How we reach detailed understanding: 5 ideas
- Top-down processing
When readers comprehend text using a top-down process they use their “previous knowledge and experience of the world” to understand what lies in the text. That means that comprehension “resides in the reader.” Top-down processing is about building meaning and the interaction between the reader and text. Every individual word might not be understood or recognized. Rather the reader reconstructs meaning based on predictions and looks in the text for confirmation. He then relates the information to his existing knowledge.
Smith says that “comprehension depends upon prediction” He claims that we have a system in our brain; “a personal theory of the world” and we understand the world around us based on this theory. This happens when we read: we predict what will come next based on what we already know and ask questions. Smith says that “comprehension is getting these questions answered.”
- Bottom-up processing
When readers comprehend text using a bottom-up process they “decode a series of written symbols into their aural equivalents.” Understanding is reached by first processing the smallest parts of language; letters, morphemes, grammatical features, and then sentences and longer texts. This view of processing text is often criticized in the literature as being unrealistic. Nunan argues that due to the limits of working memory bottom-up processing would cause readers to “forget the beginning of a sentences (and perhaps even a word) before they have reached the end.”
- Interactive model
It is argued that neither bottom-up nor top-down models realistically describe what we do when we process text. Stanovich proposes a third model called an “interactive-compensatory model” suggesting that we understand text by using information from various sources. (Nunan 1991:67) Those sources of existing knowledge are real world knowledge or linguistic knowledge. (O’Malley and Chamot 1990:37) We use our knowledge of phonology, lexis, syntax, semantics, and discourse as well as our knowledge of the world to understand text and lack of knowledge in one area can be compensated by knowledge in other areas. (Nunan 1991:67) Proponents of this model argue that “effective processing of text requires the use of both top-down and bottom-up processing.” (O’Malley and Chamot 1990:37
- Purpose for reading
Readers have a goal when they process text and they “read to find information that they are interested in using in some meaningful way.” (Schramm 2008:235) This purpose for reading might be for pleasure or to take action or make a decision. Ellis says that the purpose is “determined by the text itself.” (2003:40) Why we read also determines how we read. We use different skills such as skimming, and prediction; we scan texts to find specific information; we infer meaning and attitude, and we recognize the way texts are organized. (Harmer 1991:144-145) Schramm says that “good readers take action to secure comprehension” based on their purpose for reading. (2008:237)
- Reading fluency
Goodman says that “efficient reading tends to be relatively fast” because effective readers make sense of text with “the least amount of time, effort, and energy.” (1996:92) This means that readers do not recognize every individual word in the order they appear in a text because it is an inefficient model of processing; it distracts the reader from his goal: constructing meaning.
Learner problems
Transferring strategies from L1 to L2
In my experience, upper intermediate level learners do not always transfer reading skills they have when processing text in L1 to when they read in English. My learners are all literate and do not need to learn to read but they often are not aware of how the reading skills they use when processing text in Catalan or Spanish can be used as strategies when comprehending text in English.
Skimming and scanning
I have often observed reluctance from my learners to skimming texts quickly to get a global idea of what a text is about and I see this reluctance coming from weaker students who process text slower than stronger upper intermediate level learners. These same students seem unable to adapt their strategies to the purpose for reading. Problems also arise when I get my learners to scan texts for specific information. Weaker students focus on moving through the text one word at a time and take longer to find meaning in the text related to the task.
Elaboration and prediction
The prior knowledge learners use to comprehend text is fundamental (Nunan 1991:66) but I have observed how this prior knowledge can sometimes prevent my learners from understanding the meanings in texts because they do not always adapt their predictions accordingly. In my experience some upper intermediate learners “ignore the limitations of the text” (Aslanian 1985:20) and either draw conclusions that are contradictory to what the writer intended or draw conclusions from the text based on meanings that are not present in the text. I have observed my upper intermediate learners over-rely on top-down processing when texts communicate ideas that are contradictory to the learners’ knowledge of the world or communicate ideas that go on to contradict previous ideas expressed by the writer. I find this happens when I use authentic texts from pseudo-science magazines that often contradict common knowledge and challenge readers’ notions of the world. I have observed upper intermediate learners who are unable to adjust “their initial interpretation of such texts.” (Ellis 2003:42)
Unknown lexis
When my weaker upper intermediate level learners are confronted with words they do not recognize while reading, they do not always use prediction or inference skills to extract meaning. Instead they either ask me or a classmate sitting next to them what the word means or they get up and get a dictionary. I find this behavior counterproductive if learners have not first asked themselves if knowing what that particular word means is necessary to achieve their reading goal. My upper intermediate learners do not always use context clues to predict or infer the meaning of new words and thus stop interacting with the text. This is another example of learners not transferring skills from L1.
Inferring meaning
I find that my upper intermediate level learners have difficulties inferring meaning if they are required to understand the writer’s attitude or opinions expressed in text. They find it difficult to synthesize meaning from different parts of text to comprehend meaning. Weaker learners tend to lack the ability to read between the lines to construct meaning by making inferences about tone or humor. This can also be challenging for my stronger upper intermediate level learners as well. When I use authentic short stories in class and get my learners to discuss how the narrator feels about certain characters or comment on the narrator’s tone, my students often find it very demanding.
