Everyone has the feeling that perception is usually accurate-we apprehend the layout of the world without significant error, and therefore we can interact with it effectively. Several lines of experimentation, however, show that perceived layout is seldom accurate enough to account for the success of visually guided behaviour. A visual world that has more texture on one side, for example, induces a shift of the body's straight ahead to that side and a mislocalization of a small target to the opposite side. Motor interaction with the target remains accurate, however, as measured by a jab with the finger. Slopes of hills are overestimated, even while matching the slopes of the same hills with the forearm is more accurate. The discrepancy shrinks as the estimated range is reduced, until the two estimates are hardly discrepant for a segment of a slope within arm's reach. From an evolutionary standpoint, the function of perception is not to provide an accurate physical layout of the world, but to inform the planning of future behaviour. Illusions-inaccuracies in perception-are perceived as such only when they can be verified by objective means, such as measuring the slope of a hill, the range of a landmark, or the location of a target. Normally such illusions are not checked and are accepted as reality without contradiction.Most people feel that they see the world as it exists in true geometric reality, a concept known as naïve realism (Ramsperger, 1940). Contrary to this belief, however, people often fail to perceive the world as it truly exists before them, but rather perceive it in a manner that combines the geometric layout of the environment and their own potential to interact with it, distorted by a number of lessthan-reliable processing algorithms. This has been demonstrated in many perceptual studies (Bridgeman, 2004;Bridgeman, Gemmer, Forsman, & Huemer, 2000;Post, Welch, & Bridgeman, 2003) that examine various functions of the sensorimotor and cognitive systems and the use of visual information in each system. A focus of perception can be situating the body in space, rather than recognizing patterns in the abstract.Traditional approaches, such as that of traditional Gestalt psychology or the more modern spatial vision, do not consider the body as a participant in perceptual processes. Patterns are considered to have an independent existence, their important relationships being with one another rather than with the observer. Here, two empirical examples will show the importance of embodied cognition, considering the links between perception and action in the grounding of cognition.Correspondence should be addressed to Bruce Bridgeman,