How the Brain Learns to Read Read online

  A particularly significant two-part longitudinal study (Hart & Risley, 2003) documented the vocabulary growth of 42 toddlers from the age of 7 to 9 months until they turned 3 years old. Because parental vocabulary is closely associated with a family’s socioeconomic status (SES), Part 1 of this study looked at toddlers in families from three different groups. On the basis of occupation, 13 of the families were upper SES, 23 were middle-lower SES, and 6 were on welfare. By the time the children were 3 years old, the researchers had recorded and analyzed more than 1,300 hours of casual conversations between the children and their parents. To their surprise, the analysis showed a wide gap in the number of words the children heard from their parents and in the number present in their vocabularies, based on their SES. Table 1.2 summarizes the data. Children from the welfare families heard an average of 616 words per hour and had an average recorded vocabulary size of just 525 words. Those from the middle-lower SES heard 1,251 words in an hour and had 749 words in their vocabulary, while the children in the upper SES heard 2,151 words each hour and had an average vocabulary of 1,116 words. Furthermore, the children from welfare families were adding words to their vocabulary more slowly than the other children throughout the length of the study.

  Figure 1.2 The dotted regions represent the areas of highest event-related potentials when subjects processed image-loaded words or verbal (abstract) words (Swaab et al., 2002).

  Part 2 of the study was conducted six years later. The researchers were able to test the language skills of 29 of these children who were then in third grade. Test results showed that the rate of early vocabulary growth was a strong predictor of scores at ages 9 to 10 on tests of vocabulary, listening, speaking, syntax, and semantics. This study points out how important the early years are in developing a child’s literacy and how difficult it is to equalize children’s preschool experiences with language. These results reveal the enormous vocabulary gap that is formed between students of different socioeconomic levels. Regrettably, the gap continues to widen due to the cumulative experiences that the children have with vocabulary at their SES. In the United States, early literacy problems can be addressed successfully through the publically funded birth-to-school programs now available through the federal Head Start initiative and in several states and a growing number of school districts. In some of these programs, such as the Parents as Teachers initiative, school district personnel meet regularly with parents of infants in low-SES households and provide them with inexpensive, age-appropriate resources to use with their children during the preschool years. The idea is to build the child’s vocabulary and exposure to enriched language before they enter school. Such programs provide significant cost savings to local school districts by having fewer placements in special education, fewer grade retentions, and less remedial education. Moreover, one of the most reliable predictors of how well and how quickly youngsters will learn to read is the size of their mental lexicons. See the “Resources” section for information on Head Start and Parents as Teachers.

  “A reliable predictor of how well youngsters will learn to read is the size of their mental lexicon.”

  Table 1.2 English Words Heard per Hour at Home and Vocabulary Size by Three Years of Age in Various Economic Groups

  SOURCE: Hart and Risley (2003).

  Impact of Television Viewing. Studies have found that one major reason toddlers from low-SES homes hear less vocabulary from their mothers is that they spend more time in front of television sets than toddlers in higher-SES groups. Two major studies showed that toddlers in low-SES homes spent an average of two hours per day viewing television (Christakis et al., 2009; Mendelsohn et al., 2008). More often, the mothers sat quietly and watched the programs with their child, reducing the amount of time for verbal communication between them. Placing toddlers to watch television may explain the association between infant television exposure and delayed language development.

  Further support to this notion that television viewing during infancy can negatively affect language development came from a longitudinal study of more than 250 families (Tomopoulos et al., 2010). In this study, infants were assessed for cognitive and language development at 6 months and 14 months of age, and the results were compared to the average amount of time they watched television each day—from 0 to 360 minutes. Findings from the study were eye-opening: The more time the infants spent watching television between the ages of 6 and 14 months, the lower their cognitive development and language scores at the age of 14 months. Surprisingly, the type of television programing—that is, child-, adolescent-, or adult-oriented—made little difference. Similar long-range studies have found that average television viewing prior to 3 years of age was negatively associated with cognitive outcomes when children reached 6 years of age (Zimmerman & Christakis, 2005). These findings may seem to run counter to common sense. If the infant is spending so much time viewing and listening to television, shouldn’t these experiences increase vocabulary and enhance language development? Obviously, not. Why? Because so much of early language learning in the infant’s brain relies on other important clues to attach meaning to spoken words. The child is closely watching the parent for facial cues and listening to intonation, intensity, and rhythm. These cues are very noticeable during face-to-face communication between parent and child, but largely absent from the impersonal output of radio and television. For these reasons, the American Academy of Pediatrics recommends no media exposure prior to 2 years of age.

  “Exposing children prior to 2 years of age to media may cause significant delays in cognitive processing and language development.”

