Skip Navigation

Transcript: NICHD Research Perspectives—August 29, 2012

Skip sharing on social media links

Announcer:  From the Eunice Kennedy Shriver National Institute of Child Health and Human Development, part of the National Institutes of Health, welcome to another installment of NICHD Research Perspectives.  Your host is the Director of the NICHD, Dr. Alan Guttmacher.

Dr. Alan Guttmacher:  Hello, I'm Alan Guttmacher.  Thanks for joining us for another in our monthly series of podcasts from the Eunice Kennedy Shriver National Institute of Child Health and Human Development at the National Institutes of Health.

My guests today are Dr. James Griffin, Dr. James Mills, and Dr. Kathy Mann Koepke.  We will be talking about three recent studies:  the first on the effects of stress on a child's ability to learn; the second on the initial signs of fetal alcohol syndrome in infants and children; and the third on the ability to estimate quantities.

Our first guest, Dr. James A. Griffin, is deputy chief of the Child Development and Behavior Branch at NICHD.  The branch supported the study, which suggests that stress hormones can inhibit brain development and hinder achievement in young children.  The investigators spent several years matching hormone levels to behavioral and achievement test results in children.  They found evidence that the stresses of poverty may impair learning ability.  Jim, could you please tell us more about this study?

Dr. James Griffin:  Sure, Alan.  It's fitting we're talking about this, given so many children are now starting or going back to school, and 25 percent of them, approximately, are actually growing up in poverty.  So their parents may even be struggling to be able to pay for back-to-school supplies, clothing, etc.  But what we're really trying to find out with the research is how exactly poverty does affect learning in these children.  And Clancy Blair, a professor at New York University, has a theory and has been doing research on looking at how stress response actually affects learning and how all of this is related to executive function skills.  Let me say a little bit more about all that.  First of all, what is executive function?  It's part of the brain, the prefrontal cortex—it's kind of the air traffic controller.  It's about self-regulation, how you pay attention to things, how you regulate your behavior.  Now, what we see in children who grow up in poverty is that they tend to have a different stress response.  They have a tendency— rather than most children— where when something happens that causes them a little bit of stress, they have a peak in a stress hormone called cortisone, but then it rapidly goes down.  With these children, they have a different response.  It either is high and stays high, is low and stays low, or actually doesn't go down as quickly as we would expect.  Now we're finding that that is linked to these executive function skills and how basically the brain is developing in about the 3 to 5 age range.  What Dr. Blair has found in this research is when children are doing tasks, what you see is they often show this different type of stress response and they're showing lower levels of executive function skills.  And they're also not doing as well on these preschool, school readiness tasks, and that this does persist into kindergarten.  So there we're basically showing there is something going on with how the stress that they're under—growing up in poverty, all the things we associate with that including some crowding, people coming and going, maybe loud environments where it's hard to focus and concentrate on different things you're doing like reading and things of that nature—all may come into play in affecting their school readiness skills and then later achievement.  Now, he's done further research as part of the Family Life Project, which is a large research study being funded by NICHD, which started at birth, looking at children growing up in rural poverty.  Now what he was able to look at in depth in that study is how children's stress response is actually in part a function of the different parenting styles that we see.  And often what we're seeing is the parents, who are also in poverty and who are also stressed, have a different parenting style.  It tends to focus more on obedience.  It tends to focus more on compliance.  And so what we're seeing is with those who are from a slightly higher economic group, the parents are much more likely when they are working with the child of a young age to do scaffolding, where they actually work with the child in completing a task like building a block tower and things of that nature, where we're seeing the families and the parents from lower incomes are more likely to either complete the task for the child or otherwise be very direct, so they don't let the child actually make mistakes and complete the task on their own.  The end result of all of this is the child is more stressed, and you get again this different type of stress response.  And it also inhibits their ability to develop this self-regulation, these executive function skills.  So all this comes together in a way that we think has long-term implications for, again, both their development of school readiness skills and their later school and academic achievement. 

Dr. Guttmacher:  Jim, but isn't their evidence that a little stress can improve performance in some situations?  Can we explain why a greater amount of stress actually appears to interfere with that?

