["Journal of Child Psychology and Psychiatry, EarlyView. ", "\n\nBackground\nCallous–unemotional (CU) traits identify youth with more severe and chronic trajectories of conduct problems. However, the etiology of CU traits may be heterogeneous, undermining the search for effective treatments. The level of co‐occurring anxiety has been used to identify “primary” (lower anxiety) versus “secondary” (higher anxiety) variants of CU traits. The primary variant has been hypothesized to emerge from strong genetic influence and secondary variants as an adaptation to adversity, such as exposure to childhood maltreatment. However, little research has tested this hypothesis directly.\n\n\nMethods\nWe examined whether anxiety moderates the etiology of CU traits to determine whether this phenotypic feature can help distinguish CU traits with stronger genetic or environmental risk. In two population‐based twin cohorts (initial sample: N = 1,196, aged 6–11, oversampled for exposure to neighborhood disadvantage; follow‐up sample: N = 13,486, age 7), we used genotype‐by‐environment interaction twin modeling to examine if parent‐reported child anxiety moderated the etiology of concurrent parent‐reported child CU traits.\n\n\nResults\nAnxiety moderated the etiology of CU traits across both samples, such that nonshared environmental influences increased as anxiety increased. Additionally, in the larger sample, genetic influences decreased with increasing anxiety.\n\n\nConclusions\nThese findings support theories suggesting that co‐occurring anxiety may distinguish CU traits with different origins: CU traits with higher anxiety appear more influenced by nonshared environmental factors—potentially including adversity—and may show weaker genetic influence. Assessing for co‐occurring child anxiety is likely important for diagnosing and personalizing treatments among children with CU traits.\n\n"]