Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Explanations of Reading Comprehension: Schema Theory and Critical Thinking Theory

by Stephen Norris & Linda Phillips - 1987

The verbalized thinking of two sixth grade children while reading is analyzed using schema theory. An outline of a critical thinking theory is given and contrasted with schema theory. Conclusions for reading theoreticians are discussed. (Source: ERIC)

This paper is based on one delivered at the 1986 annual meeting of the Philosophy of Education Society and published as “Reading Well Is Thinking Well, ” in Philosophy of Education 1986, ed. Nicholas Burbules (Normal: Ill.: Philosophy of Education Society 1987). We thank Margaret Buchmann and Robert Floden for their advice and encouragement.

A widely accepted theory in the reading field is that a reader comprehends a piece of text by activating, modifying, or creating a schema into which the text information will fit. Schema theory has advanced our understanding of the interactions among the reader, the reading context, and the text. Prior to schema theory, emphasis was placed on the text and the skills needed to understand it, without regard to the characteristics of the reader and the varying contexts in which reading occurred. Although schema theory has broadened our perspective on the nature of reading, it is not without problems. In particular, the processes of activating, modifying, and creating schemata are not adequately explicated and there is no sufficient articulation of the constraints needed for directing them.

Any attempt to explicate the processes through which schemata are used to understand text has the possibility of leading to major modifications of schema theory, or even to its demise. However, for purposes of this article we will focus on the question of what constraints would need to be placed on the processes if schema theory is to work at all as an explanation of comprehending text, while in the end making no firm claims about the truth of the theory. The constraints we will suggest combine strategies and principles derived from a conception of good thinking, which we will also present. Our belief is that these constraints must be a part of any theory of text comprehension, and thus we feel justified for purposes of this article in adopting a neutral position toward schema theory.

The idea that good thinking is involved in good reading is not new. In 1908 Edmund Burke Huey bemoaned the fact “that a false ideal has taken [hold] of us, [namely], that to read is to say just what is upon the page, instead of to think.”1 Less than a decade later Edward L. Thorndike urged that reading is reasoning, likening comprehending a written paragraph to solving a problem in mathematics.2 Thorndike’s ideas provoked much thought, yet seventy years later there is no detailed theory of the relation between thinking well and reading well.3 Thus, reading theoreticians still find it necessary to speak out against the persistent views that reading is simply a matter of decoding written symbols into sounds,4 or of concatenating the meanings of individual words.5

Researchers and theoreticians of good thinking also recognize that reading involves thinking. For example, Vincent Ruggiero, in a book on critical and creative thinking, claims that reading is reasoning, and Michael Scriven claims both that reading is the prerequisite skill in reasoning and that it is in fact basic reasoning.6 We agree with the tenor of these positions but we do have problems with them. Ruggiero’s claim that reading is reasoning, like similar ones made by reading experts, is of little use to educational theory or practice unless some detail is provided to explain how the two processes are the same. Scriven’s position that reading plays either a subsidiary or a facilitative role in reasoning is, we believe, unjustified, as is the related view of some reading theoreticians that critical reading is critical thinking applied to reading.7 We wish to resist any attempt to separate reading from thinking and to urge that reading well is a species of thinking well.

In the first part of this article, we introduce two sixth-grade children who verbalize their thinking as they read a short passage. The second part first provides a brief description of schema theory, then offers an explanation using schema theory of the difference between the children’s reading and shows how schema theory breaks down by not being able to completely adequately distinguish the quality of one child’s reading from that of the other. An outline of critical thinking theory is provided, and we show how that theory can meet some of the problems faced by schema theory in distinguishing the quality of the children’s reading. The third part of the article analyzes the differences between schema theory and critical-thinking theory, and in the final section we pull together the discussion of the previous parts by pointing to some conclusions for the reading field as well as to some yet-to-be-answered problems.


Forty students were involved in a study of the reading process.8 There were equal numbers of good and poor readers as defined by their performance on a standardized reading test and the judgments of their teachers. We realize the risk of circular reasoning here in our attempt to show that schema theory cannot adequately distinguish good readers from poor ones. In order to draw this conclusion, good and poor readers first have to be identified. In order to identify them, we need a theory of what comprises good and poor reading, and schema theory is supposed to be that theory. So how does the enterprise get off the ground? We have no general answer to this question, with which philosophers of science continue to wrestle. The best we can offer at this time is that we start with our intuitive judgments and, by remaining open to change, can bootstrap our way to more advanced understandings. Descriptions of readers based on standardized tests and teachers’ appraisals are rough, impressionistic, and often inaccurate. Yet they often concur with other indicators such as overall performance in school and the intuitive judgments of teachers and reading researchers. We avoided reasoning in a circle by treating test scores and teacher judgments not as definers but as fallible indicators of good and poor readers, and by remaining prepared to alter categorizations of children with subsequent information.

We will introduce you to two sixth-grade children from Alberta, Steven and Colleen, as they read a passage related to the experience of children in Newfoundland. Steven was a good reader and Colleen was a poor one as identified by the criteria described above. The children were told that the story was about some men going to do something. The story, without a title, was presented in six episodes, which they were asked to read aloud. Immediately after reading each episode they were asked to report on what they were thinking and to answer other relevant questions. Interview approaches that have been shown to be reasonably trustworthy were used throughout.9

The sample from which Steven and Colleen were chosen showed no overall differences between boys and girls, so the sex of Steven and Colleen has no general implications for reading. These children were chosen solely because their protocols were appropriate for illustrating the points we wish to make.

In what follows we will present episode-by-episode Steven’s and Colleen’s thinking aloud as they work through the passage. The experimenter’s questions are given in brackets. We have chosen to make our example detailed, because we see this as the best route for providing specificity to otherwise vague generalizations about the relationship between reading and thinking. To simulate the task for you we present the passage without a title and one episode at a time as was done with the children.

Episode 1

The stillness of the morning air was broken. The men headed down the bay.


The men were heading down the bay, I’m not sure why yet. It was a very peaceful morning. [Any questions?] No, not really. [Where do you think they’re going?] I think they might be going sailing, water skiing, or something like that.


The men are going shopping. [Why do you say that?] They’re going to buy clothes at The Bay. [What is The Bay?] It’s a shopping center. [Any questions?] No. [Where do you think they’re going?] They’re going shopping because it seems like they broke something.

