Role of vestibular and neck inputs for the perception of object motion in space

Abstract

The contribution of vestibular and neck inputs to the perception of visual object motion in space was studied in the absence of a visual background (in the dark) in normal human subjects (Ss). Measures of these contributions were obtained by means of a closed loop nulling procedure; Ss fixed their eyes on a luminous spot (object) and nulled its actual or apparent motion in space during head rotation in space (vestibular stimulus) and/or trunk rotation relative to the head (neck stimulus) with the help of a joystick. Vestibular and neck contributions were expressed in terms of gain and phase with respect to the visuo-oculomotor/joystick feedback loop which was assumed to have almost ideal transfer characteristics. The stimuli were applied as sinusoidal rotations in the horizontal plane (f = 0.025-0.8 Hz; peak angular displacements, 1-16 degrees).

Results: (1) During vestibular stimulation, Ss perceived the object, when kept in fixed alignment with the moving body, as moving in space. However, they underestimated the object motion; the gain was only about 0.7 at 0.2-0.8 Hz and clearly decreased at lower stimulus frequencies, while the phase exhibited a small lead. (2) During pure neck stimulation (trunk rotating relative to the stationary head), the object, when stationary, appeared to move in space counter to the trunk excursion. This neck-contingent object motion illusion was small at 0.2-0.8 Hz, but increased considerably with decreasing frequency, while its phase developed a small lag. (3) Vestibular, neck, and visuo-oculomotor effects summed linearly during combined stimulations. (4) The erroneous vestibular and neck contributions to the object motion perception were complementary to each other, and the perception became about veridical (G approximately 1, phi approximately 0 degree), when both inputs were combined during head rotation with the trunk stationary. The results are simulated by an extended version of a computer model that previously had been developed to describe vestibular and neck effects on human perception of head motion in space. In the model, the perception of object motion in space is derived from the superposition of three signals, representing "object to head" (visuo-oculomotor; head coordinates), "head on trunk" (neck; trunk coordinates), and "trunk in space" (vestibular-neck interaction; space coordinates).