The adapted version of the Rosenzweig Picture-Frustration Study (PFS-AV) appears reliably and validly to measure hostility in violent forensic psychiatric patients. Twelve of the 24 pictures that had a good internal consistency, inter-rater reliability, and test-retest reliability were selected. In support of the instrument’s concurrent validity, scores on the PFS-AV were positively correlated with those on the aggressive behavior questionnaires but less strongly than the correlations between the aggressive behavior questionnaires mutually. The validity of the PFS-AV was demonstrated by the positive correlation between PFS-AV hostility and neuroticism, and by the negative correlation with extraversion, openness, agreeableness, and conscientiousness. A relatively low but positive correlation was found with social anxiety and a negative correlation was found with social skills in situations where approaching behavior may be exhibited. The PFS is based on the psychodynamic theory that frustration of basic needs can lead to aggression. The child and adult version of the test consists of 24 cartoon-like pictures. According to Rosenzweig’s frustration-aggression theory, scores on 11 response types are then combined to form 6 categories of aggression: 3 directions of aggression and 3 types of aggression. This paper describes the development of an adapted version of the PFS (PFS-AV) for the measurement of hostility in violent forensic psychiatric patients. Tables, notes and references