Size-contrast illusions deceive grasping as well as perception

Size contrast illusions are assumed to exert a smaller effect on human motor behavior than on perception, indicating different cortical pathways for perception and action (e.g., Aglioti, DeSouza Goodale, 1995 Current Biology 5 679-685). We tried to replicate these findings for the Ebbinghaus Illusion. Special effort was taken to minimize effects of motor learning and on assessment of the size of the perceptual illusion. An aluminum disc (28, 31, 34 or 37 mm in diameter, 5 mm in height) was positioned as target on a board. Around the target either small or large context discs were drawn (10 or 58 mm in diameter). Close to the board a monitor was mounted on which a comparison disc was displayed. In a visual task twelve subjects adjusted the size of the comparison disc to match the size of the target. In a grasping task subjects grasped the target. Subjects wore shutter glasses and could not see their hand during grasping (open loop condition). The grasp trajectory was recorded and the maximum preshape aperture was calculated. Preshape aperture and adjusted size showed strong and similar linear relationships to the size of the target. The mean perceptual effect of the illusion was 1.4 mm (SE = 0.1 mm) while the effect of the illusion on preshape aperture was 1.5 mm (SE = 0.4 mm). Thus, grasping was just as much influenced by the illusion as perception. Possible reasons for this discrepancy to previous studies are discussed