In healthy subjects, head tilt upon cessation of a constant-velocity yaw head rotation shortens the duration of postrotatory nystagmus. The presumed mechanism for this effect is that the velocity storage of horizontal semicircular canal inputs is being discharged by otolith organ inputs which signal a constant yaw head position when the head longitudinal axis is no longer earth-vertical. In the present study, normal subjects were rotated head upright in the dark on a vertical-axis rotational chair at 60 degrees/s for 75 s and were required to perform a specific task as soon as the chair stopped. Horizontal position of the right eye was recorded with an infra-red video camera. The average eye velocity (AEV) was measured over a 30-s interval following chair acceleration/deceleration. The ratios (postrotatory AEV/perrotatory AEV) were 1.1 (SD 0.112) when subjects (N=10) kept their head erect, 0.414 (SD 0.083) when subjects tilted their head forward, 1.003 (SD 0.108) when subjects imagined watching a TV show, 1.012 (SD 0.074) when subjects imagined looking at a painting on a wall, and 0.995 (SD 0.074) when subjects imagined floating in a prone position on a lake. Thus, while actual head tilt reduced postrotatory nystagmus, the imagination tasks did not have a statistically significant effect on postrotatory nystagmus. Therefore, velocity storage does not appear to be under the influence of cortical neural signals when subjects imagine that they are floating in a prone orientation.