Do critical periods determine what we can learn and when we can learn it? Neuroscientists and social scientists are probing children’s brains to answer these important questions.
Walk into a toy store and you’ll see a wide range of educational toys for babies and toddlers: beautiful little cards to enhance undeveloped math skills, video toys to guide babies in reading, and a dizzying array of colorful soft toys. Often, these products come from the “not-so-precise” message that if you don’t fill your baby’s brain with “learning” until he’s three, he won’t be able to reach his full potential.
The idea that the first three years of life are critical for learning exploded in the public consciousness after the 1997 White House conference on “Early Childhood Development. Based on neurobiological evidence that the infant brain continues to develop after birth, the conference identified a range of programs necessary to ensure that poor children have normal, healthy learning experiences throughout their childhood, including the first three years.
Carla Shaz, a developmental neurobiologist and professor in the Division of Neurobiology at Harvard Medical School in Boston, said, “No neurobiologist is saying that ages 0 to 3 are the most important time for learning, and there is a logical disconnect happening here.
While the conference’s message may have been misunderstood or misapplied, it really underscores the need for a scientific understanding of the role of critical periods in learning and social development. They are defined as “windows of time” when the brain not only becomes adept at receiving certain aspects of information, but actually needs that information to continue its normal development. The critical period details the development of sensory systems in the brain, particularly in the visual area. But many neuroscientists also believe that at least some brain functions that are potentially useful for complex learning and thinking also have their own critical periods. There is a wealth of evidence for critical periods, for example, in language learning, where there is a clear critical period. William Greenough of the University of Illinois at Urbana Champaign says, “There is indeed a critical period in human development, and there is no evidence for it. There are critical periods in human development, but there is little evidence to support the idea that there are many critical periods.”
What scientists agree on is that critical periods do exist, but they are not as clear as the media reports, nor are they as well positioned as educational toy makers. No critical period shuts down suddenly; they stop gradually. Nor is it true that critical periods are limited to the first three years; those types of learning that do have critical periods all have different critical periods. In many cases, the window never seems to close completely, and learning, while more difficult, continues into adulthood. For this reason, researchers prefer to use the term “sensitive period”. The baseline is that while it’s easier to learn a language or music as a child, adults can do it, too. Peter Huttenlocher, a pediatric neurologist at the University of Chicago, says, “That’s usually how science works, the truth is somewhere in between (the two extreme views).”
Early insight
Undoubtedly, there is a critical period for some brain development. The most famous example comes from David Hubel and Torsten Wiesel at Harvard University in 1960, who showed that if an eye is sutured early in a kitten’s life, that eye is permanently blind because the brain’s visual system misses visual input during a major stage of brain development. Children with congenital strabismus or cataracts in their eyes also show that there are indeed critical periods in human visual development.
In the intervening decades, several different kinds of studies have also shown that there are sensitive periods for different kinds of learning in the brain, with some evidence coming from neuroscientists who have used brain imaging and other techniques to study changes in the brain and correlate them with behavior and learning. Other evidence is purely behavioral and comes from psychiatric and educational research.
Research in this area also emphasizes the importance of the first three years. For example, a 1950 psychiatric study found that children were emotionally attached to their mothers or primary caregivers during the first year. Since then, many researchers have shown that babies who are safely and carefully cared for trust their caregivers to protect and nurture them, and that these babies form better relationships with others later in life than those in secondary care. Ross Thompson, a developmental psychologist at Lincoln University in Nebraska, believes that the “attachment” relationship is so critical to infant survival in human evolution that the critical period during which it is formed may The critical period for its formation may have been programmed into the developing brain.
Attachment” is an important underlying principle in the study of critical periods of emotional development in children 0 to 3 years of age. However, Thompson notes several studies in which children spent their early years in Romanian orphanages without normal human contact that would have enabled attachment formation, suggesting that the window for “attachment” is larger than we might think. Children who are “rescued” at ages 4, 5, and 6 can also form attachments. But many of those attachments are weak or unhealthy, and may be signs of a gradual shutdown of the sensitive period. But without evidence of true closure time, Thompson argues that it cannot be called a “real” sensitive period. Another consideration in drawing conclusions from the Romanian orphanage study is that these children were deprived in many ways, and there may have been many other substantial reasons for their “attachment”.
Another social science study that raises the importance of the first three years is that of Frances Campbell of the University of California and Craig and Sharon Ramey of the University of Alabama, who, in an analysis of over 1,000 specially designed educational programs for children from relatively poor families, found that children who attended these programs from birth through preschool showed significant improvements in IQ and school performance (compared to the previous year). Ramey says, “This is evidence that missed opportunities at ages 3-5 cannot be made up later in life.”
The impact of complex external conditions on brain development
No one can define a critical period during which the brain needs a rich environment if changes are to occur. This period is remarkable because it is associated with child development and may offer hope to children who lack adequate education early on. Practitioners in the field of child development have found that culturally deprived children develop more slowly than culturally enriched children, and others have pointed to cases in which the harmful effects of an early deprived environment were reversed in a later enriched environment. However, James Prescott of the National Institute of Child Health and Human Development argues that the rapidly accumulating evidence that these effects of rearing environments in animals on brain development may be similar to those in children is a matter for serious consideration. He believes that this study, which found different changes in the brain under different rearing conditions, is striking and changes the current common belief that only a barren environment can affect brain development in children.