Dr. Griffin:  Yeah, no, this is one of the situations where more is not always better.  A little stress actually can be a very good thing.  It helps focus your attention.  Maybe it motivates you to do something in a way to really pay more effort, and it even has some positive things associated with it, like increased curiosity or persistence.  But what happens in the children from these impoverished backgrounds is they have such a high level of stress that basically even adding a little bit more to that, they don't get that actual little bump that actually helps most children to attend and to actually do better in a task.

Dr. Guttmacher:  Great.  And I understand what you are saying about the difference in the stress response really of kids in different economic circumstances.  But is that stress response just limited to children in poverty?  Or might other sources of stress also affect children from higher income groups?

Dr. Griffin:  Yeah, now, it really is a universal, and we certainly can see similar patterns in children exposed to chronic stress for things like, again, a chronic physical illness, perhaps dealing with a learning disability, parents divorcing, or other major life events that really put a lot of stress on the child and the family over a prolonged period of time.  Again, stress itself isn't necessarily bad; it can actually be a good thing.  But too much of it and an inappropriate level of stress over sustained periods is really not good for any child, especially a young one who's going to struggle to cope with it.

Dr. Kathy Mann Koepke:  Jim, did the study explore how the stress response altered the long-term brain development of these young children?

Dr. Griffin:  Well again, we're looking at the studies now, but certainly it's suggestive of the fact that this type of stress response can be consistent as children develop, and we know that the executive function skills actually go through kind of another growth period during adolescence.  So again, there are other researchers who have looked at the prolonged exposure to stress and executive function skills, and there may be high-risk behaviors that adolescents engage in that it may actually have some implications for, again, some other behaviors that would put them at risk as they develop later into adolescence. 

Dr. Guttmacher:  Thank you, Dr. Griffin.  Our next guest, Dr. James Mills of the Division of Epidemiology Statistics and Prevention Research at NICHD, is the senior author of a study that found that most children of women who drink heavily during pregnancy do not have the distinctive facial features we often associate with fetal alcohol syndrome, but do show signs of abnormal intellectual or behavioral development.  Jim, can you tell us more about the study findings?

Dr. James Mills:  Yes, thank you, Alan.  Most women know that drinking heavily during pregnancy can cause fetal alcohol syndrome and in fact if a woman drinks heavily the chances are between 1 and 10 and 1 and 100 that she will have a child with fetal alcohol syndrome.  What women don't know, and it's because we really don't have very good information ourselves, is what happens to the rest of the children of heavily exposed mothers.  How likely is it that these kids will be damaged, and what type of damage are they likely to get?  And to answer that question, you have to start with a group of pregnant women who are drinking heavily and then see what happens to their offspring, and that's what we did.  We went to a prenatal clinic in Chile and interviewed almost 10,000 women coming for their first prenatal care visit, to identify 101 women who were drinking very heavily, and that means averaging four or more drinks per day.  Now, I have to say that the first thing we did with these women was to tell them that this was very dangerous to their fetus, and we offered them help in stopping and sent them, if they were interested, to our alcohol clinic.  We also collected data from them on their drinking habits from the time that they conceived until they knew that they were pregnant and then from that time on in pregnancy.  And incidentally, I should say that a number of these women did either stop drinking or cut down on their drinking considerably.  Then after their babies are born, we followed them for up to 8½ years, checking them for birth defects and monitoring their growth and then looking at their intellectual and neurologic development.  This enabled us to see what effect alcohol had, looking not just for fetal alcohol syndrome, but all the effects that are generally grouped as fetal alcohol spectrum disorder.  What we found was rather disturbing.  Almost none of the children had the classic features for fetal alcohol syndrome, which was good. But 80 percent showed some evidence of alcohol damage.  The first thing we found was that the abnormal facial features that are classic for fetal alcohol syndrome were present in about 17 percent, and poor growth was twice as common in the group of children exposed to alcohol as in our unexposed comparison group.  Roughly a quarter of all the exposed children had poor growth.  But brain and nervous system abnormalities were the most common problems:  44 percent had either a low intelligence test score, language learning delay, or hyperactivity.  The children whose mothers drank heavily had a wide range of developmental and neurologic problems.  They were much more likely to have small heads, which as you know, can be a sign of poor brain growth.  A third of them had low scores on IQ testing.  In addition, they had other signs that alcohol had disrupted some aspect of brain development.  This included delayed language skills, more likely to be hyperactive than the comparison group of children, and they were more likely to have attention deficit disorder.  Now these are important findings, because language delay, hyperactivity and attention deficit are just the kind of problems that lead to poor school performance; so identifying them early may help these children to succeed.  As an aside, we know from studies of fetal alcohol syndrome that alcohol can damage many areas of the brain resulting in a multitude of serious problems.  These include everything from intellectual disabilities to learning and memory problems to difficulty making friends, clumsiness, and problems with daily living that eventually can lead to difficulty holding a job or even an increased risk of being incarcerated.  So our finding that alcohol exposed children had a wide range of intellectual and neurological problems did not come as a surprise, but it was a very disturbing finding and has important implications for the future of these children.

