Bonnie J. F. Meyer

Effects of Discourse Type on Recall

Discourse can be organized in different ways; four of these ways are comparison, problem/solution, causation, and a collection of descriptions. These four discourse types correspond to schemata that vary in their organizational components; these differences were expected to result in differences in processing text. The more organized discourse types of comparison, problem/solution, and causation were predicted to yield superior recall of information than when this same information was cast as a collection of descriptions about a topic. The data from two studies support the hypothesized facilitation of the more organized types of discourse and have implications for understanding memory and writing instructional materials.

Effects of Structure Strategy Training and Signaling on Recall of Text.

Younger and older adults participated in 9 hr of either structure strategy training, interest strategy training, or no training Both trained groups reported positive changes in reading, but only the structure strategy group showed increased total recall from a variety of texts and an informative video. Structure strategy training increased the amount of information remembered as well as recall of the most important information. This training affected the organization of recall and was critical for producing readers who could use the structure strategy consistently across a variety of expository texts. In addition, it helped learners use signals in text more effectively. There was an additive effect of training plus signaling for use of the structure strategy consistently across five passages. The strategy switch hypothesis was supported, indicating that signaling affects encoding rather than retrieval processes. The findings have implications for both reading and writing.

Discourse comprehension and problem solving: decisions about the treatment of breast cancer by women across the life span.

The 1st study examined the decision making and prose comprehension of 94 women interacting with an authentic, unfolding health scenario about breast cancer. The 2nd study involved questionnaire data focusing on the decisions made by 75 women who had been diagnosed with breast cancer. Two major findings emerged from this laboratory and survey research. First, older women sought less information when making treatment decisions about breast cancer. However, the outcome of those decisions were equivalent to those of younger women. Second, older women made these decisions faster than younger women. In addition, treatment decisions were related to prose processing, the type of information underlined as important while reading as well as the type of information remembered about various treatment options.

Text Coherence and Readability.

The goal of matching texts to readers can be aided by using readability formulas in concert with considering text, task, reader, and strategy variables. For example, signaling makes sentences longer and readability scores soar, but eases readability for readers employing the structure strategy and l

Effects of Structure Strategy Instruction Delivered to Fifth-Grade Children Using the Internet with and without the Aid of Older Adult Tutors.

The authors assessed the impact of using the structure strategy as a base for an intergenerational Internet tutoring program in which older adults, with strategy training, provided Internet-based tutoring for 5th-grade students learning the strategy through an instructional Web site. Students were randomly assigned to 1 of 3 groups: structure strategy with tutors, structure strategy without tutors, and control. Both tutors and children in the structure strategy group with tutors increased strategy use, total and main idea recall, and self-efficacy. Program effects were apparent months after instruction. Posttest performance was related both to careful completion of Web lessons and amount of tutor feedback and content-related questions. Findings have implications for learning from computers, intergenerational tutoring, and reading instruction.

The interaction of reader strategies and the organization of text

A model is proposed for the interaction between reader and text in arriving at an identiflcation ofthe overall logical organization to be used by the reader in processing and rernembering the text. Different emphasis patterm provided by the writer for a text on railroads are shown to result in different organizational pattems and types of Information recalled by adults. Recallprotocoh written by adults with different reading strategies are examined for different versions of this text. In addition, other subjects reported their expectations for these versions of the railroad text when asked to read and comment on the text in a sentence-by-sentence manner; the reported expectations show similanties with the model and the recall data.

Reading Research and the Composition Teacher: The Importance of Plans.

A plan consists of a goal and steps to achieve that goal. Plans are obviously a central component of the processes of communicating and understanding. A writer must evolve some general plan of what to say; and a reader must somehow be able to follow the plan along. However, there has so far been only a small amount of work which relates the psychology of planning to the act of writing.2 It might be helpful to consider some research on reading as a potential means of exploring the role of planning. Reading research has demonstrated the important role played by the mental representation formed in the mind of human beings who read texts. This representation cannot be the text itself; that is, it cannot be a linear series of individual words as presented on the page. Although this mental representation is important for many tasks, such as writing, reporting, summarizing, commenting, and so forth, readers will seldom be able to provide a wordfor-word account of the original text. Therefore, a better understanding of what this mental representation is and how it forms in long-term memory should help a writer plan texts which enable their readers to create representations that better match the writers purpose in the communication. For a writer, the plan is like a set of directions about how to present ones materials. I shall be concerned here with three important functions that writing plans have. In their topical function, they help a writer conceive and organize main ideas on a topic. In their highlighting function, they help the writer show the reader how some ideas are of greater importance than others. In their informing function, they help the writer see how to present new knowledge while keeping readers aware of the old.

Integrating different types of information in text

This investigation was designed to answer two major questions: (a) does height of information in the content structure of text affect the probability of integrating the information, and (b) is it possible to differentiate between integration that occurs during acquisition and integration that takes place at retrieval. Results of an inference verification task suggested that information high in the structure is more likely to be integrated than information low in the structure. These effects were observed in two instruction groups (learn and read) and in two presentation modes (separate and consecutive). In addition, premises occurring together in the text (consecutive presentation) were found to promote faster correct decisions than premises occurring separately. These verification time results were interpreted as evidence that structural integration (i.e., integration at acquisition) is more likely when premises occur together in an information source than when they occur separately.

Why older adults make more immediate treatment decisions about cancer than younger adults.

Literature relevant to medical decision making was reviewed, and a model was outlined for testing. Two studies examined whether older adults make more immediate decisions than younger adults about treatments for prostate or breast cancer in authentic scenarios. Findings clearly showed that older adults were more likely to make immediate decisions than younger adults. The research is important because it not only demonstrates the consistency of this age-related effect across disease domains, gender, ethnic groups, and prevalent education levels but begins to investigate a model to explain the effect. Major reasons for the effect focus on treatment knowledge, interest and engagement, and cognitive resources. Treatment knowledge, general cancer knowledge, interest, and cognitive resources relate to different ways of processing treatment information and preferences for immediate versus delayed decision making. Adults with high knowledge of treatments on a reliable test tended to make immediate treatment decisions, which supports the knowledge explanation. Adults with more cognitive resources and more interest tended to delay their treatment decisions. Little support was found for a cohort explanation for the relationship between age and preference for immediate medical decision making.

Text Processing Variables Predict the Readability of Everyday Documents Read by Older Adults.

A model is presented to predict the readability of documents encountered by older adults. The documents studied are contained in the Educational Testing Services Test of Basic Skills (1977 edition) and require readers to answer questions about charts (e.g., bus schedules), labels (e.g., plant spray labels and prescriptions), and forms (e.g., tax forms). The components of the model came from theoretical and empirical work on discourse processing and include such factors as discourse structure, emphasis, and position of an answer in a linguistic analysis of the everyday document.A sample of 482 adults from 52 to 93 years of age took the everyday problems test as well as a psychometric ability battery. The correlation was .54 (p < .01) between the readability scores for test items predicted by the model and the percentage of older adults correctly answering those items. In addition, the more difficult test items as identified by the model were correlated more highly with fluid intelligence abilities (figural relations and induction), crystallized intelligence abilities (vocabulary, experiential evaluation), and with memory span.