The major strengths of the current study are the use of a large, mixed patient and non-patient sample with varying levels of PD traits and the use of a semi-structured interview to assess PD traits. Some limitations must be noted too. Firstly, some PD traits, like schizotypal and histrionic, were infrequent. Although the current sample can still be considered representative of a mixed patient and non-patient population where these particular traits are least common and residuals of the PD traits-aggression correlations are normally distributed, extension to in a subsample selected higher levels of schizotypal, schizoid and histrionic PD traits is required. Secondly, reactive and proactive aggression were only assessed using self-report; aggression might be particularly prone to underreporting due to its social undesirability. It would be important to assess the PD trait–aggression relationship using behavioral reactive and proactive aggression measures, like the Taylor Aggression Paradigm