The effects of either enriched environment are individual-specific and undoubtedly interact with other factors such as sex, genetic composition, and nutritional status of the individual human. For example, the assignment of mice from the same litter to individual rearing environments is designed to control for the effects of minimal genetic differences in brain development, which always interfere with the effects of the rearing environment. In addition, there is evidence that an enriched environment can somewhat eliminate the deleterious effects of brain damage, malnutrition, or hypothyroidism in mice, etc. Sackett notes that some mice reared in barren environments do not exhibit abnormal behavior, suggesting that there are unknown factors that interact with the conditions in which the animals are reared and keep them somewhat unaffected. Examination of these unknowns may provide clues as to how to reduce the effects of barrenness and enhance the effects of enrichment.
We have established that the enriched environment has definite effects on the brain, but the significance of these effects is not yet clear. Do these “very mice” have advanced abilities because they have more neural connections and, in general, more complex brains? Most researchers agree that the brain changes following enrichment have some beneficial effects on brain organs, but there are counter arguments for behavioral changes, and so far it is not certain that brain changes are responsible for the experimentally documented behavioral changes. So although studies have shown that “enriched” environments early in life have beneficial effects on the brain, Bruer believes the results have been misused, especially in some products such as educational audio and video specifically for middle-class families. He believes that for most children, there is no scientific evidence that “extra enrichment” beyond the normal environment provides many benefits. Not only are parents of middle-class families misinformed about what to offer their children, but society’s attention is also diverted from the needs of truly poor families, which do not end at age 3. To support this view, he cites the widely used animal experiments that promote educational assistance to enrich the infant’s environment.
In the previous 20 years of research, Greenough showed that mice raised in so-called “complex environments” with other mice and lots of toys to play with formed more synaptic neural connections in their brains (compared to mice raised alone in standard laboratory cages). But Greenough said his study was more experimental deprivation than enrichment, because the so-called complex environment was actually close to the normal growth environment for mice. He says his findings imply that extreme deprivation is decisive, but he doesn’t stress whether additional stimulation in the normal environment is really better.
Greenough also noted that his study had little direct relationship to the rich environment of the first 3 years of human childhood, because his mice did not enter the complex environment until after weaning, roughly equivalent to 2.5 to 5 years of age in humans, and those mice stayed there until adolescence. Also, when the researchers exposed adult mice to a similar environment, their neural connections proliferated,” Greenough said. “With young animals, the changes occur faster, the magnitude is greater, and the effects are not easily lost. Seeing these experiments, one realizes that it completely undermines the notion that everything is over by age 3.”
In short, there are a whole lot of questions to be discovered about how the brain functions before answering why a rich environment can affect the brain and what those effects mean for the animal, but there is no doubt that brain development is linked to the richness of the environment in which the animal grows.
Windows of learning
While the Ramey-Campbell, and Greenough studies have focused on the brain’s readiness for learning, other researchers have examined whether there are critical periods for certain skills, such as music and language. For example, in a 1995 brain imaging study of musicians, Thomas Elbert of Konstanz University in Germany and Edward Taub of Alabama University in Birmingham found that the left hand of string musicians was more highly represented by the tactile areas of the brain (compared to left-handed non-musicians).
The researchers’ main conclusion was that the brain’s ability to change its response to musical training extends well into adulthood. But they also found that string musicians who began training before age 12 had the largest brain areas devoted to left-handed sensation, which may indicate receptivity to musical training earlier in life. The difference in total training time between the two groups is too small to explain the quantitative differences in brain organization.
Researchers who have observed the brain’s reception of language learning control for training effects, and they believe they have clearer evidence that young brains can learn a language as fluently as very few older learners can achieve. In other words, there is indeed a sensitive period for language acquisition.
Some of the stronger behavioral evidence comes from Elissa and colleagues, cognitive scientists at Newport and the University of Rochester in New York. In the late 1980s, they studied 46 Chinese and Korean immigrants to the United States who were integrated into an English-speaking environment from ages 3 to 39. To rule out training effects, the researchers matched their use of English to the subjects’ time, then played back their oral recordings, some with grammatical errors, such as misordered words or incorrect verb tenses, and asked them if they were speaking the correct sentences.
For natives, this test was less difficult, but for immigrants, the researchers found “a systematic decline in correct responses as a function of age when people first arrived in the United States.” Those who arrived in the U.S. before age 5 or so did as well as natives, Newport said, but “each group after that systematically declined,” and the curve smoothed out for those who immigrated after their teens. “That’s the shape of the critical period that one would expect.”
