Prose Retention

views updated


As with memory for other kinds of verbal material, memory for prose is a complex function of various factors. Some of the more important factors include 1. the kind of reading or studying activity in which the learner engages while processing a text; 2. the particular type of prose that is being remembered; 3. the kinds of information and events that intervene between the initial reading of the text and when remembering is attempted; and 4. the way in which memory is assessed (e.g., through recall, short-answer, multiple-choice, or true/false tests; see Figure 1). For clarity of presentation, this entry considers each of these factors separately. In reality, however, each of these factors does not operate in isolation; instead, they work together to influence memory for prose.

Influences of Encoding and Reading Processes

One of the dominant themes in contemporary memory research is that memory benefits to the extent that the material is processed for meaning. The same holds true for prose. For example, research has shown that readers who have been instructed to count the number of four-letter words in a passage do not remember the passage as well as readers who have been instructed to rate the paragraphs for ambiguity (Schallert, 1976). Similarly, if a text is written so that its topic is obfuscated, the degree to which meaningful processing can take place decreases, and this text is recalled less well than a text in which the topic is clearly stated (e.g., through an informative title or by reference to the topic throughout the text; Bransford and Johnson, 1972). The lesson here is straightforward: A prerequisite for remembering prose is extracting the meaning that is being conveyed.

On the assumption that readers typically are extracting meaning as they read, one can further analyze the components of that meaning which are retained. It is typically the case that people do not remember prose verbatim (except with concerted effort, and even then it is difficult; Sachs, 1967). Instead, prose retention is characterized by memory for the ideas that are captured by the particular words used. Further, memory for the ideas in a text generally conforms to the following function. The main or most important ideas are recalled best, followed by ideas of intermediate importance, with unimportant details usually recalled least well.

The reader's point of view or frame of reference partly determines the importance of the ideas contained in the text. Because an idea's level of importance determines its memorability, very different patterns of memory for the same text can be observed across readers who adopt different points of view. For example, a prose passage about a household and its daily activities will be remembered differently by a reader who adopts the perspective of a prospective home buyer compared with a reader who adopts the perspective of a burglar appraising particular properties for a clandestine visit (Fass and Schumacher, 1981). The differences in recall for these two readers will directly reflect the fact that the particular ideas in the passage differ in importance, depending on the perspective adopted while reading the passage.

Consistent with findings for other types of verbal material, good prose retention depends on reading activities that engage attention toward understanding the individual ideas in a text and how these ideas are organized or interrelated. Such activities can include trying to explain the information to oneself, thinking about causes or results, relating the information to prior knowledge, and creating linkages between the presented ideas. Numerous factors influence the degree to which such reading activities will be engaged. Not surprisingly, readers seem to become more engaged for text rated as more interesting than for less interesting text, with more interesting text being better remembered (Son and Metcalf, 2000). Individual readers vary as well, with some readers spontaneously engaging in more active processing of the text than others, thereby increasing retention for the prose material. Individuals can be instructed or encouraged to use selected reading or study strategies (e.g., self-explanation, outlining) designed to stimulate greater comprehension and organization of the information in the text. These reading strategies generally improve overall recall of factual (expository) text (McDaniel and Donnelly, 1996; Slater, Graves, and Piche, 1985).

One difference between prose memory and memory for word lists is that prose allows special forms of organization that are not operative with simpler materials. The varied kinds of organizational factors that can be influential in prose memory are best revealed by considering how the genre of the prose affects remembering.

Prose Organization and Prose Type

Generally, narrative prose (e.g., a story) is better remembered than expository prose (e.g., an essay or article). The superior memorability of narratives relative to expositions is probably due in large part to readers' better ability to organize and interrelate the ideas expressed in a narrative. There are a number of differentiating factors between narratives and expository prose that could underlie the extent to which the ideas expressed in the text are organized by the reader. Though it may ultimately turn out that not all of these factors are important in accounting specifically for the memory differences between narratives and expository text, these factors do seem to play a role in prose memory.

One factor is that readers presumably activate relevant structures of world knowledge when attempting to understand a text. These structures are usually called schemata (singular, schema). In general, schemata are believed to influence the degree to which incoming text information is organized. Text content for which readers have a schema is believed to be better organized and elaborated, and thus better remembered, than text content for which readers have a poorly articulated or sketchy schema. For instance, researchers have shown that a narrative about a fictitious baseball game is better recalled by readers knowledgeable about baseball (i.e., readers who have a baseball schema) than readers who are not knowledgeable about baseball (Spilich et al., 1979).

