Magicians often trick spectators’ mind by relying on cognitive limitations. Recently, research in psychology has studied the processes at play in magic. The goal of these studies is not only to gain new insight on known cognitive processes but also to uncover those yet unexplored. Although the range of psychological biases associated with magic is vast, most of them rely on a certain exploitation of participants’ prior expectation in order to manipulate their low-level (perceptual) and high-level (problem solving) reasoning processes. The aim of the present project is to investigate the mechanisms at play in three cognitive biases commonly manipulated by magicians and relying on participants’ prior expectations: the mind-fixing effect, motion extrapolation and attribute substitution. Studying the cognitive processes manipulated by magicians should lead to major advances in the fields of low-level and high-level reasoning. The aim of this first work package of our project is to better understand the mechanisms at play in the mind-fixing effect by studying the influence of two main factors closely linked to participants’ expectations: the credibility of the source and the level of insight linked with the activation of a false/wrong solution. The aim of the second work package is to understand the specificity of the motion extrapolation involved in a famous perceptual illusion: the Vanishing Ball Illusion (VBI). Although a considerable number of studies have investigated how individual and contextual factors (e.g., participants’ expectations, participant’s age, allocated attention…) affect participants’ motion extrapolation in the widely studied representational momentum (RM), none have yet investigated the specific influence of these factors on the VBI. However, the VBI seems to rely on mechanisms that are independent from those involved in RM tasks. The aim of the third work package is to better understand the mechanisms at play in perceptual attribute substitution error. More specifically, we aim to evaluate its degree of cognitive impenetrability, to better understand the generic aspect of this perceptual attribute substitution and to investigate the role of individual differences such as ambiguity tolerance. Improving our understanding of these low-level and high-level reasoning biases, and notably the influence of prior expectations, could be beneficial in many areas where intuitive errors can have major consequences, such as road safety or aeronautics. Moreover, this could shed theoretical new lights on these yet poorly studied biases.