Dr. Guttmacher:  Jim, how might these findings influence current practice in diagnosing and treating fetal alcohol syndrome even in the absence of the distinctive facial features that are characteristic of this syndrome?

Dr. Mills:  That's a good question.  And our findings are important both for health care providers and for teachers, because they show that only a very small percentage of children who have alcohol-related problems show the classic signs of fetal alcohol syndrome—the signs that help doctors make the diagnosis.  When a child comes into a doctor's office or a classroom showing signs of intellectual disability or hyperactivity or attention deficit disorder, the physician or teacher should always have the question in the back of his or her mind:  Could this be due to alcohol?  And as we showed in our paper, 40 percent of the affected children have only neurologic problems—no growth deficit, no facial features, that would help guide people to make the correct diagnosis.  Now unfortunately, we know that alcohol damage is permanent, but there are ways that we can ameliorate it.  In fact, the earlier the problem is identified and treated, the better the child's prognosis.  The Centers for Disease Control has excellent information on their website on the diagnosis and treatment of fetal alcohol spectrum disorder, and I highly recommend that to anyone who is interested.  I should say that we are a long way from having the perfect treatments for fetal alcohol spectrum disorder, and that is excellent area for future investigation.  Nonetheless, there are drugs and behavioral therapies available that can improve the lives of affected children and adults.

Dr. Guttmacher:  Jim, given these findings, is it possible that the prevalence of fetal alcohol related disorders is higher than what has usually been estimated?

Dr. Mills:  Absolutely.  Our findings indicate that for every child we see who has the classical signs of fetal alcohol syndrome, there may be 20, 30, or even more children who have more subtle forms of alcohol damage, particularly neurologic damage.  And many of these children will be missed if no one asks about alcohol exposure before their birth.

Dr. Griffin:  Well, Jim, I guess that raises the question, though, between mothers drinking and these outcomes, was more drinking by the mother associated with more of these bad outcomes for the child?

Dr. Mills:  Good question.  We studied woman who are all drinking at what experts, in fact everyone would agree, were dangerously high levels, four or more drinks per day is a lot of alcohol.  So we suspected that we would see many children who are adversely affected by their mothers' drinking and we did.  But despite the high proportion of our children who suffered adverse effects, we were still able to show that the more women drank before they recognized that they were pregnant, the worse the outcome for the children.  There is some disagreement regarding the effects of binge drinking.  Does it cause even greater problems than consistently high levels of drinking?  And in our population, binge drinking, which incidentally is defined as five or more drinks in one day, increased the risk for poor outcomes both before and after the woman recognized that she was pregnant. 

Dr. Guttmacher:  Thank you, Dr. Mills.  Our final discussion today explores intuitive number sense, an ability we use in everyday life as well as in the classroom.  For example, people use this ability for the simple task of judging which line to join at the grocery store or estimating daily calorie intake.  This study found that the ability to estimate quantities grows across the first 30 years of life.  Like Dr. Griffin, our first visitor for the podcast, Dr. Kathy Mann Koepke is another member of the Child Development and Behavior Branch of NICHD, which also supported this study.  Kathy, could you please tell us more about it?

