Stabilization of oculomotor and postural responses results from a complex multisensory integration, which can be defined as the process of matching multiple internal representations of an external event (head and trunk rotations), obtained from different sensory modalities (visual, vestibular and proprioceptive), into a single intrinsic frame of reference, in which appropriate motor commands can be coded. Several characteristics of the stabilizing synergies make them a unique model to study how sensorimotor transformations are implemented by the central nervous system. Firstly, these synergies have a purpose, which is well defined and requires the computation of an equally well-understood quantity, i.e. self-motion. Secondly, the input (velocity of the visual world detected by the eyes, velocity of the head detected by the vestibular system, etc.) as well as the static and dynamic motor responses to these inputs (eye and head movements, change in the skeletal geometry) is quantifiable with precision. Thirdly, the stabilizing synergies have been shown to display plasticity in adult humans and animals either under the pressure of environmental changes (prism adaptation, repetitive stimulation) or pathological changes (vestibular neuritis, unilateral labyrinthectomy, vestibular neurectomy). Consequently, the neural networks implementing the sensorimotor transformations underlying gaze and postural control have been intensively studied for the past sixty years and they are well described. It remains that despite its plasticity, gaze and postural stabilization can be disturbed by senescence and several pathologies; in an ageing population, vertigo, perception of instability and falls have become of great concern. Falls are the leading cause of both non-fatal and fatal injuries among the over-65 age group (Sleet et al. 2008). With that perspective, several tests were developed to probe the function of the vestibulo-ocular and vestibulo-spinal synergies, in order to detect vestibular deficits in due time. The vestibulo-ocular reflexes of semicircular canal origin can be tested by using caloric and rotatory stimuli and the head impulse test. The cervical and ocular vestibular evoked myogenic potentials (oVEMPs) are used to test the saccular and utricular functions, respectively. A common denominator of these clinical tests is that they rely for their interpretation on electrophysiological studies performed in animal models. That is, their validity relies on the fact that the activation of a given sensor and the resulting motor behaviour was demonstrated to be similar in patients and in the animal models where the neuronal vestibulo-ocular and vestibulo-spinal networks could be described. In that context, oVEMP recordings were recently proposed as a new and increasingly popular technique to clinically test otolith function (Todd et al. 2007). oVEMPs are evoked by bone-conducted sound and vibrations stimuli to probe the vestibulo-ocular pathways and are thought to result mainly from utricular stimulation (Chiarovani et al. 2011). In animal models, selective utricular nerve stimulations evoke excitatory and inhibitory responses in the contralateral superior oblique motoneuron pool and the ipsilateral inferior oblique motoneuron pool (Uchino et al. 1996). However, it remained to ascertain the muscle of origin of the oVEMP when tested in humans. This is precisely what Weber et al. (2012) elegantly succeeded in demonstrating in an article in this issue of The Journal of Physiology, by recording single motor unit activity in human extra-ocular muscles. Reviving a heroic tradition from the first days of physiology of testing on themselves a quite invasive procedure, the authors recorded their own inferior oblique (IO) and inferior rectus (IR) muscle activity with concentric needle electrodes, following bursts of 500 Hz skull vibration and sound – two otolith stimuli. After bone vibration of the midline forehead, highly synchronous excitation of the IO and the IR muscles was recorded, with a reciprocal activation pattern after a short latency. Sound stimulation of one ear produced short-latency excitation of the contralateral IO muscle. Simultaneous needle and surface recordings proved the IO to be responsible for the oVEMP. By using oVEMP to investigate eye muscle activity during the translational vestibular ocular reflex in humans, not only did Weber and colleagues (2012) confirm a crucial physiological hypothesis, they also reopened a wide field for future research, much beyond the peculiar setting of self-experimentation. Needle electromyography of extra-ocular muscles in humans has slowly declined after its golden age in the 1960's, although technical advances now allow for more refined single motor unit recordings. While this technique is less frequently used for clinical purposes in the era of molecular genetics (Finsterer 2008), it remains essential for physiological studies, as it is the only way to help bridge the many existing gaps between animal experimentation and novel clinical tests or findings.