The chemistry of child neglect: do oxytocin and vasopressin mediate the effects of early experience?

R esearch described in a recent issue of PNAS by Wismer Fries, Ziegler, Kurian, Jacoris, and Pollak (1) reveals significant group differences in two brain-derived neuropeptides, oxytocin and vasopressin, between familyversus orphanage-reared children. These intriguing results may stimulate a new chapter in the ongoing scientific dialogue concerning the neurobiological effects of early experience. Fascination with the developmental effects of social deprivation is not new. Consider Romulus and Remus, the legendary founders of Rome who were purportedly nursed by a shewolf; or Kasper Hauser, a mysterious 19th century German boy, believed to have been isolated from human contact from early childhood. Modern interest in the impact of maternal separation or neglect on human behavior can be traced to two proteges of Sigmund Freud, Rene Spitz and John Bowlby (2, 3), who described children reared in orphanages or by inconsistent caregivers as at risk for emotional and social disorders. Behavioral studies, including research in nonhuman primates, have documented vulnerabilities in socially deprived infants (4–6). From birth onward, social interactions are essential for normal development. In all human cultures, including nonhuman animals, we find mechanisms for transmitting social experiences from one generation to the next. These experiences may in turn alter future behavior. Attempts to study the neuroendocrine consequences of early experience have tended to focus on changes in the hypothalamic–pituitary–adrenal (HPA) axis (7, 8). However, adrenal hormones are at best indirect measures of what is happening in the nervous system. In the Fries et al. study (1), some children in both groups had oxytocin and vasopressin in the normal range, and some had low levels of these peptides. However, a significant proportion of the orphanage-reared children exhibited very low levels of peptides. It is important to note that it is too early in our understanding of human neuroendocrinology to conclude that the differences seen here were causally related to behavior. For example, it is possible that differences in oxytocin and vasopressin levels ref lect metabolic differences between children in the two groups. However, at least for oxytocin, the absence of basal differences in this study suggests that this is unlikely. In Fries et al. (1), oxytocin and vasopressin were measured in urine. In theory, these peptides might have originated in tissues other than the brain. However, most of the body’s oxytocin and vasopressin is produced in the hypothalamus. In this case, the bladder serves as a convenient and accessible reservoir for centrally produced peptides. The current finding is particularly compelling, because the previously orphaned children had lived in good homes for an average of nearly 3 years and were tested under apparently comparable conditions.