Dr. Mann Koepke:  I'd be delighted to.  I guess, first, let me just say that we think of it as simple ability, but the ability to estimate how many or which has more is known as the approximate number system, also the ANS.  The approximate number system doesn't rely on language.  And perhaps surprisingly to a lot of people, previous research has shown that this ability is shared across virtually all humans and many, many animals, including mammals, birds, and fish.  So it's a very basic or fundamental ability for animals and presumably a very important survival skill.  So the investigators for this study were interested in looking at the approximate number system and whether or not it changes over lifetime.  So the first thing they did was that they asked participants to agree to be a study participant.  So they got informed consent.  Keep in mind that the study was done on the web, on the Internet, so they weren't looking at the participants face to face.  The participants agreed on the web.  And then they asked them some basic questions about their personal history: how old they are, how much schooling they had, things like that.  They also asked them to report about their math and other school abilities prior in life, depending upon their age.  So if you were a young person, obviously you were still in school.  If you were an older person, we were talking about your school performance when you were younger.  The procedure is really quite simple.  On a computer screen, they show very, very briefly, lots of colored dots—two colors, blue and yellow.  And it's on the screen for about a fifth of a second or 200 milliseconds, very short time, too quickly to allow anybody to count the actual dots.  Some of the dots are yellow, some are blue.  And then they ask, "Which did you see more of, blue dots or yellow dots?"  When the number of blue dots is almost the same as the number of yellow dots (for example, if there's five blue dots and six yellow dots), it's very hard to distinguish which color had more dots.  But when the difference between the number of the different colors gets larger and larger—so, for example, five blue dots and 20 yellow dots—it becomes pretty easy to very quickly perceive which there are more of.  Now remember, you don't have enough time to count, so this is a flash that you see, and your brain registers.  Some people can tell that there are more when there are only a few more, that small difference, pretty reliably, and other people need a greater difference:  they need to see a much broader spread before they can really tell which has more dots, which color.  So it's an ability to quickly know when there is more or less of something that the investigators were interested in studying in a wide variety of people across many ages.  People who can tell the difference in number very easily are said to have a good ANS acuity, or sharper ANS acuity, and people who need more difference in the number of dots are said to have a less acute ANS or that their Weber fraction is smaller.  The Weber fraction is just the mathematical computation of the difference.  Because this is easily done on a computer, to present the dots over a short period of time, it was especially useful to do this study on the web because they could collect many, many more data points or they could test many more subjects in a very short period of time, in a much broader range of people because they didn't need to recruit just from where they were, which happens to be in Baltimore, MD, versus the whole world.  So, in fact, in 3 months' time, 10,590 people participated in this study, in an age range from 11 years old to 85 years old, so a huge range and a very short time to collect the data.  When the researchers grouped people by their age, they found that for the people aged 15 to 30, the older the person was, typically the better their ANS or their approximate number ability or acuity was, up until 30 years of age.  That's when it seemed to peak.  Most people have best performance around 30 years of age.  However, when you got past 30—particularly for participants over the age of 60—the older you got, the less acute your approximate number system was.  In other words, they were not as sensitive to small differences.  Important to note that even though it seemed to get less well tuned after the age of 60, this was a general trend over a large population.  And there was a high degree of variability so there were certainly individuals who were over age 60 who did quite well; in fact the best performance came from a person over the age of 60 in the people who were studied.  But there did seem to be this apparent trend of aging and a declining ANS for some people.  So the study identified for us that there seems to be a change over development—in particularly, the first 30 years of life—this acuity seems to get better, and then after age 60 it appears that perhaps it gets worse in some people.  But remember that this was not a longitudinal study; this was a cross-sectional study.  And so we weren't looking at the same person many times over a long period of time, which is what we would call a longitudinal study.  We're looking at different people across many different ages and guessing that because we had so many people that we were seeing a trend that we would see if we were to study the same person over and over again. 

Dr. Guttmacher:  Kathy, we've been assessing people's math abilities since the advent of formal education, if not before.  How is this approach novel?  Why does all this matter?

Dr. Mann Koepke:  Well, first of all, remember that what we're assessing typically in schools is what we taught them in school, and what we teach people in school is what we call the exact number system or the ability to count, enumerate things individually, and manipulate those individuals so that we have exact numbers.  The approximate number system isn't doing exact number math; it's doing something very different.  So it doesn't require language, it isn't exact or enumerating, and it isn't directly taught in school.  So the measures here are very different from the kinds of things you would see in a classroom where you're measuring people's math ability.  The ANS can be measured, the approximate number system can be measured, as I said, across many animals and is, at least in humans, we know, present at birth, so it's theorized by many scientists to be the basis or the foundation for all later mathematical ability.  Several studies show strong correlations between past and future arithmetic or mathematical skills that are required in school that are correlated with your approximate number acuity.  In addition, we have found that children who have math learning disability, or MLD, have been shown to have a weaker approximate number system, and so their acuity is not as good.  And we're wondering whether this approximate number system sets people up for the ability to learn exact number.

