Neurophysiology of balance

Writing a detailed chapter on the anatomy and physiology of balance in 1999 would be quite conceited on our part. Instead, we prefer to redirect the reader to a number of publications written by other qualified individuals. This said, we do feel that for practical reasons, for a thorough comprehension of the following section it is important to provide an overview of this physiology. In other words, to explain what essential information is required to have a good understanding of both the pathologies and their symptoms. Vestibular rehabilitation after all is nothing more than the manoeuvring of the physiology. Knowing this will enable a better grasp of therapeutic procedures.


Some of the following hypotheses of the system operations do not necessarily appear in available literature. These hypotheses are the result of questions asked while monitoring patient behaviour. These hypotheses are part of the approaches employed by different research laboratories working in our particular field. That is to say that although vestibular rehabilitation has now reached a certain level of sophistication, there is a great deal of work still ahead of us and plenty of things to learn.


Balance is a function carried out by nerve centres that rely on a periphery system. The periphery system is two-fold, comprising neurosensory afferent activity and neuromotor efferent activity. Issues dealing with neuromotor matters will not be dealt with in this paper. Neuromotor pathologies are responsible for what neurologists call standing and walking disorders. Physiotherapists are well acquainted with the rehabilitation of the different ataxic disorders.

In this function, the neurosensory part, thanks to information received from the peripheral sensors, gives the order while the neuromotor part carries out the order. With regard to the pathologies that interest us, it is assumed that the neuromotor part is apathalogical.

Standing up requires three sensory inputs:


Somatosensory input primarily consists of the proprioception of the skeletal muscles and osteo-articular sensors. The tibiotarsal and coxofemeral are the principal joints.

Visual input comprises foveal vision used to watch things and peripheral vision used to bring an image that appears in the field and whose interest is significant into view of the fovea. The peripheral retina also contains sensors that capture speed, movement and the visual scene. Together, they allow us to see.

Vestibular input comprises the posterior labyrinth that includes the semicircular canals and the otolithic maculae. There are three semicircular canals per labyrinth. There is an external canal on the Frankfurt plane and two vertical canals. These two vertical axes form a 90° angle. These canals are full of a liquid: endolymph. Each one has a bulge at one end called the ampulla. This ampulla is enclosed by an anhist membrane: the cupula. The base of this cupula rests on a hair epithelium. The hair cells are the mechano-transmitters of information. Together, the canal and cupula form an angular accelerometer. Remember that:


The otolithic maculae are comprised of calcium carbonate crystals embedded in a gelatinous material on a bed of hair cells. The density of the maculae is much greater than that of the endolymph. When the head moves for example, the macula, under the effect of gravity, shifts forward. This shearing will last as long as the head remains in that position. This shearing movement will move the cilia of the hair cells and will send a message to the nerve centres. The message will indicate a change in the direction of the gravitational accelerator vector with respect to the axis of the head. We even believe that the otolithic system plays a role in inhibiting ascending eye movements.

Afferent data provided by the peripheral sensors are analysed and compared by the nerve centres. Having checked the congruity and the redundancy of the data, they send an order to the efferent systems to adapt the posture to the task in hand.

There is however both a strategic and a hierarchical approach when using said data. Some pieces of information carry more weight than others.

To simplify things, proprioception can be considered a low frequency system. In fact, if it was not for proprioception, we would not be able to remain still. Proprioception ensures the smoothness of segmental tonic muscle contraction and ensures posture control. Irrespective of proprioceptive control, postural muscle tone must overcome acceleration due to gravity to maintain an upright position. The role of the otolithic system is to permanently measure acceleration due to gravity and to therefore control, amongst other things, the tonicity of the extensor muscles.

Visual input is used much more frequently, but for a greater variety of the frequency range and of the task to be completed, either slow eye movements (tracking movements), or quicker eye movements (saccadic eye movement) can be used, but for the latter, sight would be suppressed during movement.


Vestibular input responds to high frequencies. Its role is essential as it interoperates with the other two inputs. We have seen the otolithic interaction in an unmoving position, but in real life, we do move and we do shift about. The vestibule is the organ used for moving around, for moving the head or the body within a three dimensional space.