Steven recognizes that there is insufficient information for explaining what the men are doing. On questioning, he tentatively suggests a couple of alternatives consistent with the information given, but indicates there are other possibilities. Colleen presents one explanation of the story, and seems fairly definitive that the men are going to buy clothes at The Bay, a chain of department stores in Canada, On being queried she maintains her idea that the men are going shopping but offers an explanation inconsistent with her first one that they are going to buy clothes. To do this she assumes that something concrete was broken, which could be replaced at The Bay.

Episode 2

The net was hard to pull. The heavy sea and strong tide made it even more difficult for the girdie. The meshed catch encouraged us to try harder.


It was not a very good day as there were waves which made it difficult for the girdie. That must be some kind of machine for doing something. The net could be for pulling something out of the water like an old wreck. No, wait! It said “meshed catch.” I don’t know why but that makes me think of fish and, sure, if you caught fish you’d really want to get them. [Any questions?] No questions, just that I think maybe the girdie is a machine for helping the men pull in the fish or whatever it was. Maybe a type of pulley.


I guess The Bay must have a big water fountain. [Why was the net hard to pull?] There was a lot of force on the water. [Why was it important for them to pull the net?] It was something they had to do. [What do you mean?] They had to pull on the net and it was hard to do. [Any questions?] No. [Where do you think they’re going?] Shopping.

For both children the interpretations of Episode 2 built on those of Episode 1. Steven continues to question what the men were doing. He raises a number of alternative interpretation dealing with the context of the sea. He refines his interpretations through testing hypothetical interpretations against specific details, and hypotheses of specific word meanings against his emerging interpretation of the story. At the outset he makes an inference that a girdie is a machine, but leaves details about its nature and function unspecified. He tentatively offers one specific use for the net, but immediately questions this use when he realizes that it will not account for the meshed catch, and substitutes an alternative function. He then confirms this interpretation with the fact from the story that the men were encouraged to try harder and his belief that if you caught fish you would really want to bring them aboard. Finally, he sees that he is in a position to offer a more definitive but tentative interpretation of the word girdie.

Colleen maintains her interpretation of going shopping at The Bay. When questioned about her interpretation, Colleen responds in vague or tautological terms. She seems not to integrate information relating to the terms net, catch, and sea, and seems satisfied to remain uninformed about the nature of the girdie and the reason for pulling the net. In the end, she concludes definitively that the men are going shopping.

Episode 3

With four quintels aboard, we were now ready to leave. The skipper saw mares’ tails in the north.


I wonder what quintels are? I think maybe it’s a sea term, a word that means perhaps the weight aboard. Yes, maybe it’s how much fish they had aboard. [So you think it was fish?] I think fish or maybe something they had found in the water but I think fish more because of the word “catch.” [Why were they worried about the mares’ tails?] I’m not sure. Mares’ tails, let me see, mares are horses but horses are not going to be in the water. They are out in a boat on the bay so there’s nothing around except sky and water. The mares’ tails are in the north. Here farmers watch the north for bad weather, so maybe the fishermen do the same thing. Yeah, I think that’s it, it’s a cloud formation which could mean strong winds and hail or something which I think could be dangerous if you were in a boat and a lot of weight aboard. [Any questions?] I am curious now about whether I’m right about the mares’ tails.


They were finished their shopping and were ready to go home. [What did they have aboard?] Quintels. [What are quintels?] I don’t know. [Why were they worried about the mares’ tails?] There were a group of horses on the street and they wouldn’t move and the men were afraid they would attack the car. [Any questions?] No.

Steven is successful in his efforts to incorporate the new information into an evolving interpretation. From the outset Steven acknowledges that he does not know the meaning of quintel and seeks a resolution of this unknown. He derives a meaning consistent with his evolving interpretation and with the textual evidence. In his attempt to understand the expression mares’ tails he first acknowledges that he does not know the meaning of the expression, Thence, he establishes what he does know from background knowledge (mares are horses, horses are not going to be in the water, there is nothing around except sky and water, farmers watch the north for bad weather) and textual information (the men are on a bay, they have things aboard, the mares’ tails are in the north), and inferences he has previously made (the men are in a boat, they are fishing). He integrates this knowledge into a comparison between the concerns of Alberta farmers with which he is familiar, and what he takes to be analogous concerns of fishermen. On seeing the pertinence of this analogy he draws the conclusion that the mares’ tails must be a cloud formation foreboding inclement weather. He claims support for his conclusion in the fact that it would explain the skipper’s concern for the mares’ tails, indicating that he did not lose sight of the overall task of understanding the story.

Colleen maintains her original interpretation but does not incorporate all the new textual information into it. She works with the information on the men’s leaving and the mares’ tails, but appears to ignore or remain vague about other information. For example, she says the cargo was comprised of quintels but indicates no effort to determine what these things are. She cites the fact that the men were ready to leave and suggests that they have finished their shopping, but does not attempt to explain the use of such words as skipper, cargo, and aboard in the context of shopping for clothes. She interprets mares’ tails as a group of horses that possibly would attack the men, but gives no account of what the horses might be doing on the street. Basically, she appears to grow tolerant of ambiguity and incompleteness in her interpretation.

Episode 4

We tied up to the wharf. We hastily grabbed our prongs and set to work. The catch was left in the stage while we had breakfast.


They made it safely to land, they unloaded the fish. I’m fairly sure it’s fish now but I don’t know what the prongs are, truthfully. They then left the fish somewhere and had breakfast. [Why did they tie up to the wharf?] It’s the thing to do, otherwise the boat would be swept away. [Why was the catch unloaded with prongs?] Since I don’t know what they are, I’m not sure but maybe they’re not allowed to unload with their hands or maybe it’s faster that way because they’re slimy creatures. [Any questions?] The word “stage” is interesting because I know that my meaning for the word doesn’t make sense, so I’d like to know about it.


They are on a wharf and are going for breakfast. [Why did they tie up to the wharf?] I don’t know. [So where do you think the men are now?] I don’t know. [So first the men went to The Bay to go shopping and then what happened?] They went shopping and saw a waterfall with fish. They were catching some fish with little nets like in the stores to bring home and when they were finished they met some horses on the street. [Why was the catch unloaded with prongs?] I think that they are now going to a play or some show because it says about a stage. [Any questions?] No.

