Since the 19th century, it has been widely acknowledged that persons with a psychotic disorder are more often involved in violent crimes than those without mental problems, which is confirmed by several recent review papers and meta-analytic studies. However, the caveats and limitations of these studies have been largely neglected. In the present systematic review paper, the epidemiological studies that were published since 1980 on the link between psychosis and violent behavior were critically reviewed. Electronic databases were searched to initially identify 5756 articles. A total of 26 articles were eventually selected as they met all selection criteria. Studies were categorized according to their research design (i.e., birth cohort, community, register/record-based). Although schizophrenia and other related psychotic disorders seem to be undoubtedly associated with violent behavior (with or's between 2 and 28), it should be kept in mind that underlying variables or risk factors (e.g., family history of violence, emotions such as anger and anxiety, impulsivity, childhood problems), the study design, and conceptual problems (i.e., definitions of violence/aggression and schizophrenia/psychosis) might be particularly relevant when interpreting the link between violence and psychosis.