Recently it was shown that abrupt head acceleration impulses induce short latency response in the neuronal pathways of the vestibular system and a short latency vestibulooculomotor response. This phenomenon cannot be explained only with the frame of the classical torsion pendulum model of the cupuloendolymphatic system and passive reaction of the vestibular receptors to the cupular displacement. All three stages of stimulus transduction: mechanical, receptive and neuronal, are revisited in recent studies, and these findings are analyzed in the present paper. The head impulse test of Halmagyi and his colleagues, the high frequency head oscillations using special helmet [Tabak et al., 1995, 1997a, b], the autorotation test [O’Leary, 1990; Hirvonen 1998], the technique of recording vestibular evoked potentials to acceleration impulses [Elidan at al.] are discussed. In spite of some deficits in our knowledge, the exclusive diagnostic value of the high-frequency and high-intensity (impulse) stimuli became already obvious. It is concluded that the proper differentiation between the mechanical, receptor and afferent neuronal parts of the vestibular end organ may be achieved only by applying stimuli with different frequency content. Further investigations are requested to elucidate the specific mechanisms of vestibular sensitivity to the different stimuli used for this purpose, and the possible directions of these studies are discussed.