Steven acknowledges that he is fairly sure of his interpretation, having remained tentative to this point. He then relies on his belief that the men are catching fish to propose inferences about the use of the prongs, and to recognize the inappropriateness of his known meaning of stage. From there he expresses a curiosity about the meaning of the word to make sense in this context.

Colleen begins by saying the men are on a wharf and are going for breakfast. She does not mention the intermediate information that the men set to work, nor answer the question of what they were doing. Even though she admitted not knowing answers to crucial questions, she did not take a cue from this to test her interpretation.

Episode 5

The splitting was done by the skipper. The boys did the cutting and gutting.


Well there’s no doubt in my mind now that it’s fishing because they’re splitting and gutting them. [What kind of fish do you think?] I don’t really know, maybe salmon or perch. [What do you think they did with the fish?] Maybe they froze them or just sold them as they are.


They seem to be cutting something up now. [Like what?] It doesn’t say. [Maybe it was fish. What kind do you think it was?] Maybe goldfish or guppies. [What do you think they did with the fish?] Put them in a bowl or what’s the name of it, you know the water tanks, an aquarium.

Steven confirms with certainty his interpretation that the men are fishing. Having accomplished his overall goal to understand what the men are doing, he is then in a position to provide some alternatives for the details concerning the kind and use of the fish. He uses both background knowledge and text to adjudicate his responses, while remaining tentative.

Colleen does not tie this episode to what she said about the previous one. She notes that the men are cutting something without questioning how this relates to her conclusion that they were going to a play or a show. She accepts the hint that it might be fish they are cutting, proceeds to suggest that the fish might be guppies or goldfish, and does not appear to consider further the idea of putting these cut-up fish in an aquarium.

Episode 6

Catching fish is filled with risk.


[So what did you think of the story?] It was interesting but tell me about the stage if you have a minute.


[So were you correct about what the men were doing?] Yes. [Do you have any questions?] No. [So, what was the story about?] Like I said before, they went shopping and got some fish, met some horses and then went to a show. That’s about it.

Steven’s focus is on the overall interpretation even to the end of the story when it is resolved that the men are fishing. He still wants to be informed about the word stage. He seems very much disposed toward understanding as much as he can.

Colleen continues with her original interpretation. She summarizes it and indicates that she has no questions. She concludes with a story, but one that fails to take into account a great deal of the textual information. For example, the story does not contain the notion of the relationship between fishing and risk contained in the last episode. Throughout, Colleen seemed willing to tolerate ambiguity and lack of understanding, focusing only on those elements of the text that were immediately interpretable within her original interpretation.

Having supplied the protocols in which the children gave their interpretations of the story, we now turn to the question of how schema theory and critical thinking theory enable us to explain the differences between what the children did.


What distinguishes Steven’s interpretation from Colleen’s? The answer depends on the theory of reading adopted. In this section we will consider the question from the perspective of two theories, schema theory and critical thinking theory. According to schema theory the difference between Steven and Colleen is that they bring different schemata to bear in making their interpretations. According to critical thinking theory the difference lies in the extent to which they use criteria of good thinking in making their interpretations. These are answer sketches to the question of what distinguishes Steven’s and Colleen’s interpretations. We will now elaborate the answers.


Schema theory’s tradition has extended at least from Bartlett, to Piaget, through contemporary researchers such as Anderson and Pearson and Petrie.10 While it has not been described in precisely the same way by each of these people, there are fundamental similarities in their descriptions. According to schema theory, understanding occurs when stimuli are fitted into existing cognitive structures. A cognitive structure consists of an organized set of concepts, possibly hierarchically related. For example, a person’s cognitive structure may contain the schema DOG. The schema would consist of the relationship of this concept both to more general and to more particular concepts. More general concepts might consist of canines, four-legged creatures, mammals, living creatures, or earthly creatures. More particular concepts may include domestics, wild dogs, long-haired dogs, short-haired dogs, spaniels, setters, and mutts.

An example of a schema and how it would enable comprehension is provided by Anderson and Pearson11 in their discussion of the SHIP-CHRISTENING schema. While acknowledging the oversimplification in their description they maintain that the SHIP-CHRISTENING schema

can be analyzed into six parts: that it is done to bless a ship, that it normally takes place in a dry dock, [that it is done by a celebrity, that it involves a new ship, that a bottle is usually broken on the bow, and that it is usually done just before launching]. In the jargon of schema theory, these parts are called “nodes, ” “variables,” or “slots.” When the schema gets activated and is used to interpret some event, the slots are “instantiated” with particular information.

They offer the following short story to illustrate how a reader might use the SHIP-CHRISTENING schema to understand the story.

Queen Elizabeth participated in a long-delayed ceremony in Clydebank, Scotland, yesterday. While there is still bitterness here following the protracted strike, on this occasion a crowd of shipyard workers numbering in the hundreds joined dignitaries in cheering as the HMS Pinafore slipped into the water.

Anderson and Pearson suggest that “the generally good fit” of this story with the SHIP-CHRISTENING schema is evidence that part of the message has been comprehended. Queen Elizabeth, they maintain, fits the celebrity slot, Clydebank fills the dry dock slot, and the clause “slipped into the water” fills the just-before-launching slot. “Therefore,” they conclude, “the ceremony mentioned is probably a ship christening.”

Admittedly, this is a simplistic account of schema theory and the theorists in the field use such examples as the above only to present an initial, overall view. For example, Anderson and Pearson believe that a schema must also include a specification of the relations between its component parts. The SHIP-CHRISTENING schema needs to include information on the temporal order of events in a ship christening, on the causal interaction among its components (that, for example, it is the celebrity who breaks the bottle on the ship), and on the spatial relations among events, such as that the whole ceremony takes place near a body of water.

A reader is said to comprehend a piece of text, then, when he or she “is able to activate or construct a schema that gives a good account of the objects and events described.”12 A good account is said to obtain when all the textual information can be fitted into slots in the schema, when all the major slots of the schema contain information from the text, and when the account is coherent.13 In order to achieve such a coherent account, the reader may have to do many things, not the least of which is to make inferences. For example, in the passage about Queen Elizabeth it is not explicitly stated that the Queen is the celebrity who performed the ship christening, or that there was a bottle of champagne used. A reader would have to infer these things or recall them from background knowledge, if they are to become part of the readers’s interpretation.

