Models of depth perception typically omit orientation despite the potential usefulness of the ground plane and of head position information for interpreting disparity (Blohm et al., 2008). Perceived depth is shortened when visual cues to upright are rotated relative to gravity (Mander and Harris, 2012), suggesting a role of orientation in depth perception. Here we used the York University Tumbled and Tumbling Room facilities, realistically decorated rooms systematically arranged to vary the relative orientation of visual, gravity and body cues to upright. We exploited size/distance constancy to assess perceived depth. Observers matched the perceived length of a tactile rod to a visual line (controlled by a QUEST adaptive procedure) projected on the wall of the facilities.
The line was consistently set longer when the room and subject were rotated 90° relative to gravity, compared to when subjects were upright suggesting that the opposite wall appeared closer. The effect was modulated by binocular cues, nearly doubling under monocular viewing. After systematically varying the room’s and subject’s orientation, the largest “expansion” was obtained when observers were upright in a rotated room (looking at the ceiling), suggesting that the illusion is induced primarily by rotation of visual cues. The effect was reduced when subjects were rotated toward the ceiling in an upright room, suggesting that the perceived distance change induced by perceived body tilt is modulated by actual body tilt. Thus, vestibular orientation cues are a mediating factor when there is a mismatch between visual and gravitational upright.

psychophysics; depth perception; distance perception; illusion