Psychopathy and metacognitive difficulties have been associated with the occurrence of violence in schizophrenia. However, evidence suggests that nonschizophrenic psychopaths match or even outperform healthy controls on tests of metacognition. We hypothesize that the metacognitive difficulties associated with schizophrenia may be ameliorated by comorbid psychopathy. To this end, metacognition (using the Metacognition Assessment Scale-Abbreviated [MAS-A]) and psychopathy (using the Hare Psychopathy Checklist-Revised [PCL-R]) are assessed in 79 patients with schizophrenia with a history of criminal offending. Piecewise regression reveals that the association between metacognition and psychopathy changes from a negative to a positive association at a breakpoint corresponding to a score of 24 on the PCL-R. This score is within the range of the cut-off point used for the diagnosis of psychopathy in Europe, which corresponds to a score of 26 on the PCL-R. The patients scoring above 24 on the PCL-R demonstrated better overall metacognitive abilities, suggesting that these patients constitute a specific group in which schizophrenia has an attenuated impairing effect on metacognition. However, this effect was absent for the Mastery subscale of the MAS-A, which refers to the ability to use one's own mental states to solve social and psychological dilemmas. Our results suggest that the relative preservation of metacognitive abilities in schizophrenic patients with extreme levels of psychopathy may contribute to their offending behavior as it may enable them to understand how to manipulate and extort their victims. However, enhancing the Mastery domain of metacognition in these patients may attenuate this offending behavior.