From Neurons to Neighborhoods: The Science of Early Childhood Development had a profound influence on fostering the integration of the science of early childhood development as well as promoting the use of science in early childhood interventions. In many areas the research frontier has progressed far beyond what it was when the report was released. New tools, new concepts, and even new fields of study (such as genomics, epigenetics, and functional neuroimaging) have greatly increased the ability to realize the vision described in From Neurons to Neighborhoods.
At the workshop to commemorate the 10th anniversary of the report, three speakers examined specific topics within the diverse array of research areas that constitute the overall science of early childhood development. Alan Guttmacher, Director of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), discussed the interdependence of biology and experience in the developing child. Bruce McEwen, Alfred E. Mirsky Professor at Rockefeller University, looked at the exciting work on the biological and neurological consequences of stress. And Deborah Stipek, James Quillen Dean and Professor of Education at Stanford University, discussed emerging work associated with executive functioning in the lives of young children. These three presentations focused on specific areas of research that offer future promise for the continued vitality and promise of developmental science.
Eunice Kenney Shriver National Institute of
Child Health and Human Development
One of the most influential conclusions of From Neurons to Neighborhoods is that it is not nature or nurture but nature and nurture that matter in early childhood development, said Alan Guttmacher, Director of NICHD. “The acceptance of that today compared to a decade ago is much more pervasive, and that is an important thing.”
The acceptance of this thesis has created new opportunities to investigate the complex interactions between nature and nurture that affect early childhood. As an example, Guttmacher cited the National Longitudinal Study of Adolescent Health, which combines longitudinal data on adolescents’ social, economic, psychological, and physical well-being with data on family, neighborhood, community, school, friendships, peer groups, and romantic relationships to study how such factors affect health and achievement. He also cited the National Children’s Study, which will allow researchers to examine multiple effects of environmental influences and biological factors on the health and development of approximately 100,000 children across the United States from before birth to age 21.
New Research Tools
Recent advances have given researchers new tools to examine early childhood development. In particular, Guttmacher focused on recent advances in genetics, although from an unexpected perspective. “I come here as a certified genomicist to tell you that the real thing we have to focus on is the environment. That’s because we have made great strides in the last decade in having tools to look at genetic [influences]. We have not made similar kind of strides in understanding environmental influences.”
The sequencing of the human genome has made it possible to identify genes involved in a wide variety of human diseases. For example, new tools and data have made possible investigations known as genome-wide association studies, which began to find genetic variants associated with specific diseases in 2005. Since then, genome-wide association studies have uncovered hundreds of genetic regions involved in human diseases. Guttmacher acknowledged that relatively few of the genetic regions identified so far
1 This section of the chapter is based on the presentation by Alan Guttmacher titled “The Federal Policy Perspective” at From Neurons to Neighborhoods Anniversary: Ten Years Later.
appear to be involved in human behavior or cognition, but many of the diseases associated with these genetic variants are relevant to early childhood.
However, uncovering the environmental factors that contribute to human diseases—much less human behaviors—is much more difficult. Furthermore, changing outcomes for children will involve changing their environments, not their genes. “To understand this complex interaction, we need to understand environment and genetics,” he said.
Geneticists have always assumed that genes influence behavior, but a growing body of evidence indicates that behavior influences genes in ways that were not anticipated. In particular, experiences are able to change genetic activity that once was assumed to be hard-wired. The new field of epigenetics, for example, is examining how experiences act on the configuration and modifications of the DNA molecule to affect the activities of genes. These are “examples of the kinds of things we should be thinking about,” said Guttmacher.
An important focus of research will be to understand individual variation in learning. In the past, geneticists have tended to divide people into categories, but everyone is biologically unique in terms of genome and experiences. This research will include the use of new neuroimaging techniques to explore variation in learning, longitudinal studies of learning that look at the interaction of genetic variation and sociocultural influences, examination of how or if early learning experiences modify the child’s genome through epigenetic modification, and the application of new research knowledge to improve early interventions for individuals with learning disabilities.
Another important research focus will be neural plasticity—how neuronal structure and function change in response to experiences. For example, how are neural circuits reconfigured as a result of experience? What are the environmental experiences necessary for normal or optimal development in various sensitive periods of neurocognitive development? How can the rehabilitation and adaptation of function contribute to recovery from disease or injury?