  Syntax and Semantics

  Language Hierarchy

  With more exposure to speech, the brain begins to recognize the beginnings of a hierarchy of language (Figure 1.3). Phonemes, the basic sounds, can be combined into morphemes, and through a set of conventions, morphemes can be combined into words. These words may accept prefixes, suffixes, and infixes (insertions), and may undergo a change of consonants or vowels. Words can be put together according to the rules of syntax (word order) to form phrases and sentences with meaning. The difference in meaning (semantics) between the sentences “The woman chased the dog” and “The dog chased the woman” results from a different word order, or syntax. Toddlers show evidence of their progression through the syntactic and semantic levels when simple statements, such as “Candy,” evolve to more complex ones, for example “Give me candy.” They also begin to recognize that shifting the words in sentences can change their meaning.

  The Syntactic Network

  The rules of syntax in English prohibit the random arrangement of words in a sentence. The simplest sentences follow a sequence common to many languages, that of subject-verb-object (or SVO) format, as in “He hit the ball.” In more complex sentences, syntax imposes a stringent structure on word order to provide clarity and reduce ambiguity. Brain areas near the front of the temporal lobe seem to concentrate on establishing the meanings that could emerge when words are combined into sentences (Vandenberghe, Nobre, & Price, 2002). Just look at what happens to meaning when writers neglect to follow the rules of syntax. The following examples are taken from actual headlines that appeared in the nation’s newspapers.

  • Rescue Squad Helps Dog Bite Victim

  • Safety Experts Say School Bus Passengers Should Be Belted

  • Dealers Will Hear Car Talk at Noon

  • Sex Education Delayed, Teachers Request Training

  Figure 1.3 This diagram represents the levels of hierarchy in language and in language acquisition. Although the process at the beginning usually flows from the bottom to the top, recycling from the top to lower levels also occurs, as indicated by the arrows to the left. Through each step, the child’s vocabulary continues to grow rapidly.

  Over time, the child hears more patterns of word combinations, phrase constructions, and variations in the pronunciation of words. Toddlers detect patterns of the SVO word order—person, action, object—so they can soon say,
“I want cookie.” They also note statistical regularities heard in the flow of the native tongue. They discern that some words describe objects while others describe actions. Other features of grammar emerge, such as tense. By the age of 3 years, over 90 percent of sentences uttered are grammatically correct because the child has constructed a syntactic network that stores perceived rules of grammar. For example, the child hears variations in the pronunciation of walk and walked, play and played, and fold and folded. The child isolates the -ed and eventually recognizes it as representing the past tense. At that point, the child’s syntactic network is modified to include the rule: “add -ed to make the past tense.” The rule is certainly helpful, but causes errors when the child applies it to some common verbs. Errors are seldom random, but usually result from following perceived rules of grammar such as the “add -ed” rule. If “I batted the ball” makes sense, why shouldn’t “I holded the bat”? After all, if fold becomes folded, shouldn’t hold become holded? Regrettably, the toddler has yet to learn that over 150 of the most commonly used verbs in English are irregularly conjugated (Pinker, 1999).

  Why do these common past-tense errors occur in a child’s speech, and how do they get corrected? Once the “add -ed” rule becomes part of the syntactic network, it operates without conscious thought (Figure 1.4). So when the child wants to use the past tense, the syntactic network automatically adds the -ed to play and look so that the child can say, “I played with Susan, and we looked at some books.” If, however, the child says “I holded the bat,” repeated adult corrections, repetition, and other environmental encounters will inform the syntactic network that the “add -ed” rule is not appropriate in this case and should be blocked, and that a new word held should be substituted and added to the child’s lexicon. This principle of blocking is an important component of accurate language fluency and, eventually, of reading fluency.

  Figure 1.4 These diagrams illustrate how blocking becomes part of the syntactic network. Before a child encounters an irregular verb, the “add -ed” rule applies. Thus walk becomes walked and hold becomes holded. After several instances of adult correction and other environmental exposures (repetition is important to memory), the syntactic network is modified to block the rule for the past tense of hold and to substitute held, a word that now becomes part of the child’s lexicon (Pinker, 1999).

  So how does the child eventually learn the irregular forms of common verbs? Long-term memory plays an important role in helping the child learn the correct past-tense forms of irregular verbs. The more frequently the irregular verb is used, the more likely the child will remember it. Take a look at Table 1.3, which shows the 10 most common verbs in English as computed by Brigham Young University. Known as the Corpus of Contemporary American English, the verbs are drawn from a 450 million–word database of text used in magazines, newspapers, textbooks, popular books, and other sources (Davies, 2012). Note all of these most common verbs are irregular. Interestingly enough, this tends to be true in many other languages. Pinker (1999) explains that irregular forms have to be repeatedly memorized to survive in a language from generation to generation; otherwise the verbs will be lost. He cites several infrequently used irregular verbs whose past tenses have slipped from common usage: cleave/clove, stave/stove, and chide/chid.

  Table 1.3 Frequency of Common Verbs in English in a Million Words of Text

  SOURCE: Davies (2012).

  The ability of children to remember corrections to grammatical errors—including blocking—would be impossible without some innate mechanism that is genetically guided. No one knows how much grammar a child learns just by listening, or how much is prewired. What is certain is that the more children are exposed to spoken language in the early years, the more quickly they can discriminate between phonemes, recognize word boundaries, and detect the emerging rules of grammar that result in meaning.

