Search

Groups of 24 Os estimated the slant of a circle and a rectangle (Groups I, II, and III) or a circle and a square (Group IV) by setting a tilt-rod. The forms were viewed at 45° slant, monocularly and binocularly, with fixed head, under reduced conditions. Exposure durations were: for Groups I, II, and IV, 5, 4, 3, 2, and 1 sec., imposed in descending order; and for Group III, five 3-sec. exposures. Groups I and III left the tilt-rod at the response setting, and Groups II and IV returned it to vertical, after each response. With reduced exposure, over-all estimates of slant decreased for the response setting (Group I), but decreased less (Group IV), or did not change significantly Group II) for the vertical setting. The difference between the estimates for the circle and rectangle or square increased with reduced exposure time due mainly to decrement for the angular forms. The discussion relates the over-all result to an anchor effect of tilt-rod position, and the trend difference for form to differential salience of projective outline distortion and convergence.