Dr. Guttmacher:  Kathy, I think you said that the youngest individual in the study was 11 years old.  How early in life does this intuitive grasp of numbers become apparent?  Can it be documented in very young children?

Dr. Mann Koepke:  Yes.  Actually, we have a number of studies that document that this approximate number system exists even before the first year of life.  One study in particular looked at newborns, meaning infants 7 to 48 hours old, and again demonstrated that the approximate number system was present even at 7 hours or 2 days of life. 

Dr. Mills:  Kathy, how do environmental factors such as education play a role in the approximate number sense?  Might this ability actually be improved through educational methods?

Dr. Mann Koepke:  That is a great question, and we don't know the answer.  This study can't answer that, but it's a question that's a very important question and one we want to know.  So one of the things that's obvious to do is to see if we can train the approximate number system in some way.  And we actually have several investigators who are doing those studies right now—this investigator as well as others.  So stay tuned, we hope in the next 6 to 9 months we'll be able to come back and give you the answer to that.  We're hoping that it is trainable because if it is trainable and if the approximate number system really does set you up for the ability to learn exact number, then it would be very important especially for people who are at risk for difficulty with math, if we could train their approximate number system and then hopefully move their ability to do exact math, the kind of math that we learn in school. 

Dr. Guttmacher:  Kathy, thanks for that explanation and the introduction to approximate number sense, which is quite interesting.  I can understand how it might even have an evolutionary advantage for our species and our ancestors who were hunters; for instance, to be able to look across the savannah and get some idea about the relative size of herds and predators, etc., might be useful.  Do we know anything about other species?  Is this a uniquely human quality, or are there other species that can do this?  Has anyone looked at that?

Dr. Mann Koepke:  Yes, and actually it's been studied very extensively.  So, to date, we have studied in every vertebrate that we have studied that which has included a wide variety of mammals, a number of birds, as well as fish.  And in every species we have tested and we've only tested vertebrates, we have seen the approximate number system present, so that tells us that this is a very old, followed genetically, ability and it tells us how very important it probably is.  It's important to every organism's survival, to be able to tell "stay away from that group, there is a lot of bad people over there," or "go over to this bush because there is a lot of fruit on this bush, and I'm wasting my time crossing the savannah to go see the bush that has many fewer berries on it."  So it's a really important survival skill.  And it also suggests to us that it probably at least sets the stage in the brain in terms of neurodevelopment for math ability.  Now whether or not it is what eventually becomes the math ability that lets us do, for example, algebra, we don't know yet.  And there are some scientists that would argue strongly that, yes, it is the foundation, and then there are other people who say no, it probably isn't, but the fact that it's present in so many organisms tells us that the brain is designed to be able to do at least some kinds of math.  And that's good news because it means that—even though there are a lot of people walking around, if you ask them, "oh no, I can't do math"—the fact of the matter is that their brain was programmed to do at least some kinds of math.  And that means that they should be reassured that they really can learn.  That this is something even fish can do, and they can do too, they just don't recognize it. 

Dr. Guttmacher:  Well, thank you, Dr. Mann Koepke, it's interesting to know that a survival skill that was good in the savannah is still good in the 7-Eleven.  We evolve but still keep some of the same abilities.  That brings us to the end of our podcast for this month.  I'd like to thank Dr. Griffin, Dr. Mills, and Dr. Mann Koepke, for joining us today and for sharing with us some of the research supported by the NICHD.  I'd also like to thank our podcast listeners for joining us.  For more information on any of today's topics and many related topics, visit  That's  I'm Alan Guttmacher, and I hope you will join us for more NICHD podcasts as we post them on our website each month.

Announcer:  This has been NICHD Research Perspectives, a monthly podcast series hosted by Dr. Alan Guttmacher.  To listen to previous installments, visit  If you have any questions or comments, please email

Back to Research Perspectives.

Last Reviewed: 11/30/2012
Vision National Institutes of Health Home BOND National Institues of Health Home Home Storz Lab: Section on Environmental Gene Regulation Home Machner Lab: Unit on Microbial Pathogenesis Home Division of Intramural Population Health Research Home Bonifacino Lab: Section on Intracellular Protein Trafficking Home Lilly Lab: Section on Gamete Development Home Lippincott-Schwartz Lab: Section on Organelle Biology