Chapter 25: Building Representation of Informational Text: Evidence From Children's Think-Aloud Protocols

In the article, “Building Representation of Informational Text: Evidence From Children’s Think Aloud Protocols, Natalie Cote and Susan Goldman investigate strategies children use to comprehend unfamiliar expository text. One such strategy is the think-aloud process which allows researchers to see the processes and thoughts children have while they process expository text. Research has shown the think-aloud method helps the reader to remember text and solve problems better than if the method was not used.

Several assumptions about the processing involved in constructing cognitive representations of text:
1. Sequential processing of information from text
2. A connection of new information and prior knowledge are brought together in working memory and then are stored as a partial representation
3. During the reading process, readers build different representation simultaneously. One such example is making a hypothesis about the text while taking in information directly from the page.
4. Relationships between two units of text information, or between information from the text and information received from long-term memory, can only be detected or generated if the pieces of information are active in working memory at the same time
5. Working memory has a limited capacity

For the study, the participants consisted of 16 sixth grade children ranging in ages 11 to 13. The students were asked to read four passages that contained material new to them. Of the four passages, two were on the more-familiar topics of and two were on the less-familiar topics. Students read these articles at two different times in the year: at mid-year and at the end of the year. At each session, they read two articles, one containing a familiar topic and one containing a less-familiar topic. During these sessions the student were instructed to use the think-aloud strategy so researchers could hear what the children were thinking. This allowed the researchers to know what was hard, what was understood or misunderstood which in turn shows researchers what strategies the children used in order to comprehend the test they read.

The two texts that contained unfamiliar information were about metabolism and hybrids. The Metabolism text is about factors that affect metabolism or metabolic rate and gives some explicit information about metabolic rate in separate paragraphs (such as food, climate, activity, and genetic inheritance). In order to understand how metabolic rate, energy source and energy requirements are related, the reader must integrate information from several text paragraphs and prior knowledge about the body. Readers who do not possess prior knowledge about this topic will be heavily dependent on the explicit information in the text for making the inferences and connections necessary for understanding. The second text is about hybrids and their helpfulness to Mother Nature. While the Metabolism passage implicitly expressed a complex relationship among a concept and the factors affecting it (causing a reader to use inference), the Hybrids text did not. Instead, this passage contains a number of situations and examples of hybrids and how they are helpful, which is information not requiring the use of inference.
Cote and Goldman studied the coding and analyses in terms of the types of five comprehension and reasoning strategies and processes that individuals use when coming across difficult text. Five categories emerged. They are self-explanation, monitoring, paraphrasing, predicting, and associations.
· Self-Explanations-Within this category, students work on making meaning out of the sentence from prior knowledge and contextual cues found in the text.

· Monitoring- Confirmation of comprehension (whether accurate or not) occurs in this category. These statements signify an evaluation of comprehension where students are comparing their sense of understanding to some internal criterion of satisfactory meaning.

· Paraphrasing- This is used to maintain information in working memory. By changing words while keeping the same context, this allows students to connect new information to any prior knowledge, they possess about the topic.

· Predictions- Statements that tell what students expect to find further in their reading. If confirmation of the prediction occurs, the new information helps to create a logical connection to prior information.

· Associations- Statements that do not aid in comprehension of the focal sentence, but do provide a link to the student’s prior knowledge.

Cote and Goldman found that out of the five categories presented earlier, self- explanation and monitoring protocol were most used by students. They did not often call upon the other strategies and when they did, they did not have much success in using them. For example, students who used prediction as a strategy often struggled within the information in the expository text to predict successfully what would come next.

The researchers did an analysis of types of protocol events to understand how children capture reasoning and comprehension strategies while they construct a representation of the text. During a second analysis, Cote and Goldman were looking at the degree to which children were making connections among elements in the text and between text elements and prior knowledge. Three categories of reinstatements emerged; they are according to the source of the reinstated information.

Physical Reinstatements- Computer records allow the researchers to see the order in which students accessed and read the sentences in the passage. This allowed the researchers to be able to distinguish between prior texts that were physically reinstated or use of mental reinstatement from memory Consistent findings show many of the children spontaneously and selectively reinstated previously read sentences by re-exposing them physically.

Mental Reinstatements- Two types of mental reinstatements emerged during the think –aloud protocols.

Mental–Text- reinstatements made by the reader where previous information presented in the text were mentioned without physically looking at the material again.

Mental–Student Generated- reinstatements of prior knowledge about the text in which the student verbalized on earlier focal sentences are connected to the physical reinstatements.

The researchers feel that it would be to hard to analyze the individual protocols, instead to capture the nature of an individual student’s activities, four categories were developed.

· Successful Knowledge Building- students put an extreme effort into this type of processing. They were attempting to construct a coherent model of the central concepts and relations. To help with this, students engaged in activities such as cause/effect explanations, elaborations, and cross-text integrative inferences that identified the macrostructure of the text.

· Less-Successful Knowledge-Building- While these students also made an effort to understand the central concepts and relations expressed by the text, but their actions concerning monitoring comments and requests for information show that they were less successful in making a coherent representation. They were frequently seeking information from the text that the text did not provide.

· Text-Focused Processing- These students showed varying degrees of effort through paraphrasing, interpreting, questioning, and other activities but these were generally in response to single sentences rather than constructing a global representation.

· Minimalists- Students who gave verbalization on half or fewer of the text’s sentences resulting in protocols too poor to support a reliable analysis of their text processing. Possible explanations for this are lack of motivation or being familiar with the context.

The researchers gleamed from the categories that the more knowledge building their approach is; the more engaged they are in reinstatement and self-explanations. Thus making the think-aloud approach a useful and meaningful approach for children to use in trying to comprehend expository text.