  Syntax and English Language Learners. Each language has its own rules of syntax. Consequently, children learning English as an additional language often have problems with English syntax. For example, unlike English, adjectives in Spanish, French, and other similar languages are typically placed after the noun they modify. Blue sky is spoken in Spanish as cielo azul and in French as ciel bleu. German verbs usually are placed at the end of a clause and rarely follow the SVO sequence so common in English.

  English and the Romance languages are subject-prominent in that every sentence must have a subject in the initial position, even if the subject plays no role, as in “It is raining” or “It is possible that the sun will shine today.” Other languages, like Japanese, Mandarin, and Korean, are topic-prominent in which the topic holds the initial position and there may or may not be a subject. For example, “It is cold in here” becomes in Mandarin “Here very cold.” In Korean, “The 747 is a big airplane” becomes “Airplanes [topic] the 747 is big.” As for Japanese, “Red snapper is my favorite fish” translates to “Fish [topic] red snapper favorite it is [note the subject-object-verb string].” Topic-prominent languages also downplay the role of the passive voice and avoid “dummy subjects,” such as the It in “It is raining.” For these beginning English language learners, special attention has to be paid to understanding how English rules of syntax differ from those of their native tongue.

  The Semantic Network

  As phonemes combine into morphemes, and morphemes into words, and words into phrases, the mind needs to arrange and compose these pieces into sentences that express what the speaker wants to say. Meanwhile, the listener’s language areas must recognize speech sounds from other background noise and interpret the speaker’s meaning. This interaction between the components of language and the mind in search of meaning is referred to as semantics. Meaning occurs at three different levels of language: the morphology level, the vocabulary level, and the sentence level.

  Morphology-Level Semantics. Meaning can come through word parts, or morphology. The word biggest has two morphemes, big and -est. When children can successfully examine the morphology of words, their mental lexicons are greatly enriched. They learn that words with common roots often have common meaning, such as nation and national, and that prefixes and suffixes alter the meaning of words in certain ways. Morphology also helps children learn and create new words, and can help them spell and pronounce words correctly.

  Vocabulary-Level Semantics. A listener who does not understand many of the vocabulary words in a conversation will have trouble comprehending meaning. Of course, the listener may infer meaning based on context, but this is unreliable unless the listener understands most of the vocabulary. Children face this dilemma every day as adults around them use words they do not understand.

  Sentence-Level Semantics. The sentence “Boiling cool dreams walk quickly to the goodness” illustrates that morphology and syntax can be preserved even in a sentence that lacks semantics. The words are all correct English words in the proper syntactic sequence, but the sentence does not make sense. Adults recognize this lack of sense immediately. But children often encounter spoken language that does not make sense to them. To understand language, the listener has to detect meaning at several different levels. Because adults do not normally speak sentences that have no meaning, a child’s difficulty in finding meaning may result from a sentence having meaning for one person but not another. At this level, too, the listener’s background knowledge or experience with the topic being discussed will influence meaning.

  The cerebral processes involved in producing and interpreting meaning must occur at incredible speed during the flow of ordinary conversation. How it is that we can access words from our enormous storehouse (the mental lexicon) and interpret the meaning of conversation so quickly? What types of neural networks can allow for such speed and accuracy? Although linguistic researchers differ on the exact nature of these networks, most agree that the mental lexicon is organized according to meaningful relationships between words. Experimental evidence for this notion comes form numerous studies that involve word priming. In
these studies, the subjects are presented with pairs of words. The first word is called the prime, and the second word is the target. The target can be a real word or a nonword (like spretz). A real word target may or may not be related in meaning to the prime. After being shown the prime, the subject must decide as quickly as possible if the target is a word. The results invariably show that subjects are faster and more accurate in making decisions about target words that are related in meaning to the prime (e.g., swan/goose) than to an unrelated prime (e.g., tulip/goose). Researchers suspect that the reduced time for identifying related pairs results from these words being physically closer to each other among the neurons that make up the semantic network, and that related words may be stored together in specific cerebral regions (Gazzaniga, Ivry, & Mangun, 2002).

  Additional evidence for this idea that the brain stores related words together has come from several imaging studies. Subjects were asked to name persons, animals, and tools. The results (Figure 1.5) showed that naming items in the same category activated the same area of the brain (Beauchamp, Lee, Argall, & Martin, 2004; Chouinard & Goodale, 2010; Damasio, Grabowski, Tranel, Hichwa, & Damasio, 1996; Morris & Stockall, 2012). It seems that the brain stores clusters of closely associated words in a tightly packed network so that words within the network can activate each other in minimal time. Activating words between networks, however, takes slightly longer.

  Implication for Teaching and Learning: “Because of the brain’s apparent affinity for storing related words in the same cerebral region, teachers may want to purposefully group related words into lessons aimed at acquiring new vocabulary.”