The Research Agenda at NICHD
NICHD launched a year-long process to identify scientific opportunities over the next decade across the institute’s mission, which includes pediatric health, maternal health, rehabilitation medicine, and many other topics. The aim, said Guttmacher, is to develop a scientific vision that sets an ambitious agenda and inspires the institute, the research community, and the institute’s partners to achieve critical scientific goals and meet pressing public health needs.
NICHD will hold workshops to gather input from external experts,
commission white papers for workshops to create a foundation for the vision, convene a large multidisciplinary meeting to shape emerging plans, and welcome public input throughout the process. The target date for publishing the vision is December 2011. (For more information, see http://www.nichd.nih.gov/vision.)
The institute’s staff has identified nine scientific themes to be explored in workshops:
• Developmental origins of health and disease
• Diagnostics and therapeutics
• Pregnancy and pregnancy outcomes
At least seven of these themes apply directly to early childhood development, said Guttmacher.
In addition, a number of topics cut across all of these themes:
• Analytical and measurement tools and methods
• Animal and computational models
• Biotechnologies and bioengineering, including high-throughput, assistive, and related technologies
• Developmental trajectories
• Differences and disparities across populations
• Epigenetics and metagenomics
• Functional status
• Global health
• Implementation science, including health economics
• Preventive and personalized medicine
• Stem cells
• Systems biology
• Training and mentoring
The intention is to produce a research agenda that is useful not only to NICHD but also to many other organizations that are interested in early childhood development, said Guttmacher. In this way, research will be able
to inform policy through investigations of the interactions between biology and experience in early childhood. The overall result will be to improve children’s outcomes through the application of new tools and approaches.
During the discussion period, Guttmacher was asked how NICHD can overcome the disciplinary barriers that impede needed research on early childhood development. He replied that the program review currently under way at the institute is organized around broad themes, not around existing disciplines, to conceptualize the research agenda in a broader way. Disciplinary boundaries are a problem for the National Institutes of Health (NIH) in general, he noted. “No matter what area of health and development one is talking about at NIH, [disciplinary] silos don’t tend to correlate with biology in the real world.” The challenge for transdisciplinary research is to overcome these silos both scientifically and programmatically. The program staff at NIH take great pride in their grantees and the fields they support. But this pride also can breed narrow thinking. “In academia we tend to identify with certain systems and certain programs,” he said. “We need to think more creatively. For instance, if you think about the environmental-genetics construct, the only way to do that research is if we have people who are world-class experts in both of those worlds.”
Training is one way to resolve this problem. “I would hope that we would train folks who are comfortable across this broad swath of knowledge,” he said. “Right now we don’t have those people. We need more creative ways of bringing teams together.” He noted that a historical strength of the IOM and the NRC is their ability to bring together people with very different perspectives—as the title From Neurons to Neighborhoods suggests. More of that kind of audacity is needed in the scientific community.
NICHD also needs to hear from and listen to the users of its research, Guttmacher said. This input relates both to the scientific opportunities and to the public health impact of research. A sense of the key questions that can be answered would be extremely useful to NICHD. Partnerships between agencies involved in policies and programs also can yield such input. Such partnerships can lead to experiments that reveal whether novel interventions are having a desired effect, so long as these effects are accurately measured.
People’s biological reactions to stress are a good example of the interplay between biology and the environment, said Bruce McEwen, Alfred E. Mirsky Professor at Rockefeller University. Roughly speaking, stress can be divided into three categories. Positive stress can be generated by a personal challenge, in which satisfactorily meeting that challenge gives rise to a sense of mastery and control. Positive stress points toward the existence of a healthy brain architecture, self-esteem, good judgment, and impulse control.
Tolerable stress results from the occurrence of adverse life events that are buffered by supportive relationships. The result of such stress is coping and recovery, again suggesting the presence of healthy brain architecture, self-esteem, good judgment, and impulse control along with good external supports.
Toxic stress results from unbuffered adverse events that are extended and/or sizable. The result is poor coping and compromised recovery. For children, the consequences of toxic stress include lifelong risk for physical and mental disorders, which is likely to be due to compromised brain architecture and dysregulated physiological systems.
