We are witnessing a tremendous expansion of strategies and techniques that derive from basic and preclinical science to study the fine genetic, epigenetic, and proteomic regulation of behavior in the laboratory animal. In this endeavor, animal models of psychiatric illness are becoming the almost exclusive domain of basic researchers, with lesser involvement of clinician researchers in their conceptual design, and transfer into practice of new paradigms. From the side of human behavioral research, the growing interest in gene-environment interplay and the fostering of valid endophenotypes are among the few substantial innovations in the effort of linking common mental disorders to cutting-edge clinical research questions. We argue that it is time for cross-fertilization between these camps. In this article, we a) observe that the "translational divide" can-and should-be crossed by having investigators from both the basic and the clinical sides cowork on simpler, valid "endophenotypes" of neurodevelopmental relevance; b) emphasize the importance of unambiguous physiological readouts, more than behavioral equivalents of human symptoms/syndromes, for animal research; c) indicate and discuss how this could be fostered and implemented in a developmental framework of reference for some common anxiety disorders and ultimately lead to better animal models of human mental disorders.