Subjective visual vertical (SVV): Effects of head position and visual conditions

The subjective visual vertical (SVV) is often used to evaluate asymmetry of otolithic function clinically. However, there is no established method of measuring SVV. This study assessed the effects of head location and visual (monocular or binocular) condition on measurements of SVV. The reproducibility of SVV was also examined. The normal subjects with no history of ear diseases participated in this study. When myopic or astigmatic, they were allowed to wear glasses or contact lenses during the measurements. SVV was measured with the seated subject's head loosely fixed in the upright position. A bar was placed 30 cm in front of the subject's face to assess SVV using a joystick. To eliminate clues for verticality, a hemispheric dome 60 cm in diameter and covered with a random pattern of dots covered the entire visual field behind the target. Each session consisted of a total of nine head and visual conditions (three different head locations times three different visual conditions). Ten measurements were carried out for each condition. In all, 90 measurements were carried out in each session. Two measurement sessions were carried out daily, 15 minutes apart, on five consecutive days. Our findings showed that (1) the reproducibility of the results was poor, even under the same head and visual conditions; (2) of all nine conditions used, measurement in the binocular condition while the visual target was precisely in front of the face produced the most reproducible results with the least variance; (3) when the visual target was placed to the right of the subject's field of view, the average SVV tended to shift in a counter-clockwise direction from the same perspective; and (4) the average SVV when viewed in the binocular condition was close to that when viewed in the monocular condition using the predominant eye. These results suggest that the gravity vector is not fixed, but is moving constantly, even in normal subjects. They also showed that when possible, it is desirable to measure SVV in the binocular condition, with the target precisely in front of the subject.