Social Cognition, Vol. 28, No. 1, 2010, pp. 59–73 59 This research was supported by a European Union funded project aimed at understanding the evolutionary, developmental, and intentional control of imitation (EDICI-12929). Correspondence concerning this article should be addressed to Roman Liepelt, Max Planck Institute for Human Cognitive and Brain Sciences, Department of Psychology, Stephanstr. 1A, 04103 Leipzig, Germany. E-mail: liepelt@cbs.mpg.de. LIEPELT AND BRASS PRIMING PHYSICALLY IMPOSSIBLE MOVEMENTS AUTOMATIC IMITATION OF PHYSICALLY IMPOSSIBLE MOVEMENTS Roman Liepelt Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany Marcel Brass Ghent University, Ghent, Belgium Motor priming refers to the direct matching of an observed action onto the observer’s motor repertoire (Iacoboni et al., 1999) leading to a ten- dency to automatically reproduce the action. Recent research has shown diminished automatic imitation when observing nonbiological agents, bio- mechanically impossible actions, and non-intentionally produced actions. However, the question of whether automatic imitation also occurs for phys- ically impossible actions remains open. We found motor priming effects of the same size for both physically possible and impossible movements in a choice-reaction task paradigm (Experiment 1). Both physically possible and impossible movements also elicited identical motor priming effects when attention was drawn to the difference between possible and impos- sible movements (Experiment 2). While previous research clearly showed a sensitivity of the automatic imitation system to biological plausibility and attributed intentionality, the present indings show its insensitivity to physi- cal plausibility, a inding that remains unaffected by top-down inluences. From personal experience as well as from systematic experimental research in so- cial psychology we know that humans often tend to end up in a posture similar to that of a social interaction partner (Lakin & Chartrand, 2003). Such phenomena reflect a tendency to reproduce observed actions completely unintentionally and automatically. It is now widely assumed that automatic imitative response ten- dencies result from a more or less direct mapping of the observed behavior onto the observer’s motor repertoire (Iacoboni et al., 1999). To investigate automatic imitation of others’ actions under controlled experimental conditions, a range of motor priming paradigms have been devised (see Blakemore & Frith, 2005; Brass & Heyes, 2005). Typically, participants have to execute preinstructed finger move-