Figure 2.

Differential effects of amygdala reactivity to threat on later AB among European-American versus African-American males who showed high AB in adolescence

***p<.001, *p<.05. N=242. Models controlled for income and maternal education. A. For African-American males who had high adolescent AB, there was a signficant inverse relationship between amygdala reactivity to fearful faces at age 20 and antisocial behavior at ages 22–23 (B= -2.79, SE=.80, β=.67, p<.001). The region of significance was for values of amygdala reactivity up to .15 and higher thant .75 (shown in grey shading). B. For European-American males who had high adolescent AB, greater amygdala reactivity to fearful faces at age 20 was related to more AB at ages 22–23 (B=2.68, SE=.55, β=.86, p<.001). For European-American males who showed late-moderate AB in adolescence, lower amygdala reactivity to fearful faces at age 20 was protective against future AB (B=-.1.04, SE=.39, β=-.54, p<.01), although the interaction of being in the late-moderate group x amygdala reactivity to fear was only significant at a trend-leve, limiting what can be inferred from this significant slope (see Supplemental Table 6).The region of significance was for values of amygdala reactivity below -.59 and above -.05 (shown in grey shading).