Tag Archives: cognition/learning/understanding

Chapter 5: “Reading Comprehension” Summary, Implications and Discussion Questions

Summary

• There are three levels of meaning representation: we extract ideas from sentences, we connect the ideas across sentences, and we build a general of what a text is about.
• At each level, there are rules about how meaning is made—rules that can be expressed independent of the content of ideas. But it’s also true that meaning influences how we comprehend text at each of the three levels.
• Many readers set a low criterion when assessing whether they understand a text. They do not coordinate meaning across sentences, and thus fail to notice texts that contain contradictions.
• Teaching reading comprehension strategies that require the coordination of meaning across sentences does improve comprehension, but it seems to be a one-time improvement rather than a technique that can be practiced to continually improve reading comprehension.

 Implications

• The prominent role that background knowledge plays in reading comprehension ought to make us think differently about reading tests. We might think that reading tests provide an all-purpose measure of reading ability. But we’ve seen that reading comprehension depends heavily on how much the reader happens to know about the topic of the text. Perhaps then, reading comprehension tests are really knowledge tests in disguise. The Cunningham and Stanovich experiments discussed in the text supports that idea.
• Teaching reading is not just a matter of teaching reading. The whole curriculum matters, because good readers have broad knowledge in civics drama, history, geography, science, the visual arts, and so on. But the conclusion is not just “the curriculum has a lot of stuff in it.” Sequence matters too, because students can only encounter so much new content at one time. They need to know most of (but not everything) the writer assumes the reader knows. Such precision in what students should know before they tackle a text calls for careful planning.
• Telling students to make inferences or teaching them reading comprehension strategies provides a one-time boost to comprehension. That implies that, when they are taught, they have no impact on some students. Students who still struggle with fluency are not able to use these strategies may be larger than is commonly appreciated, but applies to only a subset of students.
• Students from disadvantaged backgrounds show a characteristic pattern of reading achievement in school: they make good progress until around fourth grade, and then suddenly fall behind. The importance of background knowledge to comprehension gives us insight into this phenomenon. Reading instruction in the early grades concerns decoding, and so reading tests are basically tests of decoding ability. Kids from wealthier homes in fact do a bit better on these tests, but poorer children are still doing okay. But around fourth grade most children can decode fairly well, and so reading tests place greater weight on comprehension. The disadvantaged kids have not had the same opportunities to acquire the vocabulary and background knowledge needed to succeed on these tests and so their performance drops significantly.

 Discussion Questions

• Readers usually forget the particular phrasing of what they read quite soon after reading it. Does that mean it doesn’t matter much.
• Even struggling readers seem to do a good job of coordinating meaning when they are watching a movie; they follow the plot and put together an effective situation model. Why are movies different than texts? Is there anything to be learned from movies that might help a student’s reading comprehension?
• When we learn that comprehension depends heavily on background knowledge, that naturally invites the question: “Which knowledge should children learn?” (Note that in the experiment on the relationship of background knowledge and reading, the researchers referred to knowledge as “cultural literacy.” Whose culture does that literacy refer to?) Before addressing that question, I invite you to consider the factors that ought to contribute to your answer.
• I noted that making inferences is sometimes possible when you lack background knowledge and vocabulary the writer assumed you have, but that doing so is mentally taxing. Much of the reading expected of students (especially in the later elementary grades and beyond) is difficult. Its’s not only difficult in terms of vocabulary and knowledge; they read texts with more complex structures, texts that convey abstract and subtle ideas, and they are asked to put these texts to new purposes, like understanding the author’s technique. In short, students don’t do the type of reading where comprehension is smooth and there’s an opportunity to get lost in the story. They mostly read in situations where reading feels like work. What impact do you think that has on students’ attitude toward reading? Do they confuse leisure reading with the reading they do for school? If so, what might be done to disabuse them of that notion?
• The account of the fourth-grade slump offered above suggests that disadvantaged children perform poorly on reading tests because they lack the background knowledge that their wealthier peers have—knowledge that is required to comprehend the texts appearing on reading tests. What texts would these children read well, likely better that middle-class children? Should such texts appear on reading tests?

 Excerpted from: Willingham, Daniel T. The Reading Mind: A Cognitive Approach to Understanding How the Mind Reads. San Francisco: Jossey-Bass, 2017.