Explaining by Schema Theory the Difference between Steven’s and Colleen’s Reading

We work on the assumption that Steven’s interpretation is better than Colleen’s, not merely different in content. Some people may not share this assumption, and we will address some of their concerns later. Nevertheless, we believe that it is impossible for anyone to consistently maintain that no interpretations are better than others, because they would at least wish to believe that their view (namely, that no interpretations are better than others) is an interpretation better than the alternatives. Thus, any adequate theory of reading must be able to explain what makes some interpretations better than others and how, in general, readers come up with quality interpretations, not just any interpretations. Do better interpretations come about by happenstance or by design? If the latter, of what actions is the design composed and how are these actions guided? Schema theory, we will maintain, can answer these questions in part, but something else is needed to complete the answers. We will argue subsequently that principles and strategies of critical-thinking theory can assist in completing these answers.

According to schema theory, Steven and Colleen used different schemata to interpret the story. Steven’s interpretation was guided primarily by a FISHING schema. Such a schema might contain the following elements, among others: done on water; fish are caught; weather is an ever-present consideration; return to land; and fish are prepared. Each of these elements can be analyzed into sub-elements. For example, the fish-are-caught element might contain the following sub-elements: sometimes uses nets; sometimes uses traps; sometimes uses hooks; other equipment used to haul fish aboard; and storage place for fish on board. These sub-elements can be further divided, and so on. Colleen’s interpretation was, on the other hand, guided primarily by a SHOPPING schema. Such a schema can contain such elements as the following: there is an intention to buy something; often occurs in a shopping center; such centers often have diverse attractions; and travel to and from the shopping location is generally required. As in the FISHING schema, each of these elements can be further analyzed. The diverse attractions, for example, might include the following sub-elements: rides for children; water fountains; arrangements of flowers, plants, and trees; special exhibits; seasonal decorations; temporary concessions; and fashion shows.

Schema theory maintains that comprehension occurs by matching elements in schemata with pieces of information in what is being read or, put another way, by filling the slots in schemata with pieces of information. Therefore, Steven’s understanding occurs by his filling the slots in the FISHING schema with information, either explicitly found in or derived from the story, and Colleen’s understanding occurs by her filling slots in the SHOPPING schema. For example, if the first two episodes are reexamined, it can be seen that Steven probably filled done-on-water, fish-are-caught, and equipment is-used slots of a FISHING schema. Colleen seemed to fill intention-to-buy, occurs-in-shopping-center, and water-fountain slots of a SHOPPING schema.

Schema theory allows us to display a number of differences between the children’s reading. It can be said that Colleen activated the SHOPPING schema from the outset (in her first sentence), and then proceeded to fill what she took to be slots in that schema using information from the text and her background knowledge. Steven, on the other hand, constructed a schema over several episodes in the story, not settling definitively on the FISHING schema until the fifth episode. Another difference that schema theory can be used to highlight is that Steven managed better than Colleen to fit the textual information into slots in the schema ultimately chosen. For example, in the second episode Steven was able to incorporate the information about the girdie, whereas this was completely overlooked by Colleen; in the third episode Steven made a reasonable conjecture about how quintels would fit, but Colleen just said she did not know what quintels might be; also in the third episode, Steven incorporated the information about the mares’ tails being in the north, whereas Colleen did not deal with the compass direction of the mares’ tails.

These differences can be used to compare the quality of the children’s reading. According to schema theory, the first difference marks merely a procedural difference in acquiring understanding, but no difference in value. Schema theory is not able to address the issue of whether Colleen’s schema-activation procedure is better or worse, as a route to comprehension, than Steven’s schema-construction procedure. The second difference is one about which schema theory has something to say. The fact that Steven was able to fit more textual information into his schema than Colleen was able to fit into hers marks a qualitative difference. Schematic interpretations are better to the extent that all the textual information can be fitted into them.

Schema theory is also able to speak to the issue of the quality of the fit of information into slots. While the ideas of “good lit” and “forced fit” might be difficult to specify precisely, there does seem to be a sense in which Colleen was forcing information to lit the schema she had activated. For example, the idea of cutting up guppies and then placing them in an aquarium is not a notion that fits well (in some imprecisely specified sense) with the idea of a shopping trip.

So, then, schema theory is able to provide some account of why Steven’s reading is better than Colleen’s. However, we think that the account is incomplete, and in the following section we raise some questions that we believe schema theory in its present form is unable to answer.

Problems with Schema Theory

A primary problem with schema theory is that it provides no way for a reader to “get off the ground” in the comprehension process. In reading a piece of text, a reader must proceed from a state of relative lack of understanding to a state of relative understanding. However, if, in order to understand, a reader must have the right schema, must not the reader already have understood in order to come up with that schema? Allow us to elaborate. If a reader does not understand until the proper schema is activated, then how does the reader know how to proceed before activating a schema? Since Steven did not conclude until the fifth episode that the proper schema was a FISHING schema, what kept him on track throughout the previous episodes? The answer cannot be that he focused on filling slots in a FISHING schema, because until that point he had not confirmed FISHING as the proper schema. Furthermore, in the fifth episode, how did Steven know that he had “arrived” where he ought to be? It is question-begging to answer by saying that he fitted all the information into a FISHING schema, because there was no independent way for him to know what the correct schema was. On the other hand, while activating a schema from the outset can provide a means for a reader to keep on track (by subsequently seeking to fill slots in that schema), we can see from Colleen’s case that it does not necessarily lead to understanding. Being on a track is not good enough unless it is the right track.

Derivatively, the problem may be stated as follows. If understanding means fitting information into slots in a schema, then in order to understand by this route one must have already understood, because for something to be information, and not just ink marks on a page, it must be understood. However, for the ink marks on a page to be understood one must, by hypothesis in schema theory, have a schema. This leads to a problem wherein schemata are needed to fill slots in schemata ad infinitum. The SHIP-CHRISTENING schema can be used to illustrate this point. Suppose that one slot in the schema is “to bless ship.”14 For what is in this slot to be itself understood, schema theory would minimally require schemata for “BLESSING,” “SHIP,” and “REASON FOR ACTING.” Each of these schemata can then be analyzed into slots in the same way as the SHIP-CHRISTENING schema. These subsequent slots would also require schemata to be understood. There is no mechanism provided by schema theory for bringing this process of analysis to a halt.

