1 Introduction Vision not only allows us to perceive objects `out there' in the world; it also plays a crucial role in the guidance of the actions we carry out with respect to those objects. More than 15 years ago, Goodale and Milner (1992) proposed that these two different functions of vision can be mapped onto the two prominent `streams' of visual projections that arise from primary visual cortex. According to their account, the ventral stream, projecting to the inferotemporal cortex, delivers the rich and detailed perceptual representation of the world required for cognitive operations, while the dorsal stream, projecting to the posterior parietal cortex, provides the flexible moment-to-moment control required for the programming and guidance of visually guided actions, such as grasping. Goodale and Milner (1992) argued that the two separate streams of visual process- ing evolved because perception and action require quite different transformations of the visual signals. To be able to grasp an object successfully, for example, it is essential that the brain computes the actual size of the object, and its orientation and position with respect to the observer (ie in egocentric coordinates). Moreover, the time at which these computations are performed is also critical. Observers and goal objects rarely stay in a static relationship with one another and, as a consequence, the egocentric coordinates of a target object can often change dramatically from moment to moment. For this reason, it is essential that the required coordinates for action be computed immediately before the movements are initiated. For the same reason, it would be counterproductive for these coordinates (or the resulting motor programs) to be stored in memory. In short, vision-for-action works very much in an `online' mode. The requirements of perception are quite different, both in terms of the frames of reference used to construct the percept and the time period over which that percept (or the information it provides) can be accessed. Vision-for-perception appears not to rely on computations about the absolute size of objects or their egocentric locations. Instead, the perceptual system in the ventral stream computes the size, location, shape, and orientation of an object primarily in relation to other objects and surfaces in the scene. Action rules: Why the visual control of reaching and grasping is not always influenced by perceptual illusions Perception, 2008, volume 37, pages 35 5^366 Melvyn A Goodale, Claudia L R Gonzalez, Grzegorz Kro¨ liczak CIHR Group on Action and Perception, Department of Psychology, University of Western Ontario, London, Ontario N6A 5C2, Canada; e-mail: mgoodale@uwo.ca Presented at Interdisciplinary Conference on Pre-Emptive Perception, Hanse Institute for Advanced Studies, Delmenhorst, Germany, 15 ^ 18 October 2005 Abstract. It is generally accepted that vision first evolved for the distal control of movement and that perception or `representational' vision emerged much later. Vision-for-action operates in real time and uses egocentric frames of reference and the real metrics of the world. Vision-for-perception can operate over longer time scales and is much more scene-based in its computations. These differences in the timing and metrics of the two systems have been examined in experiments that have looked at the way in which each system deals with visual illusions. Although controversial, the consensus is that actions such as grasping and reaching are often unaffected by high-level pictorial illusions, which by definition affect perception. However, recent experiments have shown that, for actions to escape the effects of such illusions, they must be highly practiced actions, preferably with the right hand, and must be directed in real time at visible targets. This latter finding suggests that some of the critical components of the encapsulated (bottom ^ up) systems that mediate the visual control of skilled reaching and grasping movements are lateralised to the left hemisphere. doi:10.1068/p5876