Summary

• The meaning of a word is very sensitive to the context in which it appears.
• Researchers model the organization of word knowledge with simple features of meaning that are densely interconnected.
• If you don’t know the meaning of a word, sometimes (but not always) you can deduce it from the context. But people are not eager to do a lot of this work, because it’s difficult and it interrupts the flow of reading.
• Words defining other words sound circular. This problem may be partially solved through the use of grounded representations.
• It’s not only important to know a lot of words (breadth) but for the words you know to have many connections, and for those connections to be strong (depth).
• New words are used bit by bit, through exposure.

Implications

• Looking words up in a dictionary will be of limited use—not useless, but, but we must acknowledge that it will be just one context in which to understand the word’s meaning, and it’s possible that the student will misunderstand the definition. Explicit instruction of new words is more likely to be successful the way teachers usually implement it, with multiple examples and with the requirement that students use each word in different contexts. There is a good evidence that students do learn vocabulary this way.
• In addition to consistent vocabulary instruction, teachers can make it more likely that students will learn words they encounter in context. They can give students pointers that will help them use context for figure out an unfamiliar word. For example, students can learn to use the clues in the sentence about the unknown word’s part of speech, to use the setting described in the text to constrain the word’s meaning, and to use the tone of the text to help constrain meaning.
• Students are also better able to deduce the meaning of unfamiliar words if they have had some instruction in morphology. The definition of a morpheme is a unit of language that is meaningful on its own, and that cannot be further divided. Thus “dog” is a morpheme. The really interesting morphemes are the all-purpose ones that can be added to words—usually as prefixes of suffixes—to change their tense or inflection, or meaning. For example, the prefix “super” means over, the suffix “like” means having the characteristics of, and so on.
• Though important, direct instruction cannot account for all of children’s vocabulary learning. That’s because someone who stays in school up to age 18 may know as many as 20,000 word families. (Word families meaning that “talk, talks, talked,” and other obvious derivatives count just once.) If children are learning about a thousand words each year and there are about 36 weeks in a typical American school year, students would need to get instruction in about 28 words each week. That seems high, especially given that children in early elementary grades often don’t get explicit vocabulary instruction.
• Much of the vocabulary that we know is not the product of explicit study, but was learned incidentally, either through conversation or reading. We would expect, however, that reading will be more useful for learning new words than conversation will be, because writers more frequently use unusual words than speakers do.
• The difference between writing and speaking in terms of the richness of vocabulary it offers becomes really important as vocabulary grows. For the newborn, adult speech offers plenty of novelty, but they will obviously be most likely to learn the words that people around them most frequently use. So one way to boost student vocabulary is to prompt teachers and parents to use more unusual vocabulary words, and some research indicates that helps. But that may be hard to implement for older children who already know quite a few words. For them, moderately challenging reading material will be the main way they will encounter new words. And because a single instance of exposure is not enough to learn a word—learning is, after all, gradual—it would seem that the injunction to students must not just be “Read!” but “Read a lot.”

Discussion Questions

• I cited studies showing that people are reluctant to read texts that use unfamiliar words. How much do you think it helps to read and electronic text with a feature whereby touching a word brings up a dictionary definition?
• For older children, what is the responsibility of teachers of subjects other than English Language Arts to provide exposure to new vocabulary? Should it just be vocabulary particular to their subject, or broader? It sounds as though it would be useful for teachers in different subjects to coordinate to be sure that students practice the same words in different subjects, but is that really practical?
• Teachers can provide varied contexts in which students can encounter the same word, so that the meaning representation will be precise. How can someone trying to improve their vocabulary do that on their own?
• We might propose that teachers use richer vocabulary with their students. How should this be implemented? Should teachers derive a list of words that they try to use over some period of time (say, a month) to ensure students hear repetition of these words? Or should teachers just make a mental note to use words they know will challenge students, and to provide on-the-fly definitions?
• Do you think most children have good morphological knowledge? How about adults? How about teachers? If a school or district were to set the goal of improving students’ morphological knowledge, what would be required?

Excerpted from: Willingham, Daniel T. The Reading Mind: A Cognitive Approach to Understanding How the Mind Reads. San Francisco: Jossey-Bass, 2017.