Several limitations of the present study can be noted. First our results are based primarily on use self-report data. Similar to the issue of using SEM-based approaches, multi-method approaches to the estimation of both psychopathic traits and external criteria (see Blonigen et al., 2010) are optimal for identifying the manner in which contextual processes, such as the family structure, shape the development of psychopathy over time. Second, our sample included only male twins; thus, there is a need to examine the developmental and structural patterns in mixedgender samples. Third, our examination of external correlates was limited to the family environment and externalizing behavior factors. These factors, however, represent two of the more robust correlates of psychopathic features in adolescence (Lynam et al., 2007). Fourth, our interpretations must be limited to the timeframe of late adolescence (age 17) to young adulthood (age 23). The importance of this age period aside, the present findings may not characterize the developmental patterns during the formative years of early childhood (Frick et al., 2003; Barry et al., 2008).