As a consequence of this endless process, schema theory does not, in the final analysis, provide an explanation of understanding. Allow us to illustrate more clearly how this failure comes about through a sequence of questions and answers. Schema theory derives its initial motivation from the question of how a reader comprehends a piece of text. The short answer to this question provided by schema theory is that the reader comes up with a schema into which the text can fit. What then is a schema? The shortest answer is that a schema is an explanation, that is, “an account of the data in the text.”15 What is an explanation? As a first approximation, an answer to this latter question is that an explanation is something that provides understanding. However, the upshot of this series of questions and answers is that a reader comprehends a piece of text by understanding it! This conclusion appears tautological, and suggests a fundamental problem with schema theory.

The problem arises from the fact that presentations of schema theory typically do not explicitly differentiate the tusk sense of understanding from the achievement sense of understanding.16 In the task sense, understanding refers to the mental process or activity that is undertaken to reach some end. In the achievement sense, understanding refers to the product or result of some mental processes, the end that is reached. In the apparent tautology above, there would be no problem if what was meant was that in order to understand (achievement sense) a reader engages a process of understanding (task sense). Schema theory, however, does not make this distinction explicit and, as a matter of fact, seems to be more product- than process-oriented. While schema theory countenances a process of filling slots in schemata, that is, “finding a mental ‘home’ for the information in the text,”17 the emphasis in accounts of the theory seems to be placed more on describing the products of the process, the schemata themselves. Thus, there is much said about the detailed structure of schemata, but there is little to be found about how such schemata come about, other than that they are “constructed” or “activated.” Even then, the processes of construction and activation, as described, are subservient to particular schemata. For example, the process of slot-filling implies that slots are there to fill, implying further that the end product (the schema) is already in place, because schemata are defined by the slots that they have. Now, while the process of schema activation and construction cannot be conceived intelligibly as completely separate from the products, there has to be some way to speak of the quality of mental processes apart from the products they produce.

If quality processes are only those that lead to quality products, we are led to the view that Isaac Newton’s thinking was poor because many of his conclusions are false. Of course, this conclusion is absurd. Given what was known in Newton’s time, the conclusions he reached about motion were as profound as could be expected from any human being. It is only with a retrospective look from the twentieth century and working with a substantially greater knowledge-based that Albert Einstein could find fault with Newton’s failure in the seventeenth century to take account of the relative motion of objects when developing laws of motion. The fact that we continue to praise Newton’s ideas even though we consider them to be false presents the germ of a complete answer to the question of what makes Steven’s reading better than Colleen’s. The elaborated answer requires an exposition of critical thinking theory, to which we now turn.


Critical-thinking theory finds its roots in ancient Greek philosophers such as Anaximander and Thales, who sought to approach truth by means of critical discussion. A critical discussion is one in which ideas are pitted against one another. The underlying belief is that it is only through criticism and critical discussion that we can get nearer to the truth. In the present century one of the strongest proponents of this theory is Sir Karl Popper, who believes that scientific ideas progress by being pitted against one another in a process of conjectures and refutations.18 In education, critical-thinking theory is grounded in John Dewey’s idea of reflective thinking and in B. Othanel Smith’s studies of how to correctly go about accepting and rejecting statements.19

Currently, one of the most comprehensive theories of critical thinking is offered by Robert Ennis. Ennis began his theorizing over two decades ago with the idea of critical thinking as “the correct assessing of statements.”20 Finding this conception of critical thinking too narrow, he has extended the meaning to include not only the appraisal of statements but the whole process of reasonably and reflectively going about deciding what to believe or do.21 According to Ennis’s theory, thinking critically involves the use of both abilities and dispositions. In a context of deciding what to believe or do, critical thinking dispositions include tendencies to seek a clear statement of the thesis or question, to take into account the total situation, to remain relevant to the main point, to seek alternatives, to take a position and to change it when the evidence and reasons are sufficient to do so, and to remain open-minded. However, having dispositions to think critically is not enough. A person may have a disposition to remain relevant to the main point, in the sense of not wanting to stray from it, but just not know how to accomplish it. In order to succeed in remaining relevant to the main point, the person requires certain abilities.

Ennis classifies these abilities into live main types: elementary clarity-related abilities; advanced clarity-related abilities; inference-related abilities; abilities related to establishing a sound basis for inference; and abilities involved in going about decision making in an orderly and useful way.22 Each of these categories contains a large number of abilities, not all of which can be described here. For illustration, the category “elementary clarification abilities” includes abilities for focusing on a question; abilities for analyzing arguments by identifying their conclusions, stated and unstated reasons; and abilities for asking and answering questions of clarification or challenge dealing with the main point, the facts of the case, and how the main point might be exemplified.

When abilities such as these are combined with the critical-thinking dispositions discussed above, there is a comprehensive picture of how to go about the process of deciding what to believe or do. We will illustrate the process by examining a short poem and trying to decide what to believe the poem means. The poem is by X. J. Kennedy, entitled “Who to Pet and Who Not To,” and is taken from a primary school basal reading textbook. It thus represents a real task of deciding what to believe that children in school may face.

Go pet a kitten, pet a dog,

Go pet a worm for practice,

But don’t go pet a porcupine—

You want to be a cactus?

In deciding what to believe this poem means, abilities from several of Ennis’s categories are required in addition to several of the critical-thinking dispositions. In the first place, it seems that there must be a disposition to seek a clear statement of the poem’s meaning and a disposition to take into account the total situation relevant to the poem, a disposition to be open-minded in considering alternative meanings, and a disposition to form conclusions and to alter them when warranted. When people are so disposed, they have the requisite motivation to exercise their abilities in seeking the meaning of the poem.

To do this, abilities falling under at least four of the live categories of Ennis’s classification scheme would be needed. Under decision-making there is a need to define what problem exists in interpreting the poem and to formulate alternative solutions to the problem. That is, the reader has to reach some conclusion of how the poem is to be taken. The reader might wonder: Is the poem a warning? Is it a set of instructions? Is it a joke? Does it have a moral? Posing such questions is a strategy or tactic that both shows there is a problem of interpretation and suggests some possible solutions to the problem.