The vestibulary system (and by vestibulary system we mean the complex sequence of events from the periphery to the centres; the peripheral sensor being nothing more than a link on this chain) is to stabilise the visual scene during movement and/or a change in the position of/and the head or body. If the eye was fixed in the eye socket or if it only used its own movements, when walking we would experience a condition called oscillopsia. In other words, the surrounding landscape would appear blurred or would appear to jump each time we took a step. When walking, the head is subject to a passive lateral movement of approximately 1Hz. So we would therefore see vertical oscillation of the visual scene at a frequency of 1Hz. Expressed differently: the vestibular system with its vestibular-ocular interaction allows us to enjoy a stable image of the target object on our retina. The fact that the eye and the vestibule are located in the head enables us to state that the vestibule stabilizes the gaze and also acts as the proprioceptor for the head.


Stable vision is essential for balance. It is impossible to maintain balance in a visually unstable universe.

That covers the fundamentals. But in real terms, the coherent operating of the chain as a whole is more complex. Adapting the standing position is a result of learning. It can be accepted that all actions are the result of learning. By extension, it can also be accepted that vestibular rehabilitation is not rehabilitation per se, but rather the learning of a new way to use the systems. Systems that no longer function according to logic acquired through everyday use, but with a functional anomaly of one of its parameters. Through a process of learning in normal subjects, this multimodal adaptation is essentially quite simple for one reason: to complete a task as quickly as possible, without conscious and /or voluntary participation, as best as possible, or as perfectly as possible while expending the least amount of energy possible.

Normal subjects are free to use whichever of the different inputs is most effective. This use is subject to a strategy that is unique to each subject. The difference in activity of each one, the difference in speed of the action requires the systems to be adapted in order to be able to comply with the requisites when carrying out a task. The multiplicity of strategies explains the extraordinary inter-individual differences observed in normal subjects during the different experiments and balance assessment tests.

For a subject with a pathological condition, missing information, or erroneous information, appears when the system is used. This anomaly, when it involves the vestibular system, causes a great stir due to the tremendous number of interactions of the system on the other inputs. This anomaly interferes with the logical functioning that is “habitually” used. If a solution is not found quickly, either by spontaneous recovery or by compensation, a kind of panic attack will be produced and the brain will go into a state of alarm. Action and movement will therefore become conscious. This new situation consumes a considerable amount of energy.


The problem can only be resolved if:

This can be attested to in the experiments carried out by Michel LACOUR (National Scientific Research Council (CNRS) Marseille). He has shown the compensation of a monkey following unilateral labyrinthectomy in a number of situations:

The return to “normalcy” of the monkey that underwent sensory restriction took longer and was incomplete. On the other hand, the monkey who had to carry out an “activity” underwent a quicker and better recovery. We therefore know that with vestibular rehabilitation, what has been acquired has been acquired and will remain as long as a need for compensation exists. Although admittedly there is some degree of decompensation, it is often consequential such as:

In these cases, diagnosis should be re-examined.


We are very aware that for a subject to be considered a "vestibulopath", to return to a normal social and professional life, one must:


The vestibular system has another role, one of many: the anticipatory positioning of the gaze. In effect, we move towards what we are looking at (Alain BERTHOZ). The vestibular system ensures that the gaze is aimed in the direction we are moving in before the rest of the body changes direction.


So from the above, it can be deduced that balance starts in the head and then descends down to the feet. From a rehabilitation perspective, this means that the priority is to recover the ability to direct and anticipate the gaze. The body followed by the chain reaction of a set of reflexes. In summary, and succinctly, there are:


We should also not forget the vestibulospinal reflex which controls the tonicity and is in charge of making postural readjustments, whereby the vestibule becomes the trigger for the parachute reaction. The best example of this reflex is during what Georges FREYSS describes as: vestibular omission. Further on we will see what that entails in detail but for now, briefly, it describes the non-employment of vestibular input.

