How the Brain Learns to Read Read online

  Figure 1.5 This diagram is a representation of the combined imaging scan results showing that naming persons, animals, and tools mostly activated different parts of the brain’s temporal lobe (Chouinard & Goodale, 2010; Damasio et al., 1996).

  How can we best represent these networks? Several different models have been proposed. One that seems to garner substantial support from contemporary neuroscientists is based on an earlier model first proposed by Collins and Loftus in 1975. In this model, words that are related are connected to each other. The distance between the connection is determined by the semantic relationship between the words. Figure 1.6 is an example of a semantic network. Note that the word lemon is close to—and has a strong connection to—the word grapefruit, but is distant from the word bird. If we hear the word lemon, then the neural area that represents lemon will be activated in the semantic network. Other words in the network such as lime and grapefruit will also be activated and, therefore, accessed very quickly. The word bird will not come to mind (Grainger & Ziegler, 2007; Marupaka, Iyer, & Minai, 2012).

  From Words to Sentences

  We have just discussed how the brain acquires, stores, and recognizes words. But to communicate effectively, the words must be arranged in a sequence that makes sense. Languages have developed certain rules—called grammar—that govern the order of words so that speakers of the language can understand each other. In some languages, such as English, different arrangements of words in a sentence can result in the same meaning. “The girl ate the candy” has the same meaning as “The candy was eaten by the girl.” Of course, different word arrangements (syntax) can lead to different meanings, as in “The boat is in the water” and “The water is in the boat.”

  As a child’s syntactic and semantic networks develop, context plays an important role in determining meaning. When hearing the sentence “The man bought a hot dog at the fair,” the youngster is very likely to picture the man eating a frankfurter rather than a steaming, furry animal that barks. That’s because the rest of the sentence establishes a context that is compatible with the first interpretation but not the second.

  How does the young brain learn to process the structure of sentences? One prominent model suggests that words in a sentence are assigned syntactic roles and grouped into syntactic phrases (Pinker, 1999). For example, the sentence “The horse eats the hay” consists of a noun phrase (the horse), a verb (eats), and another noun phrase (the hay). A rule of grammar is that a verb (V) can be combined with its direct object to form a verb phrase (VP). In the preceding example, the verb phrase is eats the hay. The combination of the noun phrase (NP) and the verb phrase comprises the sentence (S), which can be represented by the syntactic model shown in Figure 1.7.

  Figure 1.6 This is a representation of a semantic network. Words that are semantically related are closer together in the network, such as lemon and yellow, than words that have no close relationship, such as lemon and bird. Similar geometric figures identify semantically related words. The lines connect words from different networks that are associated, such as lemon and yellow.

  Figure 1.7 This model illustrates how the brain may process sentences to establish meaning. By grouping, or chunking, individual words into phrases, processing time is decreased.

  Figure 1.8 This illustrates how the brain proceeds to make additional chunks into phrases to ensure rapid processing and accurate interpretation.

  As sentences become more complicated, each module can contain another module within it. For example, the sentence “The parent told the principal her son is ill” contains a verb phrase that is also a sentence (her son is ill). To ensure rapid processing and accurate comprehension, the brain groups the phrases into the hierarchy as represented by the diagram shown in Figure 1.8.

  How Can We Speak So Rapidly?

  This module-within-a-module pattern (Figure 1.8) has two major advantages. First, by rearranging and including different phrase packets, the brain can generate and understand an enormous number of sentences without having to memorize every imaginable sentence verbatim. Second, this pattern allows the brain to process syntactic information quickly so that it can meet the demanding comprehension time required for normal conversation. The efficiency of the system is amazing! The young adult brain can determine the meaning of a spoken word in about one-fifth of a second. The brain needs just one-fourth of a second to name an object and about the same amount of time to pronounce it. For readers, the meaning of a printed word is registered in an astounding one-eighth of a second (Pinker, 1999).

  Recognizing Meaning

  The brain’s ability to recognize different meanings in sentence structure is possible because Broca’s and Wernicke’s areas and other smaller cerebral regions establish linked networks that can understand the difference between “The dog chased the cat” and “The cat chased the dog.” In a functional magnetic resonance imaging (fMRI) study, Dapretto and Bookheimer (1999) found that Broca’s and Wernicke’s areas work together to determine whether changes in syntax or semantics result in changes in meaning. For example, “The policeman arrested the thief” and “The thief was arrested by the policeman” have different syntax but the same meaning. The fMRI showed that Broca’s area was highly activated when subjects were processing these two sentences. Wernicke’s area, on the other hand, was more activated when processing sentences that were semantically—but not syntactically—different, such as “The car is in the garage” and “The automobile is in the garage.”