The brain is the central organ in the stress response, including adaptation. The brain perceives and judges what is threatening and therefore stressful. In this way, the brain determines the subsequent responses to stress, including adaptation. The term “allostasis” refers to the physiological process of producing hormones in the body that produce a reaction to stress. The term “allostatic load” refers to the cumulative burden on the body as a result of stress and harmful health behaviors resulting from a stressful life.
There are many types of stressors, including trauma, abuse, major life events, family conflict, workplace environments, and community factors. In addition, each person has a unique response to stress determined by genetics, early developmental history, and experiences throughout life.
Stress causes the release of hormones in the brain that travel to other parts of the body, which in turn cause subsequent hormonal responses and other physiological reactions. In some cases, these hormones have beneficial effects. For example, in the short term, these hormones can
2 This section of the chapter is based on the presentation by Bruce McEwen titled “Implications for Physical and Mental Health” at From Neurons to Neighborhoods Anniversary: Ten Years Later.
enhance immune responses, memory, energy replenishment, and cardiovascular function. But if stress is prolonged or severe, these same hormones can cause harmful inflammatory and immune responses, impair memory, elevate mineral loss from bones and muscle wasting, and contribute to metabolic syndrome. In addition, other physiological reactions can disrupt brain, metabolic, immune, and cardiovascular function, creating a complex and nonlinear network of stress responses.
The brain also governs behavioral responses. For example, it can react to stress through fighting or fleeing. It can motivate personal behaviors that cause adverse outcomes like eating too much, drinking too much, not getting enough sleep, or not engaging in health-promoting behaviors like regular physical activity. The behavioral responses encompassed by the term “lifestyle” are also important, because they can influence the allostatic load (see Figure 2-1).
Finally, toxic chemicals in the environment can feed into the network of allostasis. For example, toxicants in the environment can increase inflammatory responses, which in turn can increase hormonal secretion and create imbalances in the allostatic network.
Diseases exacerbated by the allostatic network include cardiovascular disease, depression, diabetes, Alzheimer’s disease, arthritis, and cancer. All are diseases of modern life, McEwen pointed out, and many others could be included.
FIGURE 2-1 The brain plays a central role in mediating between a person’s experiences and physiological reactions, including the stress reaction.
SOURCE: Reprinted from McEwen, B. S. 1998. Protective and damaging effects of stress mediators. New England Journal of Medicine 338(3):171-179. Copyright 1998 with permission from Bruce McEwen.
Stress and Brain Plasticity
The brain is capable of considerable remodeling, even in adults, McEwen observed. For example, the dendrites that receive electrical impulses from other nerve cells and then help pass them on to others can grow or shrink throughout life. The synapses through which nerve cells communicate with each other also can change as a result of experience. The brain can even form new neurons in particular brain regions such as the hippocampus, which plays a key role in certain kinds of memory. In fact, particular kinds of stress suppress the generation of new neurons, and the regulation of this process is thought to be a key factor in depressive illnesses.
Research on animals and, to a more limited extent, imaging techniques in humans have shown that stress can cause neurons to shrink or grow. For example, in the hippocampus, stress can cause neurons to shrink, so that dendritic networks become simplified. But in a part of the region called the basal lateral amygdala, which is critical in memories of fear, stress causes an expansion of the dendritic tree. Similar processes can occur in a cortical area called the orbital frontal cortex, which is involved in determining the salience of reward or punishment, and in the medial prefrontal cortex, which has important roles in decision making, working memory, and top-down control of mood, anxiety, and autonomic and neuroendocrine functions. “What you have is growth in some areas, shrinkage in some areas,” said McEwen. “Some parts of the brain become more activated and more responsive. Others become less responsive.”
The good news is that animal models have produced some evidence that the brain can recover when the stress ends, so “this is not an example of brain damage per se.”
