Monoamine Oxidase A, Gender Differences, and Social Exclusion: Response to Gallardo-Pujol et al.

T hank you for the opportunity to respond to the letter by Gallardo-Pujol et al. concerning our article. We believe that they raise some important issues, but we also believe that their response is based on some misconceptions. Gallardo-Pujol et al. expressed concern that choosing a sample that includes mostly women may lead to misleading conclusions when trying to understand the relationship between monoamine oxidase A (MAOA) and antisocial behavior. They explain that this is due to the fact that 1) only women can be heterozygous for the MAOA gene; and 2) women are less likely to show antisocial behavior. We are less worried about this concern for several reasons. First, our purpose in the article was to examine aggression, not antisocial behavior as discussed by Gallardo-Pujol et al. Aggression is a broader construct than antisocial behavior and has been shown to be just as prevalent among women as it is among men (1–3). Our measure of aggression and the task that we use to assess neural reactivity tap the kinds of aggression that do not show gender differences, and thus we feel that our inclusion of women in this study is appropriate. (Moreover, our sample is 59% female, not " mostly female " as stated by Gallardo-Pujol et al.) In addition, when we rerun the analyses looking only at individuals (men and women) who are homozygous for the MAOA gene (low-activity MAOA [MAOA-L] or high-activity MAOA [MAOA-H]), our findings do not change. We still find that MAOA-H individuals show significantly higher levels of trait aggression, higher levels of trait interpersonal sensitivity, and greater dorsal anterior cingulate cortex (dACC) responses to social rejection than MAOA-L individuals (all p's Ͻ .05). In addition, we find no gene-by-gender interactions in predicting these outcome variables (all p's Ͼ .47). The fact that analyzing the data in this manner did not change the results is not surprising, in light of the group differences apparent in Figures 1, 2, and 3 of the original manuscript. Although these findings are reassuring, they are also based on small sample sizes (n ϭ 20), and thus we did not feel it was appropriate to report them in the original article. As stated in the article, such results should be interpreted with caution until these findings are replicated in larger samples. Gallardo-Pujol et al. also expressed concern regarding our measure of trait aggression; however, there is no generally agreed …