menu Contact Us Dizzy Patients Health Care Providers Research BPPV DVD Tai Chi DVD Understanding Dizziness Acknowledgements Disclaimer Quoting

Subjective Horizontal and Vertical Testing

Timothy C. Hain, MD

Please read our disclaimer Return to Index. Page last modified: April 24, 2017 button Return to testing index

Otolithic organs. Schematic of the utricle (horizontal) and saccule (vertical). These sensory organs in the inner ear primarily respond to linear acceleration such as due to orientation to gravity. Utricular lesions may shift the subjective vertical.

The purpose of this test is to detect abnormal subjective tilt. In normal persons, the ability to perceive verticality is quite good. This ability is dependent on input from visual, vestibular and somatosensory systems. It also depends on a functioning central nervous system.(Yelnik, Lebreton et al. 2002).

The otolithic organs (see above) in the vestibular system sense gravity. Both the utricle and saccule (see figure above) contribute to the sense of verticality. After injury to the otoliths, or to the nerve that transmits impulses from the otoliths and other parts the ear to the brain, judgement of vertical may be altered. The inner ear may falsely suggest that the head is tilted while the eyes and somatosensory systems suggest that one is upright. Thus there is a sensory conflict. There can be an interaction between vision and the otoliths in that an otolith imbalance may transiently cause the eyes to counter-roll, which literally tilts ones vision.

Visual influences on verticality may be measured by putitng a frame around a bar. Alterations of the angle that the frame makes with vertical may disturb a persons judgement of verticality of a bar. One may eliminate the visual component to vertical by putting a person in a dark room, and asking them to orient an illuminated bar. Smaller effects can be measured using the Maddox Rod.

The subjective vertical and horizontal is inclined towards the direction of the preset angle, so this should be randomized. (Pagarkar et al, 2008)

There are several types of subjective vertical.

The subjective vertical can be subdivided into several components. The subjective visual vertical (SVV) is determined by having subjects adjust a visible luminus line in complete darkness to what they consider to be upright, earth vertical. The "haptic" vertical (HV) is assessed by manipulation of a rod to the earth-vertical position with both eyes, while the subject's eyes are closed. Haptic means derived from sense of touch. The subjective postural vertical (SPV) is the position of the head or body with respect to true vertical.(Sharpe, 2003)

Subjective visual vertical

Friedman, in 1970, studied subjective vertical in a variety of clinical situations. It is well known that normal subjects can adjust an illuminated rod in an otherwise completely dark room to vertical within a mean error of less than 2 deg. The test works in children as well as adults (Brodsky et al, 2016). Friedman concluded that severe derangement of this test is confined to brainstem lesions and the immediate postoperative period of peripheral vestibular lesions.  The SVV tilts toward the side of lesion.

Persons with vestibular lesions may orient the bar tilted as much as 10 degrees. (Garcia and Jauregui-Renaud, 2003; Vibert, Hausler et al. 1999). The SVV reverts to normal in labyrinthectomy by one year. In vestibular nerve section, a small deviation may persist after neurectomy even after 4 years (Vibert and Hausler, 2000). In patients with Menieres disease, operated with labyrinthectomy, a marked deviation toward the operated side was found acutely, with resolution over weeks.

Subjective vertical deviation was further assessed, with or without unilateral centrifugation by Schonfeld and Clarke (2011). They reported that a substantial subset of patients have deviations ONLY during UC, and suggested that these are the compensated patients.

Patients with bilateral loss of labyrinthine function can also adjust the vertical on average, but show greater individual differences. Patients with cerebellar lesions generally showed good accuracy of the subjective vertical, suggesting that in patients with spontaneous nystagmus, the lack of a deviation of the subjective vertical substantiates a cerebellar lesion.  Patients with brainstem lesions, however, frequently show extremely profound deviations, some as great as 8 degrees.

Studies have also shown a counter-roll of the eyes.(Curthoys, Dai et al. 1991).

Subjective postural vertical

The subjective postural vertical (SPV) may be disturbed by cerebral lesions.(Perennou et al. 1998). Tilts of SPV are reported while the SVV remains accurate. These individuals may exhibit contraversive pushing: they lean or push themselves towards the side of hemiparesis and hemisensory loss, perhaps in an attempt to compensate for the tilt of the SPV.

Subjective haptic vertical

The haptic vertical (HV) may be unaffected by vestibular lesions(Bronstein et al. 2003) that disrupt the subjective visual vertical.

Clinical Testing:

SVV may be useful in clinical practice. Devices to measure it are in development. The "barrel" test is presently the most promising.

Unilateral centrifugation (UC) is a reasonable method of confining the stimulus to one ear (Schonfeld and Clarke, 2011), but it requires equipment that is prohibitively expensive. Response is measured by subjective vertical.

Ocular VEMP testing has also been proposed as a method of detecting unilateral utricular lesions.

Retinal photo's and OCT testing can also be used to establish objective ocular tilt.


The subjective vertical has been the topic of a large amount of research -- more than 100 papers can be found in Pubmed using the keywords subjective vertical AND vestibular. This may be related to the simplicity of this test.

SVV is slightly biased during eccentric rotation in persons with BPPV. This probably reflects unilateral utricular disturbance. (Hong et al, 2008)


Copyright (c) Timothy C. Hain, MD 2002-2006
Copyright April 24, 2017 , Timothy C. Hain, M.D. All rights reserved. Last saved on April 24, 2017