DWB Site

Improving College Teaching
Using an Interactive, Compensatory
Model of Learning

Gregory Schraw
David Brooks
University of Nebraska-Lincoln
Lincoln, NE


Most college science teachers have little or no training in education. When they study materials written by educators, they often are overwhelmed by the jargon and underwhelmed by the utility of the material offered. At the same time, nearly all of us agree that good teachers make a difference, and that some teachers are better than others -- in given settings, and for certain clienteles. It's hard to have been a college teacher for more than a year, and not hear terms like motivation and intelligence used when discussing learning. This paper makes an attempt to use empirical knowledge about learning to create a useful synthesis for teachers. One might be surprised to learn that there are gaps -- often enormous gaps -- in our knowledge about learning. We don't try to jump or bridge these gaps. Instead, we go from what is known or, at least, seems to be known. Three notions will emerge repeatedly in our paper and our discussions. First, knowledge really is power. Knowledge is what really counts most! Next, it's the 'whole package' that counts, and not just ability or motivation. Finally, but perhaps most importantly, strength in any one of these areas can help to compensate for weakenss in another.


College students use a variety of integrated skills and attitudes to regulate their learning. This short course describes five general components as parts of an interconnected model of learning. We focus on cognitive abilities, an organized knowledge base, a repertoire of regulatory skills that includes strategies and metacognitive knowledge, and motivational beliefs. Cognitive abilities refer to one's general capacity to learn. The knowledge base refers to organized domain-specific and general knowledge in long term memory. Strategies refer to procedures that enable learners to solve specific problems. Metacognition includes knowledge about oneself as a learner and how to regulate one's learning. Motivation refers to beliefs about one's ability to successfully perform a task, as well as one's goals for performing a task.

The purpose of this short course is to describe the Interactive Compensatory Model of Learning (ICML) and discuss its implications for classroom instruction. We begin with a brief overview of the ICML, then discuss the relative importance of each of its five main components. We next consider how each of these components is related to other components in the model. Learners compensate for deficits in one component by using other components. We conclude with a discussion of instructional strategies that classroom teachers can use to improve each component and promote learning.

We emphasize the following key points throughout our discussion: a) effective learning depends on the dynamic interrelationship among a variety of learning skills, b) no single skill can totally support or interfere with self-regulated classroom learning, c) many skills make important contributions to learning, d) it is possible for most learners to compensate for weaknesses in one area using strengths in other areas, and e) it is possible to improve skills through classroom instruction. [Self-regulated learning: ability to explicitly understand and control all aspects of learning.]

The Interactive Compensatory Model

The purpose of the interactive compensatory model of learning (ICML) is to provide a framework for understanding and improving classroom learning. Though speculative, the model is consistent with a wide variety of empirical data. Figure 1 shows a schematic diagram of the ICML. The model includes five main components: cognitive abilities, knowledge, strategies, metacognition and motivation. Knowledge and regulatory skills such as strategies and metacognition are combined into one overarching module because of the close relationship among the three components included within it. We envision three interrelated modules within the model: cognitive abilities, knowledge and regulation, and motivational beliefs. Each of these modules includes a number of subcomponents. For example, the motivational beliefs module includes self-efficacy and attributional beliefs. We assume that each of the modules contributes directly or indirectly to learning, and compensates for potential deficits in other components. The focus of this paper is on classroom practice; we make no specific claims here regarding the location or structural properties of these modules in memory.

Figure 1. Interactive Compensatory Model of Learning (ICML)

The model postulates that cognitive ability is related to learning both directly and indirectly via knowledge and regulation. Strategies and metacognition typically co-develop and are quite strongly related. Knowledge and regulation are related to motivation. Cognitive ability is not related to motivation. Knowledge, regulation and motivation each are related directly to learning. In a later section of this paper, we provide a more explicit rationale for the proposed structural relationships among components.

Aspects of this model may seem to be oversimplified. One obvious problem is the isolation of some measureable entity we call learning. Our model assumes that, whatever learning takes place, it is integrated within the learner's components as suggested in Figure 2 (a dynamic figure that changes when viewed on the WWW).

Figure 2. Learning is Integrated into the Individual's 'Package.' (View using browser.)

Components Within the Model

This section provides a brief overview of each of the five main components in the ICML. Each of these components has an extensive research literature associated with it. We focus on the everyday educational applications of this literature.

Cognitive Ability

Researchers have been interested in the study of cognitive abilities for over 100 years. A number of theories have appeared in that time, and considerable debate remains as to which of these theories is best. For present purposes, we focus on three families of theories related to cognitive abilities. These include psychometric, modular, and componential theories, respectively.

Psychometric theories emerged early in the study of human abilities. These theories usually postulate one or two general components of intelligence. Writing at the turn of the century, Charles Spearmen proposed one general type of intelligence he referred to as g (i.e., a general intelligence factor). Spearman and many other early theorist's assumed that g was due to heritable differences among people. Spearman believed that g was instrumental in all intellectual endeavors across all possible domains. Many of Spearman's assumptions about g can be found today in the writings of Arthur Jensen (1992) and books such as The Bell Curve (Herrnstein & Murray, 1994).