Deciding how the poem is to be interpreted is a complex task. Essentially, it is one of determining an illocutionary point for the poem, and there are at least the four possibilities listed in the previous paragraph. Each of these possibilities provides an explanation of the author’s intended meaning for the poem and the comparative quality of these explanations must be adjudicated. The procedures of offering and adjudicating explanatory conclusions fall under Ennis’s category of inference-related abilities. The procedures involve taking into account all the relevant information and applying criteria for judging the adequacy of the explanatory conclusions. These criteria require judging whether a proposed conclusion explains the evidence and is consistent with known facts, whether alternative conclusions are inconsistent with known facts, and whether the proposed conclusion is plausible.

Suppose the conclusion is that the poem is a set of instructions. If this were the case, critical thinking theory would suggest that the conclusion be assessed according to the criteria in the last paragraph in the context of alternative conclusions. The first criterion says that the conclusion must explain the evidence. The first three lines of the poem are stated as directives, thus providing support for the conclusion. However, there are certain anomalies. The directive tone of the poem changes in the last line. This line asks a question, suggesting that the poem might not be a set of instructions. Also, there is a suggestion of humor in the last line, another piece of evidence not easily explicable if the poem is a set of instructions. The conclusion, then, does not explain all the evidence in the poem and cannot, therefore, serve as a completely adequate explanation. If there were no alternative explanations, this one might be temporarily held, but there are alternatives. Critical thinking theory would suggest that alternatives be created and considered.

Given the suggestion of humor in the last line, maybe the poem is intended to be a joke. Is this conclusion consistent with the evidence? While consistent with the last line, this conclusion is weak where the previous one is strong. That is, the conclusion that the poem is a joke cannot readily explain the existence of directives in the first three lines. Also, while the thought of looking like a cactus can be considered humorous, being stuck with porcupine quills is not very funny. The conclusion that the poem is a joke cannot explain why the thought of a painful experience is included in the poem.

There are still other alternative conclusions, of which we will consider two. Is the poem a warning? Does the poem have a moral? There is a degree of overlap in the concepts of warning and moral. Some morals can provide warnings. Warnings, however, can be either specific or general in nature, whereas morals are always general. Thus, we can expect to need some very clear evidence to distinguish between these two conclusions. The third line of the poem provides a specific warning against petting porcupines, but the title of the poem, ‘Who to Pet and Who Not To,” suggests something more general. With a title like this you would expect to be told who in general is safe to pet. There is thus a contrast between the general warning advertised in the title and the very specific warning in the body of the poem. How can this contrast be explained? Is the poem about not petting porcupines, or is it about something more general? Other criteria of critical thinking theory become relevant in answering this last question. It is a known fact about poetry that it is widely used as a medium for expressing general lessons about human experience. Therefore, the conclusion that the poem is a general warning is consistent with this fact and is plausible in light of it.

However, does the poem also provide a moral? It is clear that the poem can be easily interpreted as a moral. When we asked a number of colleagues to say what they understood the poem to be about, some interpreted the poem as providing a moral, but some did not. Here are some samples:

1. Be careful whom or what you pet. Some people or things might be “unpetable.” “Appropriateness” is the key word here.

2. Beyond the simplistic interpretation of what is and is not safe to touch, this poem deals with life situations and warns of the hurt that can be caused by some.

3. Be sure you understand the consequences of your actions before acting.

4. I would see it (especially given the title) as an essay on personality types, some prickly, some warm and soft, some neither—and how we relate to them.

The first and the third interpretations derive morals from the poem, but the second and the fourth do not, at least not explicitly. The second and fourth colleagues interpreted the poem as providing a general commentary on particular aspects of life. We are not sure that the poem provides sufficient information for adjudicating these four alternatives. It is neither clear how to decide definitively whether the poem provides a moral or a general commentary nor, within either of these categories, which specific moral or general commentary is being provided. There are, we are certain, possible morals and general commentaries that could be eliminated as candidates. However, among these four, we have difficulty choosing.

The point we wish to make, however, is that in reading for understanding it often happens that definitive conclusions cannot be reached. Text is often ambiguous, sometimes intentionally so, as is often the case in poetry. Critical thinking theory provides a means of dealing with such ambiguity. It is to raise alternative interpretations, weed out interpretations to the extent that the available information will allow, and then to remain with multiple possibilities, either awaiting more information or accepting each as legitimate.

Explaining by Critical Thinking Theory the Difference between Steven’s and Colleen’s Reading

In explaining the difference between Steven’s and Colleen’s reading using schema theory, we worked on the assumption that Steven’s interpretation was better than Colleen’s, not simply different from it. The problem is whether this intuition can be supported within a theoretical framework from which it would follow that Steven’s reading is better than Colleen’s. We have seen that schema theory is partly successful in dealing with this problem. It allows us to say that Steven’s interpretation is better because he managed better than Colleen to fit the textual information into the schema he chose, and to tit it without forcing an interpretation. We have also seen, however, that schema theory has nothing to say about the relative merit of the different approaches to comprehension used by Steven and Colleen, loosely described as constructing-a-schema and activating-a-schema, respectively. A theory that can speak to the merit of these different approaches as well as to the fit of information would be more comprehensive and thus more acceptable than schema theory.

At a general level, a major difference between Steven’s and Colleen’s reading lies in the fact that Steven progressively constructs his ideas as he works through the text, while Colleen’s ideas show little progression. By this we mean that Steven iteratively refines his interpretation to seek greater clarity, consistency, and completeness, while Colleen is highly tolerant of ambiguity and vagueness, inconsistency, and incompleteness of interpretation. At a more specific level, Steven recognizes whether or not information is sufficient for reaching conclusions. When it is not sufficient, he uses his imagination to conjecture alternatives, while suspending judgment on which alternative is correct. He then negotiates these alternatives with evidence from the text and his background knowledge, all the time keeping in mind the story. Colleen, on the other hand, concludes from insufficient evidence. Typically, she offers one explanation when many others are plausible. While she does end up with a story, often she stretches the information to make vague and weak connections to her interpretation. As a consequence, she ignores new information, does not question the fit of the information with her interpretation, and provides no overall integration of the textual evidence and her background knowledge.

A perusal of the previous paragraph will show that the same criteria that distinguish Steven’s reading from Colleen’s distinguish good critical thinking from poor thinking as outlined in the preceding section. For example, both readers make inferences. The difference is that Steven makes inferences in accord with traditional accounts of good inference-making, whereas Colleen’s inference-making displays many widely known fallacies. For example, if we consider the two principles involving consistency and completeness, there is a marked difference. Colleen’s interpretation was both incomplete and inconsistent and she was either tolerant of that fact or, as we suspect, unaware of it. Steven, on the other hand, sought completeness and consistency by constantly adjudicating his evolving interpretation with respect to these two criteria.