In contrasting narratives and expository text, it is arguably the case that narratives have content for which most of the readers in the culture have relevant schemata, whereas for expository texts, the content articulates with schemata that only a subset of readers might possess. Narratives are about people or other animate objects, their goals and conflicts, and their other activities—concepts for which readers have schemata. Narratives frequently entail events that, if not directly experienced by the reader, relate to basic cultural values and goals (e.g., finding happiness, wealth, and love). The influence of schemata in organizing and guiding memory of narratives can be observed when readers try to recall narratives from a culture with which they have little familiarity. Recall of such narratives is characterized by the omission and transformation of ideas and details in the narrative that do not fit with the conventional schema of the reader's culture (Bartlett, 1932).

Prose can also be organized by linking the elements in terms of causal relations. Causal relations capture relations between events (described in a text) that involve motivation, psychological causation (e.g., greed causes certain behaviors), and physical causation (e.g., rain causes things to become slippery). Understanding a narrative seems to involve the formation of a sequence of causal relations into a chain that links the opening of the narrative to its outcome. In remembering a narrative, people tend to remember those parts of a narrative which fit into the causal chain and forget the parts which are not incorporated into it (Fletcher and Bloom, 1988). More generally, better memory for narrative prose may occur because causal sequences are more clearly defined in narratives than in expository passages.

A third factor that contributes to prose organization is the reader's knowledge about the structure of conventional text forms. That is, when reading a fairy tale, regardless of the particular content, one expects to see certain components: a setting, an initiating event posing a problem, an attempt at its solution by the protagonist, and a consequence. This knowledge presumably allows the reader to better organize the information found in the passage.

In contrast, expository prose does not seem to entail a consistent structure, thereby reducing the extent to which the reader has a prestored structure that facilitates organization. Late-twentieth-century research indicates, however, that if readers are explicitly trained to notice and utilize some conventional expository forms (e.g., argumentative structures), then memory for such material is improved. Further, this positive effect is obtained for older as well as younger adults (Meyer, Young, and Bartlett, 1990).

Remembering Prose over Time

The organizational processes outlined previously exert greater influence as the interval between reading and attempted remembering increases. If recall is attempted within minutes after reading, then information that does not fit an initial perspective or schema can be recovered. If recall is delayed for at least one day, however, then recall of information that does not fit the schema may drop substantially, whereas information that fits the schema will still be well recalled (Fass and Schumacher, 1981). Similarly, in some cases study strategies designed to increase organizational processing of a text (e.g., outlining) will not improve memory for a text (relative to reading alone, with no studying) when recall is tested immediately, but will increase recall when testing is delayed for several days (Einstein et al., 1990). This increasingly beneficial mnemonic effect of organization with longer retention intervals is another aspect of prose memory that parallels memory for simpler verbal material.

Another factor that affects memory for prose is information that the learner encounters between reading a text and trying to remember it. Such intervening information can have facilitative as well as interfering effects on retention. To illustrate both kinds of effects, consider a biography about a poet named Susan. In the biography it is stated that Susan's father was a servant who died of diphtheria when Susan was five years old. After reading this biography, another biography is encountered in which Ann's blacksmith father dies of lung cancer when Ann is two years old. Research has shown that readers who are given the two biographies and are then asked to recall the first biography, recall the theme of the first biography (e.g., "the main character's father died when she was young") better than readers who are not given the second biography. But readers given the second biography do not recall the specific details of the first biography (the father's occupation, what he died of, and when) as well as the readers not given the second biography (Bower, 1974).

Perhaps one of the more interesting features of prose recall is that intervening information, or even a long retention interval, can promote reconstruction in remembering. Reconstruction refers to the finding that memory of a text can be distorted by the inclusion of information consistent with the theme of the text but not actually mentioned in it. Reconstruction also includes alteration of information that was in the text to bring it more in line with information encountered subsequent to the text or with readers' schemata that were activated in comprehending the text. For instance, in one study (Spiro, 1980) participants read a narrative about two college students who started dating seriously but disagreed about their desires to have children. After the participants read the text, the experimenter casually mentioned that the students had ended up getting married. Several weeks later, the participants attempted to recall the narrative. Their recall included reconstructions and distortions consistent with the new information mentioned by the experimenter but not actually in the narrative (e.g., "The students underwent counseling to correct the major discrepancy").