You might argue, Newport says, that people who learn English later in life don’t learn it well not because the critical period for learning the language has passed, but because their years of experience with their native language have severely interfered with the learning of another language. To determine this, Newport and her husband, Ted Supalla, worked on a special group of people who did not learn any language at an early age, who were deaf, whose hearing parents did not use American Sign Language (ASL), and who did not learn ASL until they were 5 or 12 years old when they entered a boarding school for the deaf. Newport and Supalla, who are deaf themselves, compared learners who began learning ASL later to deaf individuals who learned ASL sign language at birth. To exclude the effects of training, they chose individuals who were 50 to 70 years old at the time of testing, and who had been using ASL for at least 48 years.
The researchers tested ASL for sentence construction and sentence comprehension, and they found the same kinds of pictures as in the second language learning process. Those who did not begin using ASL until age 5 scored slightly lower on average than those who had been in an ASL environment since birth, but those who did not begin learning until age 12 scored even lower.
The behavioral findings on language development are supported by neurobiology. Neuroscientist Helen Neville of Oregon Eugene University has looked at the brain tissue of Chinese and Spanish immigrants who began learning English between the ages of 2 and 16. Using brain imaging, she and her colleagues observed patterns of brain activity when those people heard sentences with grammatical errors, like the ones Newport also used. “Among those who learned English later, even if they started at age 4, we found that their brains were organized differently in response to “novel” grammar,” Neville says. Neville said. Among those who learned a second language before age 4, the response occurred entirely in the left brain, which is the normal language area, while those who learned later showed more right hemisphere activity. This implies that the specific physical representation of the brain differs between late and early language learning.
These results are consistent with what Newport and her colleagues found when they looked at grammatical abilities. But grammar is only one element of language learning; other elements include phonology (how language is pronounced) and semantics (what words mean), which do not need to have a sensitivity period. For example, Neville says that when she and her colleagues observe people’s brain and behavioral responses to “novel semantics,” i.e., a sentence in which one of the words does not have the same meaning, people who learn the language later react the same way as those who learn it earlier. There does not seem to be a critical or sensitive period in the strict sense.
Even for just one aspect of language, such as phonology, there can be different windows of learning. One part of speech has to be learned early, while another part can be
other parts can be gradually improved over a long period of time. This means that language is not a mere monolithic circuit, similar to systems in which one part has a critical period and one part does not.
Basic principles of the brain
The brain matures slowly during childhood, and Neville and others have suggested that the pace of maturation affects the timing of the critical period. For example, Chicago’s Huttenlocher and his colleagues have studied the neural connections in the postmortem brain of children of different ages. They found that neural connections proliferated in most brain regions during the first year, after which there was a period of high neural connection density, ranging from 6 months to 12 months to 5 or 15 years, depending on the region. After that, neural connectivity levels begin to decline, with visual areas being the first to lose their neural connections, followed by higher cognitive areas dropping to adult levels. Harry Chugani and colleagues at Wayne State University in Detroit have used positron emission tomography (PET) imaging to measure metabolic effects in the brains of infants and children as an indirect way to observe the proliferation and loss of neural connections, and have come to the same conclusion.
Huttenlocher noted that when he saw neural connections begin to proliferate, the basic functions of brain regions were formed at that time. For example, when the neural connections in the visual cortex begin to proliferate, the child begins to have binocular vision. The pruning of neural connections seems to be related to, or at least extremely similar to, the “upper limit of time to learn a task easily”. For example, although there are different sensitivity periods for different aspects of language learning, 12 to 14 years of age is probably when the ease of language learning begins to decline in general, and this is also probably when the density and number of neural connections in language areas of the brain decline.
Despite these interrelations, some neuroscientists and psychologists have suggested that some of the apparently sensitive periods may have more to do with cumulative learning itself than with the physical development of the brain. Alison Gopnik, a psychologist at the University of California, has found that children as young as four years old recognize that others have thoughts and opinions different from their own. Her research implies that this recognition occurs when it is supposed to, because children have accumulated enough experience to reach this conclusion. She was able to accelerate the learning of these concepts, for example, by giving students special training that emphasized that others have different ideas. This implies, she says, that there is no “certain maturation event” in the brain that proves a skill will emerge at a certain time, and that the new things you learn allow you to learn newer things.
The question now is whether, at least for some kinds of complex learning, learning drives change in the maturing brain, or whether the maturation process controls the ease of learning. These questions can be highlighted as the brain structure associated with different kinds of learning. For example, she is currently experimenting with children to see if training that accelerates their language learning leads to measurable changes in brain organization, and several research groups are beginning to use brain imaging to examine changes in the organization of brain regions that include “attachment” formation. She predicts that no answer will be ambiguous. We know the full kit of the brain system. The answer may be that any one person’s system is different.
Summary
Because researchers have pooled their resources to clarify the role of critical periods in learning, the theme emerges that while younger brains may be more susceptible to change, older brains have not lost the ability to change. While childhood is clearly an exceptional time for learning, there is no reason to give up hope of learning at any age. Indeed, Newport says, research may shed light on whether the mechanisms of late learning are different from those of childhood. With a better understanding of these differences, people can come up with different approaches and strategies to improve adult instructional programs. And that would be good news for anyone who is eager to learn at any age.