In one sense, both children began at the same point, lack of knowledge of what the story was about. In another sense, they began at different points because Steven was disposed to think critically about the meaning of the text. He was disposed to remain open-minded; to seek alternative interpretations; to stick to the main point of trying to understand the story while at the same time dealing with the specific problems of interpreting particular words; to take a position and to change his position when the evidence warranted such action; and to take into account the total situation. In short, Steven had a way of getting from his position of lack of knowledge about the story to comprehension of the story. In broad outline, the way consisted of conjecturing, imagining, or supposing what might be the case; suspending judgment on those conjectures while searching for and awaiting sufficient evidence; and deciding to discard, modify, or sustain interpretations in the light of that evidence.

Colleen also had an approach to the problem of understanding the story and the approach was systematic. Her approach was to take certain pieces of information from the total information available and to add them in a linear fashion to form a chain of pieces of information. This approach was not going to lead to understanding because it provided no means for the separate pieces of information to interact. Since, by definition, a story is a unified set of information, rather than simply a concatenation of events, any approach to understanding the story that does not seek unity among all the information is bound to fail.

Critical thinking theory rates Steven’s approach more highly than Colleen’s independently of the conclusions they reached. Critical thinking theory provides a plan or a route for achieving understanding. The plan roughly is this: Be prepared to recognize that you may not understand; if you do not understand something, begin by imagining an explanation that would explain what you do not understand; be open to other possible explanations and seek them out; be prepared to withhold judgment on which explanation is correct while additional ideas are examined; evaluate the alternative explanations for consistency and completeness in light of available information; and choose to discard, modify, or accept explanations as a result of the evaluation. According to critical thinking theory, Steven’s reading is better than Colleen’s because he more nearly implemented this plan or process.


In a sense, much overlap can be seen between schema theory and critical thinking theory. To a point, both theories merely use different language for talking about the same phenomenon, understanding textual information. If schemata are taken to be the same as explanations then, by definition, having a schema is to understand. On this account, critical thinking theory would be consistent with schema theory, since thinking critically is a way of arriving at explanations.

However, in the discussion of problems with schema theory we suggested that schema theory’s orientation toward the products of reading rather than toward the process leads to some problems. In particular, schema theory provides no means for distinguishing the quality of reading processes apart from the quality of the products they produce. Such a distinction is necessary, we argued, because we know of products of the process of human understanding that we take to be false, but that were found using the best processes of understanding possible.

Schema theory does not provide a way of dealing with a fundamental problem of human understanding: How do you make progress when you do not know where you are going? Allow us to provide an example. Imagine a teacher and a student working together on a story. The teacher knows the story—has written it, let us suppose. From the perspective of the teacher, the reading of the student can be judged with respect to whether it is leading to a correct understanding of the story. However, from the perspective of the student, this can never be a criterion. The student, by hypothesis, does not know the story and, therefore, requires access to criteria for judging his or her reading that do not rely on having an understanding of the story. Schema theory, we think, provides a view of the reading process from outside, from a god’s eye view, from the perspective of someone who already understands. What is needed is a theory that provides a view from the inside, from the perspective of someone who does not understand, because the outsider’s perspective can never be used to advantage by the insider.

Critical thinking theory helps with this puzzle by being process-oriented, as opposed to the product orientation of schema theory. In the description of determining the meaning of the poem, several of the features of this process were revealed. The first thing that can be said about the process is that it is iterative. By this we mean that the process proceeds through a number of stages, with each stage aimed at providing a more refined interpretation. A stage consists of more or less the following set of steps, not necessarily in the order provided: Lack of understanding is recognized; alternative interpretations are created; judgment is suspended until sufficient evidence is available for choosing among the alternatives; available information is used as evidence; new information is sought as further evidence; judgments are made of the quality of interpretations, given the evidence; and interpretations are modified and discarded based on these judgments and, possibly, new alternative interpretations are proposed, sending the process back to the third step.

Second, the process of thinking critically is interactive. Interaction takes place among information in and about the text, the reader’s background knowledge, and interpretations of the text that the reader has created. These interpretations arise from an interaction of the text and the reader’s background knowledge but, once created, become separate entities in the comprehension process. This discrete existence lasts while judgment is being suspended on them. However, once an interpretation is accepted by the reader it becomes a part of his or her background knowledge and can serve in that capacity in further interpretations. The exact origin of such interpretations has been problematic since the time of Plato and we will not be able to offer any solutions here. Essentially, the problem is one of explaining how human beings can use their current knowledge to go beyond that knowledge. How do human beings understand things that are different from anything they presently know?

What is known is that if, as schema theory would have it, people can understand only what fits into the schemata they have, then there is no process available for coming to understand something nove1.23 While not providing a complete answer, critical thinking theory suggests that in coming to understand novel situations people use the information that is available, including their background knowledge, in creative and imaginative ways. They make progress by first realizing that what they know does not fit the current situation; by conjecturing what interpretation would or might fit the situation; and by suspending judgment on the conjectured interpretation until sufficient evidence is available for refuting or accepting it. Critical thinking theory thus provides for progression in the face of novel problems through the reader’s active use of imagination, and negotiation between what is imagined and available textual information and background knowledge. That is, one possible route to novel understanding is through the use of metaphorical thinking, but this only points to a possible solution, since metaphor is itself enigmatic.24

In order to carry out a negotiation between the products of imagination and available information, critical thinking theory is principled. The principles provide a means for determining how the conjectured interpretation is to be weighed and balanced with respect to the available information. The principles provide guidance in weighing and balancing, but they cannot provide definitive conclusions, that is, conclusions that do not admit, at least theoretically, of other possibilities. The word conjecture is purposely used to highlight the expansion of knowledge in the face of incomplete evidence, that is, in the face of uncertainty.