  How is it that Wernicke’s area can so quickly and accurately decide that two semantically different sentences have the same meaning? The answer may lie in two other recently discovered characteristics of Wernicke’s area. One is that the neurons in Wernicke’s area are spaced about 20 percent farther apart and are cabled together with longer interconnecting axons than the corresponding area in the right hemisphere of the brain (Galuske, Schlote, Bratzke, & Singer, 2000). The implication is that the practice of language during early human development resulted in longer and more intricately connected neurons in the Wernicke region, allowing for greater sensitivity to meaning.

  The second discovery regarding Wernicke’s area is its ability to recognize predictable events. A magnetic resonance imaging (MRI) study found that Wernicke’s area was activated when subjects were shown differently colored symbols in various patterns, whether the individuals were aware of the pattern sequence or not (Bischoff-Grethe, Proper, Mao, Daniels, & Berns, 2000). This capacity of Wernicke’s area to detect predictability suggests that our ability to make sense of language is rooted in our ability to recognize syntax. The researchers noted that language itself is very predicable because it is constrained by the rules of grammar and syntax.

  Development of Memory Systems

  Acquiring spoken language would be impossible if the brain did not have some means for remembering what it hears. While the brain is gaining competence in manipulating phonemes and morphemes, memory systems are also developing, giving the child’s brain the equipment it needs to store and recall sounds, morphemes, words, and sentences and their meaning. Moreover, the brain establishes several different types of memory banks to make access to stored information faster and more accurate. It is not a perfect system, but with practice, even young children can use vocabulary and syntax with astounding accuracy. In the next chapter, we will discuss more about memory systems and their importance in learning to read.

  The Components of Speaking and Understanding Language

  Any model for speaking and understanding language has to address the various stages of sound interpretation, beginning with the auditory input and ending with the formation of a mental concept represented by the word or words. Figure 1.9 shows the various neural components that linguistic researchers and neuroscientists believe are required for spoken language comprehension. It is a complex process, but the efficient organization of the linguistic networks that is built up through practice allows it to occur very quickly.

  To understand the differen
t components, let’s take the word dog through the model. After the spoken word dog enters the ear canal, the listener has to decode the sound pattern. In a part of the brain referred to as the word form area, acoustic analysis separates the relevant word sounds from background noise, decodes the phonemes of the word (duh-awh-guh), and translates them into a phonological code that can be recognized by the mental lexicon. The lexicon selects the best representation it has in memory and then activates the syntactic and semantic networks, which work together to form the mental image of a furry animal that barks (concept formation). All this occurs in just a fraction of a second thanks to the extensive network of neural pathways and memory sites that were established during the early years of speaking and listening.

  Notice that the flow of information in this model is from the bottom up and, thus, appears linear. However, feedback from higher to lower levels often occurs. For example, if the lexicon does not recognize the first set of signals, it reactivates the phonological coding component to produce another set before they decay. Likewise, if the semantic network finds no meaning, it may signal for a repetition of the original spoken word to reprime the process. It is important to understand how this process works because, as we shall see in the next chapter, the process of reading words shares several steps with this model of spoken language processing.

  Figure 1.9 This schematic representation shows the major neural components required for spoken language processing. Feedback from higher to lower levels also occurs, as indicated by the arrows on the left.

  SOURCE: Adapted from Dehaene (2009); Gazzaniga et al. (2002).

  LEVELS OF LANGUAGE COMPREHENSION

  Parents speak differently to their children than to other adults. Elementary teachers use different language with their students than with their principal. Speech can be formal, as in the classroom, or informal, as around the dinner table. When young children use informal language, it is often context dependent; that is, the conversation focuses on the immediate situation or activities at hand. On the other hand, formal speech may be more context independent or abstract in that the child may be relating different possible endings to a story. Sometimes people say one thing but really mean something else, and they hope that the listener will catch on to the subtler meaning. These different language forms are a recognition that there are several types and levels of spoken language and of language comprehension.

  Explicit Comprehension

  The most basic type of language comprehension is explicit comprehension—the sentence is clear and unambiguous. When someone says “I need a haircut,” the interpretation is unmistakable. The listener knows exactly what the speaker means and does not need to draw any inferences or elaborate further. Adults tend to use explicit sentences with children to avoid ambiguity. “Eat your vegetables” and “Please be quiet” are clear statements. Whether the child complies, of course, is another story.

  Inferred Comprehension

  A more sophisticated form of language comprehension requires the listener to make inferences about meanings that go beyond what the speaker explicitly said. A principal who says to a tardy teacher, “Our school really gets off to a great start in the morning when all the staff is here by 8:15,” is really saying, “Be on time.” The teacher has to infer the statement’s real intent by reading between the lines of what the principal explicitly said.

  Young children have difficulty with inferred comprehension. If the parent says, “Vegetables are good for you,” the child may not pick up on the underlying intent of this statement—eat your vegetables. Consequently, the child may not finish the vegetables, and the parent may mistake this behavior as disobedience when it is really a lack of inferred comprehension.

