Consistent with prior work, we found evidence of threat-related amygdala habituation, whereby amygdala activation decreased across blocks (Figure 3a). A Genotype × ELS interaction predicted threat-related habituation in the right, but not left, basolateral amygdala (Right: Table S7; Figure 3b; bGxE = 0.77, 95% CI [0.01, 1.53], ΔR2 = .013, ΔF(1,299) = 3.97, p = .047; Left: bGxE = -0.05, 95% CI [-0.70, 1.70], ΔR2 < .001, ΔF(1, 299) = 0.01, p = 0.915). A allele carriers, who were protected against the effects of CA on cannabis dependence symptoms within CATS and SAGE, displayed heightened right basolateral amygdala habituation in the context of increased ELS (bE = 0.82, 95% CI [0.14, 1.51], p = .018); G homozygotes, who were vulnerable to cannabis dependence in the context of childhood abuse in the prior two samples, did not exhibit a relationship between ELS and amygdala habituation (bE = -0.20, 95% CI [-0.28, 0.40], p = .744). Post-hoc Johnson-Neyman tests revealed that the association between genotype and amygdala habituation was significant at mean-centered log-transformed CTQ values of -0.06 (i.e., raw score of 26.68,
