A Cognitive Model of Learning

Figure 1, made by the author from Sharon Derry's review of cognitive learning theory, provides a starting place. Notes 11 - 12

Part of a working paper, Serving the Strategic Reader: Cognitive Reading Theory and its Implications for the Teaching of Writing

by Gerald Grow, Ph.D.

Professor of Journalism
Florida A&M University
copyright © 1994, 1996

How to reference this publication

Figure 1. A Cognitive Model of Learning

[To continue the Short Tour, click here for a summary of this section.]

Steps in Figure 1 are discussed in the following section.

Steps 1-3. Comprehension

When faced with a new text, readers do not begin by "reading" in the sense of starting at the first word and moving sequentially toward the last word, they first predict what the passage will mean. Prediction, which plays a key role in Frank Smith's cogent account of reading, can be understood as "the prior elimination of unlikely alternatives" or "questions we ask the world." Note 13 Readers not only predict what an article will contain or what a statement will mean, they predict:

Such predictions may be simultaneous with the first decoding of the letters on the page, and, since a reader can be led to know what to expect by illustrations, the nature of the publication, or other contextual cues, such predictions may even precede reading. Most of this activity is unconscious and appears to be part of the way we orient ourselves in the world.

For example, a glance at the heading "Sports" in a section of the newspaper will cue one reader to relax, settle in for a leisurely perusal of sports news, and begin remembering what she already knows (The playoffs were held last night.), perhaps generating questions (Who won the basketball playoffs?) or otherwise focusing attention, knowledge, and reading strategies (what to skip, which teams to read about) onto the new information (researchers call this activating relevant schemas).

Another reader may see the heading "Sports" and take it as a cue to skip that entire section of the paper. Like all readers, these two are active and strategic: They actively interpret what they read; they choose what to read and how deeply to read it.

Researchers use the word "comprehension" to label what takes place when the reader connects the new information with prior knowledge. Information alone, no matter how well written, does not create comprehension. Comprehension depends on the reader's prior knowledge and reading strategies.

The sports fan has a large body of organized information about sports and can quickly fit new information into this framework. Another reader, with a different body of organized knowledge, may actively seek out a music review to find out how well a visiting chamber group played the opening fugue of Beethoven's C-Sharp Minor Quartet. Much information in the sports pages is incomprehensible to someone lacking prerequisite knowledge of sports; much information in a music review is incomprehensible to someone lacking prerequisite knowledge of music.

Even though both the sports report and the music review contain well-written "information," that information becomes comprehensible only to readers who can combine the new information with organized existing knowledge (a knowledge network) on the subject matter. The comprehension of new information requires a meeting of the new with the known. This meeting of the known with the new is one of the fundamental concepts of cognitive learning theory.

Notice that the known does not wait passively in the mind; it actively goes forth to meet and make sense of the new. Indeed, unless a reader is able to predict, to ask relevant questions, and to know how to find the answers, comprehension is not possible. Note 14

Steps 4 and 5. Learning

Comprehension does not necessarily lead to learning--at least, not to learning of a meaningful, useful kind. How many people can remember the actual words of books they read five years ago? As Bartlett demonstrated in the 1930s, people do not ordinarily remember much of the exact information they read. Instead, they learn the "gist" of it. They select. They use selected portions of the information to address issues important to them. (Never mind the national scores, did the local team win?) For all practical purposes, the vast bulk of information one reads is simply forgotten. We use it to "service" our model of the world, then discard the details.

Once new information has been comprehended--by linking it to what is already known--cognitive theorists say that the new information can then be learned through activities which enrich the connections between the new and the old knowledge. Researchers have studied some ways students convert "comprehended information" into "learned information," through such activities as taking notes, summarizing, outlining, making analogies, relating the information to yourself personally, creating mental imagery, and similar activities known as elaboration.

Knowledge can be called "meaningful" only after it is richly interconnected with related knowledge. Knowledge can be called "useful" only if you can access it under appropriate circumstances. Meaningful knowledge is filed and cross referenced with other knowledge to which it is connected. Useful knowledge is filed and cross-referenced so that you can find it when you need it.

Some of these points may seem obvious, but studies strongly suggest that this kind of mental housekeeping makes the difference between good and poor readers. Note 16

Steps 6 and 7. Recall and Reconstruction

People apparently do not store knowledge as long, complete strings of text but rather in a dynamic, interlinked network in which the elements have been analyzed into categories linked by multiple relationships that may be organized as schemas, scripts, narratives, or other forms.

The organization of memory seems to be a good deal more like multidimensional hypertext than like paragraphs of linear prose. People not only abstract the gist from what they read, they often do not recall what they read verbatim; instead, they reconstruct what they "know."

Just as a well-organized knowledge network enables you to scan for and read only what you do not already know in a new text, the knowledge network enables you to regenerate the essence of what you know, rather than having to remember it as complete texts of information. Memory--again, the discovery dates to Bartlett in the 1930s--is reconstructive. People reinvent as they recall, and they appear to reinvent on the basis of some deep structure they have used to hang a few key facts on. A good knowledge structure, in fact, can enable you to "remember" things you never learned--by inferring them from what you already know.

The blurred boundary between inference and perception is one thing that makes eye-witnesses unreliable, and a skillful questioner can induce a witness to "remember" things never experienced. In spite this blurriness, we derive a good deal of our knowledge of the world by applying reason to the things we already know, extending those by inference, and making good guesses about what "must" exist in the gaps of our knowledge. In Kenneth Boulding's more organic terms, "Knowledge grows also because of inward teachers as well as outward messages." Notes 17 - 18

The basic theme we are developing can be stated this way:

In a nutshell: Cognition is an active, recursive, integrated process by which we continuously model the world and continuously modify the model.
[If you are on the Short Tour, click here for a Summary of cognitive reading theory.]


(Not shown in the diagram.) Under ideal conditions, learners receive feedback at various stages. A teacher, for example, may:

Because they are not interactive, the major media cannot provide direct feedback to ensure the accuracy of what readers learn. Later, we will consider whether media can offer alternatives to interactive feedback.


Gerald Grow's Home Page