibular function not only alters low frequency stabilization and to some extent also reflex-like fast motor patterns, but also interferes with the selection of preprogrammed motor synergies.

Methods and parameters are described to quantify body sway as measured by a force-transducing platform. Analogue data representing the coordinates of the body's center of force (COF) are fed into a digital computer. Th following parameters are then calculated and tested for their diagnostic significance: sway path (SP), mean amplitude of sway (MA), mean sway frequency (MF), their lateral and sagittal components, and the quotients sagittal/lateral of these as well as the sway area (SA) circumscribed by the COF. Quotients of eyes open/eyes closed for all these parameters determine the visual stabilization of posture. Sway position and sway direction histograms allow for a more detailed analysis of MA and SP. Despite considerable inter- and intraindividual variance of these parameters (in 28 normals), some of them seem of clinical significance not only for documentation and follow-up studies but also for differential diagnosis. In patients with cerebellar lesions (n = 12), SP and MA were up to 10 times larger with a marked antero-posterior instability, MF being above normal. Patients with labyrinthine lesions (n = 10) showed significant instability only with eyes closed, MF being slightly below normal.

Feedback and feedforward (anticipatory) adjustments of stance were investigated in 58 children with normal development aged between seven months and 14 years, and in 14 children with motor and mental retardation aged between 21 months and seven years. Feedback responses after tilting the body were elicited in all but the two youngest children, aged seven and nine months. The observed rapid decrease in feedback response latency with age suggests considerable acceleration of central transmission. Effective anticipatory postural adjustments (feedforward) were elicited only in the normal children over four years of age. The development of both feedback and feedforward continued until adolescence.

Postrotatory nystagmus I and II (P I, P II) were evoked in four normal humans by velocity steps (prior velocity of rotating chair 90 degrees/s). 4 s after the stop the head was actively tilted by the subject 90 degrees forwards, backwards, to the shoulder of the previous direction of rotation -'ipsilaterally', or to the other shoulder-'contralaterally'. In control trials, the head was kept in the previous erect position. Compared with the control experiments, P I was significantly reduced by all head tilts. Inhibition of P I was strongest with forward and weakest with backward tilts. This difference is explained by the inclination of the utricular base by 30 degrees backward with respect to the horizontal of the skull and by the elastic properties of the sensory matrix. A smaller amplitude (45 degrees) of head tilt about the roll axis leads to a weaker inhibition (28.5%) than a 90 degree tilt, which corresponds to the difference of the sine of the tilt angle and thereby reflects the mechanical force acting on the receptor layer.

The excitability of hindlimb extensor and flexor motoneurons is tonically modulated by animal tilt and a large visual stimulus rotating about the animal's line of sight. This direction-specific modulation is opposite for extensor and flexor motoneurons and opposite for optokinetic and vestibular stimuli, thus combining to a functionally significant pattern of postural reflexes.

Neurological examination and posturography showed cerebellar signs in 13 of 50 unselected patients with bronchogenic carcinoma not complicated by other diseases. The occurrence of cerebellar signs did not depend on the histological type of tumour or the extent of tumour spread. Most of the clinically affected patients had mild to pronounced cerebellar atrophy, revealed by CT. The correlation between the amount of CT-confirmed atrophy and the severity of clinical symptoms, however, was poor. Since other reasons for cerebellar dysfunction (e.g. chemotherapy, chronic alcoholism, metastases) were excluded, cerebellar signs were attributed to paraneoplastic cerebellar degeneration or to a consequence of severe neoplastic illness. The high incidence of cerebellar dysfunction in patients with bronchogenic carcinoma confirms the frequent histopathological finding of cortical cerebellar degeneration in malignant disease.

The functional role of spinal and supraspinal EMG-responses for the maintenance of upright human posture was investigated in ten healthy subjects standing on a force measuring platform, which could be rotated in pitch around an axis aligned with the subject's ankle joint. Voluntary changes of body posture prior to the platform movement by leaning forward or backward led to a change in the amplitude and temporal organization of EMG-responses as compared to platform movements starting from a neutral position. Tilting the platform toe-up while leaning backward led to an increase of the latency of the short- and medium-latency responses in the triceps surae muscle and to a decrease of the latency of the stabilizing response in the anterior tibial muscle. Functionally, a cocontraction of both antagonistic muscles could be observed which partly compensated for the destabilizing action of the "reflex" response in the stretched triceps surae muscle. In analogy, leaning forward and tilting the platform toe-down led to a cocontraction of the two antagonistic muscles. The observed changes of latencies of short-, medium-, and long-latency response show the functional variability of segmental and suprasegmental "reflex" mechanisms. EMG-activities, which are functionally destabilizing posture, can be suppressed or compensated by reflexive cocontractions of antagonists.

The functional role of short-, medium- and long-latency responses for the maintenance of upright posture was investigated in twenty healthy subjects standing on a platform which could be rotated in pitch around the subject's ankle joints. Tilting the platform toe-up evokes a stretch reflex in the triceps surae muscle (TS, latency 55-65 ms) and at higher speeds and amplitudes of platform displacement a medium-latency response (latency 108-123 ms). Both responses functionally destabilize posture, since they enforce the induced backward displacement of the body. Compensation of body displacement in this situation is achieved by a long-latency EMG response in the anterior tibial muscle (TA 130-145 ms). Platform movement toe-down elicits a rather small medium-latency response in TA (103-118 ms), but no short-latency response. A late compensatory response occurs in the triceps surae muscle (latency 139-170 ms). The mean latency of the late antagonistic EMG response was significantly shorter than that of a voluntary movement triggered by a somatosensory stimulus. Integrals of rectified EMG responses from the two muscles were linearly related to the amplitude and to a smaller degree to the velocity of platform displacement. The slope of this function (gain) varied depending on the direction of ankle displacement and the functional importance of the subsequent EMG responses. Destabilizing short- and medium-latency responses of the stretched muscle had a lower gain relative to amplitude than the late stabilizing response of the antagonist. This functionally adaptive modulation of gain was not seen in relation to the rate of platform displacement.