Psychologists have shown that recall of sentences describing previously performed actions is enhanced compared to recall of heard-only action-phrases (enactment effect). One interpretation of this effect argues that subjects benefit from a multi-modal encoding where movement plays a major role. In line with this motor account, it is conceivable that the beneficial effect of enactment might rely, at least in part, on procedural learning, thus tapping more directly implicit memory functions. Neuropsychological observations support this hypothesis, as shown by the fact that the enactment effect is quite insensitive to perturbations affecting declarative memories. i.e. Alzheimer disease. Memory for subject performed tasks in patients with Korsakoff syndrome. The present study attempts to evaluate whether pure motor activity is sufficient to guarantee the described memory facilitation or alternatively, whether first-person experience in carrying out the action (i.e. true enactment) would be required. To this purpose, in a first experiment on healthy subjects, we tested whether sentence meaning and content of the executed action should match in order to produce facilitation in recall of enacted action-phrases. In a second experiment, we explored whether the enactment effect is present in patients suffering from psychiatric disorders supposed to spare procedural memory but to alter action awareness (e.g. schizophrenia). We show that better recall for action phrases is found only when the motor component is a true enactment of verbal material. Moreover, this effect is nearly lost in schizophrenia. This latter result, on the one hand, queries the automatic/implicit nature of the enactment effect and supports the role of the experience of having performed the action in the first-person. On the other hand, it questions the nature of the memory impairments detected in schizophrenia.