The first principle is that the conjectured interpretation is acceptable to the extent that it is complete. By this we mean that the interpretation explains all relevant textual information. The second principle is that the conjectured interpretation is acceptable to the extent that it is consistent with known facts. That is, the interpretation does not contain statements that are known to be false or require as assumptions statements that are known to be false. In short, the interpretation is plausible given what is known. Completeness and consistency are thus the two criteria for judging interpretations. Neither criterion by itself is sufficient; they must be used in tandem. In order to deal with situations where there are competing interpretations, the criteria must also be used comparatively. We must ask which interpretation is more complete, and more consistent, because often neither interpretation will be fully complete and fully consistent.

It is difficult in the context of the Steven and Colleen examples to see how one theory, schema theory or-critical thinking theory, is more effective than the other in distinguishing better from worse reading. This is so because the evidence in the story is sufficient for Steven both to use correct critical thinking processes and to arrive at the schema that most of us would agree best captures the story. The test comes when there is a situation in which equally legitimate processes lead to different conclusions. A case in point is the poem about whom to pet. Different readers using equally legitimate critical thinking processes could, as we illustrated, arrive at different interpretations of the poem. In fact, the most legitimate interpretation is that it is not precisely clear what the poet means, but that the poet may have intended one or several of a set of specific options. In such a case where there is no definitive end point or schema that best captures the meaning, the way in which the reader goes about constructing possible interpretations becomes preeminent in judging reading quality. The interpretations cannot be used by themselves in judging quality of reading because people reading equally well can arrive at different interpretations. What is important is the support people offer for their interpretations in the process they followed to reach them.

To see better what we are trying to say, comprehending a piece of text needs to be taken as a subset of achieving understanding in its broadest sense. Consider again the reference to Isaac Newton. We do not say that Newton was one of the greatest scientists of all time because of the conclusions he reached. We now take many of his conclusions to be false. Newton’s ideas were so good because they were the result of a very rigorous process in which they emerged as the best ideas on such topics as motion and optics, given what was known at the time. We can examine Newton’s ideas from the perspective of the knowledge of this century and find his ideas wanting. This is the god’s eye view of the outsider. However, in Newton’s day there was no other perspective but the one that knowledge of that day could provide. There was no god’s eye view of the right interpretations to be reached. The best Newton could do was to follow a rigorous thinking process.

Thus, not having a god’s eye view does not necessarily leave people stymied. Steven did not know from the start that the story was about fishing. In fact, he did not settle on fishing until near the end. Nevertheless, he could still proceed in a way that, in general, is known by experience to lead to good interpretations. That is, he proposed possibilities, tested them in competition with one another, and selected the one that performed best on those tests. Think now about Colleen. Without bringing the perspective of a FISHING schema to interpreting the story, she still should have seen the inconsistencies and incompleteness in her story and she still should have concluded that she was not understanding the story well. Instead, she appears to have had full confidence in her interpretation, and hardly even considered other possibilities. Her procedure, in contrast with Steven’s, is known by experience generally not to lead to good interpretations.

Our premise throughout this article is that Steven’s understanding is better than Colleen’s. Our task has been to say what makes it better. While most people may agree with this premise as it stands, they may also wish that credit be given to Colleen for what she has done. On this point we can make some concessions, but we also wish to resist being pushed too far. We concede that Colleen had an overall view of what she was to do, namely, come up with a story. She succeeded in this. We also concede that she showed considerable skill in fitting selected pieces of textual information into her shopping story. However, her skill at fitting information was really a skill at forcing information to fit the story theme she conceived from the outset, rather than allowing new information to challenge her original interpretation.

If people can understand only what fits their schemata, as schema theory would have us believe, then, if Colleen had no schema for FISHING, she could not be faulted for what she did and she could not be helped except by providing her a FISHING schema. However, we know that Colleen made mistakes: She jumped to conclusions on the basis of insufficient evidence; she failed to consider alternative interpretations of the evidence; she failed to take all relevant information into account; and she did not alter her views when the evidence warranted so doing. We also know that attempting to help Colleen by providing her a FISHING schema is at most a stopgap measure for this particular story. What is she to do the next time she faces a story for which she has no schema?

Colleen would have been better off had she realized that, when reading, one is allowed to be wrong, to change one’s mind, to wonder, to ask questions, and to fail to know. The fundamental problem here is most likely not Colleen, but the schooling that led her to conceive of reading the way she does. That is, Colleen needed instruction in critical thinking dispositions consistent with the nature of reading, which is not an exact process, which can move in fits and starts, and for which the goal is not to come up with merely an understanding but with the best understanding. The nature of arriving at understanding in reading is thus no different in principle from the nature of reaching understanding in general. Critical thinking theory is one attempt to provide a mechanism for how such understanding is reached.

We realize that, in offering critical thinking theory as an alternative explanation to the schema theory of good reading, many issues remain to be resolved. One of the most important is how good interpretations of texts might, in general, be distinguished from poor ones. On the surface, some people might hold that this is not an issue at all. However, if one is not prepared to maintain that some interpretations are better than others, then there are uncomfortable consequences. If no interpretations are better than others, then all interpretations are equally good and, therefore, any interpretation should suffice. We believe, however, that such a relativistic position must be avoided, or else one is led to the educational quagmire that says “anything goes.” Although a complete solution to this problem is not at hand, we offer the notion of good thinking described by critical thinking theory as a step toward a solution.

Another unresolved issue is whether the two theories are at all complementary. Schema theory seems to provide a plausible explanation of understanding in situations where the reader can fit information into an already existing schema. That is, schema theory seems to work tolerably well for cases of assimilation to existing structures. Even here, however, there is the problem of how the correct, existing structure is activated. When there is no existing schema to incorporate information, schema theory seems to have no method for constructing one—that is, it works less well for cases of accommodation.

Assuming that schemata are at least approximately equivalent to explanations, critical thinking theory provides methods both for activating existing schemata and for constructing new ones. According to critical thinking theory one arrives at explanations by supposing possible ones, weighing them against the evidence, and suspending judgment until sufficient evidence is in.

If schema theory is correct in saying that schemata are needed for comprehension, maybe a role for critical thinking theory is in saying how appropriate schemata are selected and constructed. Perhaps schema theory buttressed with critical thinking theory can offer a better understanding of reading comprehension.

Cite This Article as: Teachers College Record Volume 89 Number 2, 1987, p. 281-306
https://www.tcrecord.org ID Number: 530, Date Accessed: 5/22/2022 10:18:14 PM

Purchase Reprint Rights for this article or review
Member Center
In Print
This Month's Issue