And finally, it is widely known that the vestibular system acts as an angular accelerometer. This means that at a constant speed, the system is at rest. But this is not the only situation; in fact, a reading of zero is not without reason. The zero reading means rest so we are motionless. This final intimation is the most obvious but there is also the condition where the system does not respond because it has been destroyed. So in conclusion, the value of zero corresponds to three scenarios:

What will resolve the problem? : The eye


There appears to be consistency:

On the other hand, if the system has a lesion, there will be conflict of information and consequently symptoms will manifest themselves.


The reflex that appears to eliminate any doubt when the vestibular system is silent is the optokinetic reflex: when we gaze upon a visual scene while moving, the eye is subject to a tracking movement that is attuned to the movement of the scenery. The moment arrives when:


But there is another optokinetic nystagmus which in this case is caused by the activation of the optokinetic reflex. This nystagmus is purely an involuntary reflex that follows the movement of the scenery. It is caused by stimulation of the whole visual field. In other words, the stimulus fully envelopes the stimulated subject. The stimulation surface is at a maximum. The optokinetic nystagmus, which we will call cortical nystagmus to simplify things, can be caused by the movement of a single point of light. It will have a gain of one. On the other hand, the subcortical optokinetic nystagmus caused by the activation of the optokinetic reflex will have a gain of about 0.75. It will appear following a period of around twelve seconds of stimulation. During the first twelve seconds, there is a cortical optokinetic nystagmus presumably associated with the subcortical nystagmus. After twelve seconds, only the subcortical nystagmus remains. The reflex is produced in such a way that the system functions on its own. What is particular about this reflex is that it generates additional symptoms: a sensation of movement and an ipsilateral postural deviation in the direction of the stimulus.


The sensation of movement is proof of reflex engagement. Postural deviation has this peculiarity: it is not a loss of balance but rather an active correction of the sensation of movement of the room the subject is in; apparent movement in the opposite direction to that of the stimuli. It is as if the subject perceives a motionless stimulus, a movement opposite that of the room which acts as a screen. To avoid falling with the room, which from a spatial perspective is no longer the Euclidean reference, the subject attempts to maintain their position. This extremely powerful reflex has plenty to teach us when assessing damage to the balance function. It is likely that there are still things to discover and to learn about this subcortical reflex.



In conclusion, it can be seen that everything is interconnected. The correct functioning of the system so as not to have any symptoms is complicated. What characterizes the use of all these afferents is neuroplasticity. In 1999, it is becoming difficult to be satisfied with a subcortical physiology and to simply gloss over the role of the nerve centres. Besides the cerebellum, an organ that contributes to precision and measuring, we can not simply ignore all the cortical regions, especially the parieto-insular vestibular cortex. To simplify things, all these regions come under the single heading "vestibular cortex ". Contrary to the visual cortex, it is not a geographically specific region, but rather a collection of regions that take part in the creation of routines. The routines that are executed by the peripheral system aim, as we have already explained, to complete a task in the minimum amount of time expending the least amount of energy.


The regions located in the vicinity of the cortex are responsible for the spatial representation of the function of balance. We believe that the vestibular system, as a whole, allows us to locate a front right egocentric deviation. This front right egocentric deviation would, compared to a front right reference deviation, split into two parts (right and left), and achieved through learning from experiences since childhood, and this smoothly operating ensemble would make its way through a three dimensional universe. Therefore: any dysfunction is reflected in a bias. This bias is responsible for front right egocentric deviation instead of front right reference deviation, which would therefore enable us to understand all the lateralisations and other deviations observed in patients.


Furthermore we believe that the front right reference deviation would be "suspended" from gravity by the otolithic system. Experiments on vertical optokinetic stimulations by T. TSUZUKU, A. SEMONT, A. BERTHOZ (in press) make us think that the otolithic system acts as an inhibiter on ascending ocular movements. If as we think, the front right referent is under the control of the otolithic system, it is understandable that unilateral otolithic damage might be perceived as lateral collapse. (Tilt).


As a result, we believe, as far as spontaneous nystagmus is concerned, that the slow phase would be, as we have always taught, the result of vestibular asymmetry. The slow phase would be "passive" while the quick phase would be under the control of the cortical regions and reflects a stimulating action to correct the front right egocentric deviation. The fast stage would therefore be “active”.


All these hypotheses form the basis of the new approaches to clinical research.

Neurophysiologie de l'équilibration