Ip. Howard et L. Childerson, THE CONTRIBUTION OF MOTION, THE VISUAL FRAME, AND VISUAL POLARITY TO SENSATIONS OF BODY TILT, Perception, 23(7), 1994, pp. 753-762
Three types of visual information contribute to the sense of self orie
ntation with respect to gravity: visual polarity of objects with a dis
tinct top and bottom, the principal vertical and horizontal lines of t
he visual environment, and visual motion. Three visual displays were d
esigned to investigate the contribution of each visual feature to illu
sory self tilt: a large sphere lined with dots, a cubic room lined wit
h dots, and a furnished room with floor and ceiling. In experiment 1 t
he dotted room and the furnished room were tilted to various angles ab
out the roll axis of the erect subject who set a visual line and an un
seen rod to the apparent vertical. In the dotted room, settings were m
ade either with respect to the nearest surface to the horizontal or wi
th respect to the nearest diagonal of the room. In the furnished room,
settings were made with respect to the nearest horizontal wall but no
t with respect to diagonals. In experiment 2 each of the three display
s was rotated at constant velocity and subjects' responses were classi
fied into four categories: illusory self tilt at a constant angle, alt
ernating self tilt with the body becoming erect each time a surface be
came horizontal, continuous head-over-heels self rotation, and a feeli
ng that the body was supine. Almost all responses were of constant til
t in the sphere. Constant and alternating tilt were the most common re
sponses in the dotted room. In the furnished room 60% of subjects expe
rienced full head-over-heels self rotation.