In conclusion, findings from adoption studies should probably be generalized cautiously to other populations as adoptees tend to show higher scores on aggressive behavior compared to controls. On the other hand, most of the assumptions of the classical twin design seem to hold for aggressive behavior. The EEA has been tested and found to hold for various phenotypes including aggressive behavior, and twins and singletons have been found to display similar scores on aggressive behavior, suggesting that findings from twin studies can be generalized to other populations. Most twin studies report finding little or no shared environmental influences on aggressive behavior, suggesting that random mating is of little importance for aggressive behavior. Only a few studies have examined the influence of G × E correlation on aggressive behavior, suggesting that more research is needed on this topic before we can draw any firm conclusions. Last, in the classical twin design, it is assumed that genetic and environmental influences combine additively and do not interact. This assumption is probably violated to some extent when it comes to aggressive behavior, as several