Unlike Spearman, recent psychometric theories have distinguished between two types of g (Cattell, 1987). One type is commonly referred to as fluid g, and refers to abstract reasoning abilities that are unaffected by training and education. Examples of fluid g include abstract spatial and analogical reasoning, number series, and concept formation tasks. A second type of intelligence is known as crystallized g, which refers to knowledge and skills that are learned in formal settings. Examples include mathematical knowledge and reasoning skills, reading proficiency, vocabulary knowledge, and most types of domain knowledge. A large number of studies (see Carroll, 1993, and Gustafsson & Undheim, 1996, for recent reviews) support the fluid/crystallized intelligence distinction. Results suggest that fluid ability peaks in one's mid-twenties and remains stable until one's early seventies. Crystallized ability generally increases throughout one's life and shows no apparent point of decline.

Modular theories of intelligence typically propose several different kinds of intelligence that function independently of one another. Howard Gardner's (1983) theory of multiple intelligences provides a good example. In its initial form, Gardner proposed seven kinds of intelligence that were assumed to be physiologically and intellectually distinct from the remaining six intelligences. The seven intelligences included mathematical, verbal, spatial, kinesthetic, musical, interpersonal, intrapersonal abilities. Gardner has argued that each kind of intelligence is trainable in educational settings. Also, each intelligence must be trained separately because programs that promote one kind of intelligence (e.g., mathematical) may have little or no impact on other intelligences (e.g., musical).

Componential theories of intelligence typically propose several cognitive components that interact in a systematic way to regulate intellectual performance. Robert Sternberg's (1988) componential model is an example. Sternberg postulates three main components. Processing components include knowledge acquisition skills, regulatory knowledge such as metacognition, and automated performance skills needed to complete tasks such as reading and writing. Contextual components enable one to adapt to one's immediate environment. Experiential components determine the extent to which intelligence is changed and modified through experience. Experiential components change intelligence in part by introducing individuals to other skilled learners who can serve as mentors and provide explicit feedback that improves intellectual performance.

Generally, psychometric theories emphasize the role of nature over nurture. Modular theories suggest that nature contributes to a number of separate intelligences that, in turn, can be modified through nurture. Componential theories emphasize the primary role of nurture in intellectual development.

Brody (1992) provides a comprehensive recent review of theories of intelligence and empirical research supporting these theories. His review suggests the following: a) there is evidence for multiple forms of intelligence consistent with fluid and crystallized ability and the work of Robert Sternberg, b) there is little data to support a wide variety of independent intelligences as proposed by Howard Gardner, c) nature and nurture play approximately equal and important roles in intellectual performance, d) intelligence is malleable to a substantial degree, and e) intelligence is related to occupational and educational success.

Knowledge Base

Every task we undertake depends on knowledge; it is impossible to understand and perform a task without some degree of knowledge. Knowledge can be synthesized into broader conceptual structures such as schemata that enable us to think and reason at a more sophisticated level of understanding. [Schema: a mental framework in long term memory that helps learners organize knowledge and guides retrieval.] A third reason is that what we know enables us to construct new knowledge. Constructing higher-order knowledge is assumed to contribute in many important ways to self-regulation (Pressley, Harris & Marks, 1992).

At least two kinds of knowledge are important in learning. Domain-specific knowledge includes facts, concepts and organized knowledge structures such as schemata that are necessary to learn within a specific domain such as chemistry or physics. For many students, mathematics has an easy to identify domain-specific knowledge base that includes information related to algebra, geometry, trigonometry and calculus to name just a few. Much of this knowledge is acquired in a particular sequence that facilitates the learning of related, but more difficult, concepts.

Research on expertise indicates that domain-specific knowledge is the most important single component in effective learning, outstripping intelligence and other components (Ericsson, 1996). It typically takes five to ten years to produce a 'true expert.' Expertise consists of domain-specific knowledge at the factual, conceptual and theoretical levels. Being able to understand theories and use them as reasoning tools is especially important in higher-level understanding. Developing a deep understanding of a domain is due, in part, to the amount of time spent within the domain, deliberate practice, and mentoring from more advanced individuals (Ericsson, 1996).

General knowledge is broad knowledge that is not linked to a specific content domain but is appropriate to a wide variety of tasks or domains. Examples include knowledge about how to read, how to study effectively, knowledge of current events and general historical trends.

Most research focuses on the acquisition of domain-specific rather than general knowledge. As a consequence, it is not entirely clear how long it takes to acquire general knowledge or under what circumstances it is learned and constructed most effectively. Most general knowledge may be constructed tacitly rather than explicitly because it is generalized over a variety of times and settings (Seger, 1994; Sternberg & Wagner, 1994). [Tacit knowledge: knowledge that is learned implicitly and is not consciously accessible.]


Strategies refer to the mental tactics used to make a cognitive task easier to understand or perform. Experts place a high premium on strategic knowledge for one important reason: even a modest repertoire of strategies can improve learning and performance significantly (Pressley & Wharton-McDonald, 1997). In addition, strategy instruction increases positive motivational beliefs and may compensate for lack of intellectual ability or knowledge.

Research on strategy instruction has been an important part of educational research for two decades. Implementing and evaluating strategy instruction programs is extremely expensive and time consuming. Nevertheless, several recent reviews suggest that strategy intervention programs can be extremely effective ways to improve learning and self-regulation (Hattie, Briggs and Purdie, 1996; