Posterior semicircular canal nystagmus is conjugate and its axis is parallel to that of the canal.

A patient with a postoperative fistula of the left posterior semicircular canal is presented. Negative pressure in the external ear canal produced upbeat-torsional nystagmus, which was recorded in three dimensions using binocular scleral search coils. The nystagmus was conjugate, without skew deviation, and its trajectory corresponded to the anatomic axis of the left posterior canal. The current study helps validate Ewald's first law in humans: the axis of nystagmus should match the anatomic axis of the semicircular canal that generated it. This law is clinically useful in diagnosing pathology of the vestibular end-organ, such as benign paroxysmal positional vertigo or the superior semicircular canal dehiscence syndrome.

Real-time rotation vectors.

Rotation vectors are a useful way of describing eye position without reference to arbitrary axes of rotation since any eye position can be reached from the reference position by rotation about a single axis. A real-time display of rotation vectors would not only help to acquire more reliable data, but would also widen the range of possible eye movement experiments. We describe a novel PC based data acquisition and analysis system which calculates and displays rotation vectors, velocity vectors and Listing's plane in real-time using voltages obtained from a two field coil system. The system was implemented using LabVIEW and optimised using Code Interface Nodes. Off-line processing can be sped up by varying parameters that indicate the amount of available RAM. During processing Listing's plane data can be rotated horizontally, vertically and torsionally. A computer controlled laser target changes position randomly every half second and so the targets are evenly spread, producing an appropriate range of eye positions which are used to calculate Listing's plane.

Vestibulo-ocular reflex pathways in internuclear ophthalmoplegia.

We measured the vestibulo-ocular reflex (VOR) during head impulses in a patient with right-sided internuclear ophthalmoplegia. Head impulses are rapid, passive, high-acceleration, low-amplitude head rotations in the direction of a particular semicircular canal (SCC). Adduction of the right eye was abnormally slow during right lateral SCC head impulses. The VOR during left posterior SCC impulses was severely deficient in both eyes, but the VOR during left anterior SCC impulses was only slightly deficient. We suggest that the vertical vestibulo-ocular pathways in humans are connected in SCC-plane coordinates, not the traditional roll and pitch coordinates, and that anterior SCC signals do not travel exclusively in the medial longitudinal fasciculus.

Effects of unilateral vestibular deafferentation on the linear vestibulo-ocular reflex evoked by impulsive eccentric roll rotation.

The effects of unilateral vestibular deafferentation (UVD) on the linear vestibulo-ocular reflex (LVOR) were studied by measuring three-dimensional eye movements in seven UVD subjects evoked by impulsive eccentric roll rotation while viewing an earth-fixed target at 200, 300, or 600 mm and comparing their responses to 11 normal subjects. The stimulus, a whole-body roll of approximately 1 degrees, with the eye positioned 815 mm eccentric to the rotation axis, produced an inter-aural linear acceleration of approximately 0.5 g and a roll acceleration of approximately 360 degrees /s(2). The responses generated by the LVOR comprise horizontal eye rotations. Horizontal eye velocity at 100 ms from stimulus onset in UVD subjects was significantly lower than in normal subjects for all viewing distances, with no significant difference between ipsilesional and contralesional responses. LVOR acceleration gain, defined as the slope of actual horizontal eye velocity divided by the slope of ideal horizontal eye velocity during a 30-ms period starting 70 ms from stimulus onset, was bilaterally significantly reduced in UVD subjects at all viewing distances. Acceleration gain from all viewing distances was 1.04 +/- 0.28 in normal subjects, and in UVD subjects was 0.49 +/- 0.23 for ipsilesional and 0.63 +/- 0.27 for contralesional acceleration. LVOR enhancement in the first 100 ms by near viewing was still present in UVD subjects. LVOR latency in UVD subjects (approximately 39 ms) was not significantly different from normal subjects (approximately 36 ms). After UVD, LVOR is bilaterally and largely symmetrically reduced, but latency remains unchanged and modulation by viewing distance is still present.