Results about associations between God representations and well-being/mental health can be questioned because they are predominantly based on studies with self-report instruments. There are no well-validated implicit measures of God representations. Therefore we developed the Apperception Test for God Representations (ATGR). In a clinical (n = 75) and a nonclinical (n = 71) sample, we found patterns of associations of scales of the ATGR and of an explicit God representation measure with implicit and explicit measures of object-relational functioning that undergirded the validity of most ATGR scales. Differences in patterns of associations between patients and nonpatients could theoretically be explained by the concept mentalization.