The Origins of Adult Diseases
Many adult diseases have their origins in adverse events early in a child’s life (Shonkoff et al., 2009). Studies have shown, for example, that living in a chaotic home can lead to greater helplessness and distress, poor self-regulatory behaviors, and perhaps deficits in prefrontal cortical function. Other studies have documented elevated levels of obesity, blood pressure, cardiovascular reactivity, and systematic inflammation as well as poor dental health as a result of chaotic home situations in early childhood, leading to shorter average lifespans (Brown et al., 2009). Adverse childhood experiences can exacerbate obesity, illicit drug use, mental health problems, sexual and reproductive health issues, and general health and social problems. Low socioeconomic status is associated with poor language skills, poor executive function, and other effects on learning ability. Maltreatment of children has a cascade of consequences that influence cognitive and in-
CASCADE OF CONSEQUENCES
FIGURE 2-2 The experiences that a child has as part of a family and a community have a cascade of consequences for biological and neurological functioning.
SOURCE: McEwen, 2010.
tellectual ability, social and behavioral skills, psychological and emotional processes, and physical and neurological functions (see Figure 2-2).
Animal models have played a large role in studying the physiological effects of the environment. For example, in rodent models prenatal stress retards the development of the hippocampus, whereas good maternal care improves the emotional state and even seems to affect the lifespan of offspring. With a background of good maternal care, novel experiences can benefit cognitive and social development in rodent models. Similarly, studies of rhesus monkeys have shown that maternal anxiety can be transmitted to offspring. And animal research is showing that the intrauterine environment has immense importance not only for things like metabolism and obesity but also for behavior.
The new field of epigenetics is looking at gene–environment interactions that have physiological and behavioral consequences, including the effects of stress. Epigenetic modifications to DNA include the actions of proteins called histones that determine the folding or unfolding of DNA to expose or hide genes, the binding of protein regulators to DNA and the addition of
methyl groups to DNA that influence gene expression, and the production of small RNA molecules that regulate the survival and translation of messenger RNAs. Some of these effects can be passed from one generation to the next and influence such traits as obesity and parental behavior. McEwen observed: “To reduce the obesity epidemic, for example—which by the way has effects on cognitive function as well as other diseases—we have to deal with nutrition, stress, and other things in families expecting children. We can’t just start after the child is born.”
In these ways, the experiences people have during their life become embedded in their brains and in their physiological reactions to events. Heavy allostatic loads, for example, can alter and sensitize the brain and the body to respond in certain ways to similar experiences. Experiences involving the stress response also can have positive consequences. For example, regular physical activity demonstrably improves executive function. “We all know that getting up and walking around helps to jog our cognitive function. That is a very simple and direct solution to help a lot of people help themselves, and particularly kids.”
The centrality of the stress response means that interventions to decrease stress and the harmful effects of stress can produce major benefits for society. For example, interventions influencing early childhood development can produce major financial returns through increased earnings to individuals, lower crime rates, savings on special education and welfare, and increased income tax revenues (Center on the Developing Child at Harvard University, 2010a). “The returns on investment … are significant and important,” said McEwen, “besides the fact that it is the right thing to do.”
During the discussion session, McEwen pointed out that top-down policy directives need attention, in addition to bottom-up interventions. Government policies and work-related policies of private enterprise that affect how families and children live are essentially health policies. Therefore, policies and health interventions need to work in tandem. “Societal change takes a long time and requires a political process,” he said. Specific programs can be the consequence of overarching policies, he said, citing the Nurse Family Partnership, the Perry Preschool Project, and the Abecedarian Project. A combined approach “is the most important way of dealing with things,” he continued.
Policy makers and the public need to know how the brain is involved in early learning and development. In particular, McEwen mentioned that the Society for Neuroscience is organizing a public outreach effort on early childhood development to capture the attention of the public and policy makers.
From Neurons to Neighborhoods provided incontrovertible evidence for several key conclusions, said Deborah Stipek, James Quillen Dean and Professor of Education at Stanford University. It showed that children’s experiences shape the architecture of their brains as well as their functioning. It also demonstrated the value of early interventions, including high-quality early childhood education.
These conclusions inevitably raise questions about leverage points in the lives of young children. An especially important question is, What domains of early skills can be used to enhance later skill development? The task is complicated by the interdependence of skills. For example, when children who are less prepared enter school, they lack not just academic skills. They also tend to be behind other children in social, emotional, and cognitive skills. These dimensions are highly interactive, Stipek observed, and each contains many subdimensions.
