The Effect of the Muller-Lyer Illusion on the Planning and Control of Manual Aiming Movements

Abstract

Three experiments were conducted to investigate the two visual systems hypothesis (Milner & Goodale, 1995) and the planning-control hypothesis (Glover, 2002). Experiment 1 required the participants to make rapid aiming movements to 25 cm and 35 cm tails-in, no tails, or tails-out Müller-Lyer stimuli following a 0 ms or 5000 ms no-vision delay. In Experiment 2, the participants executed their movements with full vision of the Müller-Lyer vertices that either remained the same or changed to a different configuration upon movement initiation. Vision was occluded either 350 ms or 450 ms after the onset of the movement. Experiment 3 was similar to Experiment 2, except the amount of visual feedback for on-line control was constrained to 200 ms, 400 ms, or 600 ms. The results of these experiments are problematic for both hypotheses. The participants exhibited a similar pattern of illusion-induced bias in both short and long delay conditions. In addition, the magnitude of the aiming bias increased as the movement unfolded (Experiment 1). Furthermore, even though participants were engaging in on-line control the illusion continued to exert its effects on aiming during the latter stages of the movement (Experiment 2). This effect was also observed when participants had sufficient time to process visual feedback in order to modify their movements (Experiment 3). Taken together, the results suggest that on-line control is biased by visual illusions.