  Teachers sometimes use language requiring inferred comprehension when explicit comprehension would be much easier. A teacher who says, for example, “Do you think I should speak if someone else is talking?” may provoke a variety of responses in the minds of the children. One could think absolutely not, while another might hope she would just speak louder than everyone else so the lesson could move along. A few might get the real intent—oh, she wants us to be quiet.

  Context Clues. We discussed earlier how context can be an important clue for determining the meaning of vocabulary words in a sentence. Context can also help with inferred comprehension. A first-grade teacher who is telling her spouse over dinner how crowded her class is and that there are too many students who need special help may just be seeking sympathy. But in having the same conversation with her principal, she is really saying she needs an instructional aide. She never says that explicitly; the principal must infer the teacher’s intent from her statement and the context.

  Children need to develop an awareness that language comprehension exists on several levels. It involves different styles of speech that reflect the formality of the conversation, the context in which it occurs, and the explicit as well as underlying intent of the speaker. When children gain a good understanding of these patterns in speech, they will be better able to comprehend what they read.

  QUESTIONS FOR DISCUSSION/REFLECTION

  • Why does spoken language come so easily?

  • How does a child’s brain detect language sounds from background noise?

  • Does a family’s socioeconomic status really affect a child’s vocabulary growth?

  • What impact does television have on an infant’s cognitive processing and language development?

  • How does a child learn the irregular forms of verbs?

  • Can children tell the difference between explicit and inferred comprehension?

  What’s Coming?

  The child’s brain has now acquired the fundamentals of spoken language. Neural networks are developing rapidly in Broca’s, Wernicke’s, and the word form areas, and every day brings new vocabulary and understanding to the expanding mental lexicon. How will these newly acquired language skills and knowledge help the child accomplish the next major cognitive task: learning to read? All the steps the brain must go through to progress from spoken to written language are unveiled in the next chapter.

  2

  Learning to Read

  I wish you to gasp at not only what you read but at the miracle of its being readable.

  —Vladimir Nabokov, Pale Fire

  Humans are by nature curious animals. From our beginning as a separate species, we constantly explored our environment, deciding what actions would enhance our ability to survive. Our brain evolved a large frontal lobe to help us process incoming information and make those vital decisions. We acquired spoken language, learned how to make tools, hunted for prey, and tilled the fields to increase our food supply. Clearly, we were born with an instinct to learn, and this instinct plays an essential role in our capacity to learn to read. In this chapter, we will explore the latest research on how scientists believe the brain learns to read. This chapter gives you a strong background in the findings from the research, and how those findings can be applied to teaching children to read will be discussed in Chapters 3 and 4.

  Before going any further, it is worth noting that, as adult readers, most of us do not recall how arduous it was for us to learn to read. That may make it more difficult for us to understand why so many children struggle with reading. You may be thinking, “I did it, so why can’t they?” But keep in mind exactly what is in store for these children when they enter kindergarten. Because they have already been speaking for several years, they all arrive with some degree of phonemic awareness, depending on literacy exposure at home. They likely understand the differences in some onset sounds and know that bet has a different meaning than pet, and bat is different from pat. But are they ready for our surprise? Here is the essence of it.

  “Boys and girls, remember those sounds you play with in your head when you speak? Well, do we have a surprise for you! Those sounds can be matched to written symbols (which, by the way, do not exist in the real world—we made them up, and not everyone in the world uses them). You may
know them as the letters of the alphabet. Well, we are going to learn them, but we have just one teensy-weensy problem. There are about 44 sounds in your head, but we have only 26 letters to represent them. So sometimes the same letter can be matched to different sounds, and sometimes the same sound can be matched to different letters. Doesn’t that seem like fun?” Now the girls, many of whom are more advanced in language facility than the boys at this age, are delighted with this challenge. The boys, on the other hand, are asking themselves how they got into this mess and how they can get out of it. Given the anxiety such a situation can raise in beginning readers, it may help to keep this scenario in mind as you continue reading.

  THREE PHASES OF LEARNING TO READ

  In simplest terms, learning to read involves connecting two cerebral capabilities that are already present in young brains: the spoken language networks and the visual recognition circuits. Trying to understand how this connection occurs in the brain has been a challenge for many researchers over the past decades. In 1985, British psychologist Uta Frith developed a three-phase model of how the brain acquires the ability to read (Frith, 1985). Much new research about how we read has emerged since Frith’s model. Reading is far more complicated than we thought just a few years ago. Although the model’s simplicity belies the complex cerebral processes that we now know are involved in learning to read, it nonetheless still provides a useful outline of what the brain experiences (Dehaene, 2009). Here is a brief summary of the three phases:

  • The first phase can be considered a pictorial stage, when the child’s brain photographs words and visually adjusts to the shape of the alphabet’s letters.