Prose recall need not always be reconstructive. If the reader expects a memory test on what has been read, then recall is more reproductive. That is, details are recalled more accurately, and there is very little inclusion of extra information not actually in the text. Also, in situations in which the reader's initial interpretative schema is invalidated at the time of recall, recall becomes more reproductive (Hasher and Griffin, 1978).

Assessing Prose Retention

In discussing prose retention, much of the focus has been on remembering as evidenced in recall. The influence of many of the aforementioned factors changes, however, if remembering is assessed with recognition memory tests. In recognition tests, the learner is presented with facts stated in the text and facts not stated in the text, and must determine which facts were presented and which were not. Common kinds of recognition tests include true/false and multiple-choice tests. In general, if prose memory is tested with recognition tests, the robust influence of organizational factors like those discussed earlier can be mitigated or eliminated. Thus, recognition is the same regardless of whether the text is written so that its overall theme can be clearly identified. Further, information that does not fit an encoding schema is recognized as well as information that does fit it, even if memory is tested one week after the text was read (McDaniel and Kerwin, 1987). Thus, as is the case with simple verbal materials, organizational factors play a more prominent role when prose retention is measured with recall than when measured with recognition.



Bartlett, F. C. (1932). Remembering. Cambridge, UK: Cambridge University Press.

Bower, G. H. (1974). Selective facilitation and interference in retention of prose. Journal of Educational Psychology 66, 1-8.

Bransford, J. D., and Johnson, M. K. (1972). Contextual prerequisites for understanding: Some investigations of comprehension and recall. Journal of Verbal Learning and Verbal Behavior 11, 717-726.

Einstein, G. O., McDaniel, M. A., Owen, P. D., and Coté, N. C. (1990). Encoding and recall of texts: The importance of material appropriate processing. Journal of Memory and Language 29, 566-581.

Fass, W., and Schumacher, G. M. (1981). Schema theory and prose retention: Boundary conditions for encoding and retrieval effects. Discourse Processes 4, 17-26.

Fletcher, C. R., and Bloom, C. P. (1988). Causal reasoning in the comprehension of simple narrative texts. Journal of Memory and Language 27, 235-244.

Hasher, L., and Griffin, M. (1978). Reconstructive and reproductive processes in memory. Journal of Experimental Psychology: Human Learning and Memory 4, 318-330.

McDaniel, M. A., and Donnelly, C. M. (1996). Learning with analogy and elaborative interrogation. Journal of Educational Psychology 88, 508-519.

McDaniel, M. A., and Kerwin, M. L. E. (1987). Long-term prose retention: Is an organizational schema sufficient? Discourse Processes 10, 237-252.

Meyer, B. J. F., Young, C. J., and Bartlett, B. J. (1989). Memory improved: Reading and memory enhancement across the life span through strategic text structures. Hillsdale, NJ: Erlbaum.

Sachs, J. (1967). Recognition memory for syntactic and semantic aspects of connected discourse. Perception and Psychophysics 2, 437-442.

Schallert, D. L. (1976). Improving memory for prose: The relationship between depth of processing and context. Journal of Verbal Learning and Verbal Behavior 15, 621-632.

Slater, W. H., Graves, M. F., and Piche, G. L. (1985). Effects of structural organizers on ninth-grade students' comprehension and recall of four patterns of expository text. Reading Research Quarterly 20, 189-202.

Son, L. K., and Metcalfe, J. (2000). Metacognitive and control strategies in study-time allocation. Journal of Experimental Psychology: Learning, Memory, and Cognition 26, 204-221.

Spilich, G. J., Vesonder, G. T., Chiesi, H. L., and Voss, J. F. (1979). Text processing of domain-related information for individuals with high and low domain knowledge. Journal of Verbal Learning and Verbal Behavior 18, 275-290.

Spiro, R. J. (1980). Accommodative reconstruction in prose recall. Journal of Verbal Learning and Verbal Behavior 19, 84-95.

Mark A.McDaniel