Decade of the Brain
[Proclamation] [Activities] [Publications & Tapes] [Other Links]
[Decade of Brain Logo]
Discovering Our Selves: The Science of Emotion
Executive Summary

Panel: The Affect of Emotions:
Laying the Groundwork in Childhood

"Developing Self and Emotion
in Extreme Social Deprivation"

Mary (Maya) Carlson, Ph.D., is Associate Professor of Neuroscience in Psychiatry at the Harvard Medical School, Associate Professor of Population and International Health at the Harvard School of Public Health, and a Bunting Science Fellow at Radcliffe/Harvard.

A number of investigators have described the serious developmental deficits suffered by infants raised in hospitals and other institutions. Research with nonhuman primates, which underpins my recent studies of infants raised in orphanages in Romania, illustrates the importance of social interaction in the development of emotional regulation and in the development of the sense of self.

In the early 1990s, evidence began accumulating that sensory stimulation plays a crucial role not only in the early development of the brain's sensory- and motor- and memory-processing circuitry but also in the development of the neural and hormonal systems that regulate our response to stress.

  • Infant monkeys deprived of maternal touch develop the behaviors often referred to in human beings as autistic.

  • Infant rats handled frequently by human experimenters or groomed by their own mothers secrete less stress hormone when restrained compared with nonhandled infant rats; they also show less loss of cells in the hippocampal cortex and less loss of memory associated with that hippocampal cortex as they age.

    Production of the stress hormone cortisol, which can be measured noninvasively by sampling saliva, follows a diurnal pattern. Normally, blood levels of cortisol peak just before we awake in the morning, then drop abruptly and remain almost undetectable for the rest of the day and into the night, rising again just before waking. The cortisol is regulated through a "negative feedback circuit." As blood levels rise under conditions of acute stress, they inhibit the system, reducing or shutting off further production. Touch is critical to the establishment of this mechanism for maintaining the body's homeostatic balance.

    In examining the children in Romania, we worked with two institutional groups and one group of family-reared children, looking specifically at the diurnal patterns of cortisol secretion. Thirty children, the control group, were raised in the standard conditions of the orphanage meaning, among other things, that one caregiver looked after as many as twenty infants. Many of them exhibited behaviors, such as social withdrawal and repetitive movement, that resembled autistic behavior. Another 28 children were assigned to an enriched intervention group in which each caregiver looked after only four children; these caregivers were trained in early child education techniques.

    Children in the control group received adequate custodial and medical care, but they were not referred to by name, and they were frequently changed from room to room and from caretaker to caretaker and group to group. The children in the intervention group were called by name and remained in stable groupings, with the same caregiver.

    We found that from the average age of about 6 to 9 months of age, the intervention group grew more physically and progressed more in mental and motor abilities than the control group. Unfortunately, by the time we measured cortisol, funding for intervention had been discontinued, and the children had been back in standard care for about six months.

    When we compared the levels of cortisol in the two institutional groups to that of family-reared children, we saw that both institutional groups had abnormally suppressed cortisol levels in the morning. When children's cortisol levels at each time of day were correlated with their behavioral development, we found that those with the highest morning levels showed the best performance on motor and mental tests.

    At later times of day, when cortisol should be low or almost undetectable, the most deprived children--the control group--showed significantly more elevated cortisol levels at noon compared to the intervention group. High cortisol levels at noon and in the evening in both of the institutional groups were associated with poor mental and motor performance. Cortisol levels measured by parents at home on weekends were similar to those of American children.

    The most surprising and disturbing finding came from measuring the cortisol levels of family-reared children while they were in the bleak and highly structured context of their state-run day care. We found abnormalities similar to those seen in orphanages, and again those with abnormally elevated levels in day care were the ones who had the lowest behavioral scores.

    These findings, plus studies of children who experienced the Armenian earthquake and adults suffering from post-traumatic stress disorder, offer evidence of the relationship between abnormal cortisol secretion, hippocampal shrinkage, and related memory deficits.

    Clearly, the social networks in which children grow up bear on the development of the neural networks that mediate memory, emotion, and self. As parents, as citizens, and as policy makers, we must ensure that our social policy pathways do not lead to social deprivation. We perhaps cannot prevent a child from experiencing an earthquake, but we can ensure immediate and appropriate intervention when and if one occurs. And when we are faced not with a natural disaster but with choices about children in day care and residential care, we must ensure that our decisions have positive consequences.


  • Go to:
    If you have questions or comments on the LC/NIMH Decade of the Brain Project, please contact: nimhinfo@nih.gov.

    Library of Congress
    Library of Congress Help Desk (01/03/2000)