Abstract
Three experiments investigated the organization of linear vection (the visually-induced illusion of self motion). Previous studies have shown that optokinetic stimulation of the peripheral visual field in a fashion specifying upward or downward self motion yields shorter vection onset latencies (VOLs) than equivalent visual stimulation for forward or backward self motion (Giannopulu & Lepecq, 1998; Lepecq et al., 1999). This finding has most often been attributed to differential visual-vestibular conflict, that is, the less sensitive saccules generating less of a conflict signal than the utricles. The major goals of this study were to replicate this finding and to probe its frame of reference by finding out whether the advantage of upward/downward vection is defined relative to the retina or relative to gravity.
A vection chamber was constructed to generate lamellar peripheral optokinetic stimulation specifying either forward, backward, upward, or downward self motion. Participants sat at the open end of the chamber and reported when they first experienced linear vection (i.e., VOLs) and after 30 s reported a subjective rating of peak vection potency (PVPs).
Experiment 1 explored the optimal stimulus conditions for inducing linear vection. Experiment 2 replicated the finding of shorter VOLs for upward/downward vection relative to forward/backward vection and also found an equivalent pattern of results for PVPs. In Experiment 3, participants both looked forward into the chamber while seated and upward into the chamber while supine. Stimulus direction interacted with body position such that images translating vertically on the retina produced stronger vection (i.e., shorter VOLs, higher PVPs) than images translating horizontally on the retina. These results suggest that linear vection has a retinotopic rather than gravitational (environmental) frame of reference. Results are discussed in terms of congruency and conflict models of visual-vestibular interaction.