The question then becomes whether to intervene along all of these dimensions or to direct attention to particular dimensions that leverage other kinds of skills. “How [can] a skill that you develop at the age of 3 contribute to the kinds of academic demands that you are expected to meet at the age of 5 or 6 or 8 or 11?” she asked.
Stipek argued that interventions in particular domains can act as particularly effective leverage points, noting that “Piaget taught us that cognitive development is a system of understandings, and that changes in any part of that system are likely to influence changes in other parts of the system.” As examples of potentially important leverage points, she discussed verbal skills, social skills, mathematical skills, and the executive functions that underlie many developing skills in young children.
The verbal input that children receive—both in terms of the number of words and the number of different words they hear—has a strong correlation with their vocabulary, Stipek observed. Verbal learning also exhibits a reciprocal loop. As children develop better vocabulary and literacy skills, their executive functions are enhanced. For example, recent research has
3 This section of the chapter is based on the presentation by Deborah Stipek titled “Implications for Motivation and Learning” at From Neurons to Neighborhoods Anniversary: Ten Years Later.
shown that the verbal environment of children even in infancy predicts their working memory, their processing speed, and their underlying cognitive skills, which in turn predict their vocabulary development. In addition, the levels of skill children have when they enter school influence the kinds of educational environments they experience. These skill levels influence teachers’ expectations for the kind of work they can do and help determine which other students they are grouped with in classrooms.
Because of these effects, verbal input has a multiplicative influence on children’s lives. “I would propose that the verbal environment might be one of those leverage points that we need to take a look at,” said Stipek.
Social skills also predict learning. Children’s ability to get along with peers, their skills at interacting socially, and the absence of aggressive or disruptive behavior are believed to facilitate their learning. Similarly, the relationships that children form are important predictors of how well they learn, and they learn better when they have a secure and caring adult available to them.
Research in the past decade has begun to explore these connections more thoroughly. For example, increased aggression has been shown in children who are having difficulty learning to read. In this situation, learning predicts social skills better than social skills predict learning. This does not imply that social skills are unimportant, Stipek emphasized, but the common assumption that social skills cause academic skills needs further analysis. The research suggests that improving academic skills may serve as a lever for promoting positive social behavior.
The development of mathematical skills has been part of childhood since antiquity, but these skills are relatively new to preschool. Stipek and her colleagues have had to visit preschools multiple times to find any evidence of mathematical activities. “It is something that has been neglected,” she said, “but recent evidence suggests that it may deserve more attention than we have been giving it.” A recent report by the National Research Council (2009) highlighted research in this area.
A recent study compared literacy and mathematics skills for children entering kindergarten to predict their skills in reading and mathematics in third grade. The mathematical entry skills were much stronger predictors of both mathematical and reading skills in third grade than were reading entry skills. The reasons for this correlation remain unclear. Does learning in mathematics facilitate later thinking skills and the ability to perform in
school, or do early mathematical skills reflect some underlying executive functions that influence learning? “We [don’t] really know,” said Stipek, “but data like this suggest we better find out … because mathematics might be much more important than we have thought.”
Research also has demonstrated that young children are more capable of learning mathematics than was previously believed. They can do much more than just “macaroni math”—gluing three macaronis in one box and five in another. They can learn to use numbers to describe order and measure, to use patterns to recognize relationships, and to use geometry to represent objects. “There is a lot to mathematics even for young children,” Stipek observed.
The natural progression of mathematical understanding is fairly well understood, she said, as are effective strategies to teach mathematics to young children, and research-based preschool mathematics curricula exist. Stipek observed, “It is not that we don’t have the tools, but we have not been convinced of the value of mathematics.” If this is a productive leverage point, as current research suggests it might be, teachers and caregivers will need to develop skills to help children develop fundamental mathematical skills.
As a final example of a potential leverage point, Stipek spoke more broadly about executive functions (see Figure 2-3). Research has shown that executive functions are important for social skills as well as learning. For example, studies that hold constant aspects of executive functions, such as attention and impulsivity, no longer detect a relationship between aggression and academic performance, which suggests that executive functions may be critical in both domains. In social interactions, children need to be able to remember the rules of the game. They need to inhibit the impulse to push a child out of the way when they run for a ball or to grab a toy that they want. They need to plan strategies for entering play or for engaging other kids in play. Inhibitory control, emotional self-regulation, memory, attention—which are all executive functions—are all important in social interactions as well as intellectual work.
