The effect of a pictorial illusion on closed-loop and open-loop prehension.

It has been proposed that movements to visible and remembered targets are sensitive to qualitatively different types of visual information. When the target is continuously visible, prehensile movements are thought to reflect veridical object size, whereas memory-dependent prehension is sensitive to the perceived size of the object. This hypothesis was explored by assessing the influence of illusory target width on prehension kinematics in three visual conditions: closed-loop (CL; full vision during the response), open-loop brief-delay (OL; visual occlusion coincident with the movement initiation cue) and open-loop 3-s delay (OL3; visual occlusion 3 s prior to movement initiation). To modulate illusory target width, objects were placed on backgrounds consisting of three forms of the Müller-Lyer (ML) figure. Peak grip aperture was sensitive to the ML figure in the OL and OL3, but not CL conditions, suggesting that perceptual information is used to modulate this grasping parameter when the movement is programmed and executed on the basis of visual memory. Peak-aperture velocity was affected by the ML illusion in all three visual conditions, suggesting that perceived object size might be important for modulating this aspect of prehension, independent of memory requirements. The different sensitivity of grip aperture and aperture velocity to illusory target width in the CL condition suggests that grasp preshaping might reflect multiple visuomotor processes. The results of this study are consistent with the tenets of the two-stream model of visual processing.