As an example, Stipek cited a classroom of preschoolers in which a teacher asks a question. Every child raises his or her hand, but when the teacher calls on someone, that child has no idea what to say. “This is an example of poor inhibitory control,” said Stipek. They want to participate so they raise their hand, but they have not thought through the fact that they do not have an answer. “We see these executive functions play out on the playground in a social context and in the classroom all the time.”
In intellectual work, children need to remember instructions. They need
FIGURE 2-3 Executive functions include both cognitive and emotional skills and attributes.
SOURCE: Stipek, 2010.
to be able to pay attention to the task at hand despite distractions. They need to be able to shift their attention if the strategy they are using does not seem to be working. They need to inhibit the immediate action that comes to mind and consider alternatives.
Researchers need to study executive functions to learn more about how they influence behavior in both the learning context and the social context, Stipek said. Today very few interventions that target executive functions are available, despite their potential value in building childhood skills.
The evidence on leverage points in the lives of young children supports the need for effective interventions. A strong foundation “influences children’s opportunities and abilities to learn from the very first year of their life,” said Stipek. The existing evidence does not necessarily indicate exactly where to intervene, but it identifies questions that deserve much more attention.
Intervening in the lives of young children is inevitably complex. Young children are learning language (and sometimes additional languages in addition to their native language), pre-literacy skills, social and emotional skills, impulse control, and many other skills. They are developing relationships with parents, caregivers, educators, and other adults. People who have major responsibility for young children need much more support and training to guide development in the face of this daunting complexity.
Several outstanding questions have important implications for the research needed to prepare for future updating of From Neurons to Neighborhoods, said Stipek. First, the interconnections between skills need to be better understood. How do interventions in one area affect skills in other domains? For example, very few researchers assessing the effects of interventions targeting either academic or social skills have assessed the effects of the intervention on the other domain. She expressed her view: “I strongly recommend that we look at these cross-domain effects. It will help us to understand the interconnections among domains of development and to identify those important leverage points.”
Researchers also need to develop a better understanding of what kinds of skills teachers and caregivers need and how to provide them with those skills. The focus on credentials such as academic degrees can detract from the important questions: What do people need to know and be able to do and how can these skills be developed?
The only way to convince policy makers of the need to invest in early childhood development is to convince the public, Stipek concluded. Advocates need to start where people’s hearts are, which is with their own children. They need to understand that “their child’s future depends as much on other people’s children’s future as on their own child’s future. And if they want their child to thrive, they need to make sure that they are promoting and supporting the welfare of the people [working with] their child,” she noted.
Stipek was asked during the discussion session how executive functions can be measured in preschoolers. She commented that for many domains of development there is a confusing array of overlapping and differently named measures. Stipek had begun compiling a list of measures of learning-related behaviors, but quit when she reached 30 measures. One of the things the field needs to do, she said, is develop greater consensus about the conceptualization and measurement of different domains of development.
Stipek also was asked how to provide incentives to draw the best minds into teaching young children. She answered that people need to be paid more. Care providers are not going to sacrifice the needs of their own
families to care for other children. “We are going to have to increase pay. That to me is a no-brainer,” she observed.
Specific interventions also bear promise. For example, a program at Stanford called Jumpstart gives students a year-long seminar in early childhood education along with 6 hours per week spent in a low-income community working with preschool children. “These bright, energetic undergraduates are totally hooked by the time they finish the year,” Stipek said, “if they are not going into early childhood education, many of them end up going into K-12 education.” However, they will not stay in early childhood education unless their pay and the respect they are accorded by the rest of society improves. In countries that compare favorably to the United States in educational performance, educators are paid well relative to other professions and education is a highly respected profession. “We don’t have that in the United States,” Stipek said. “Partly it is because people underestimate the difficulty. They think that [working with] young children is babysitting. They do not understand the many interconnected domains of development that teachers need to understand and promote, especially if we want to close the achievement gap.”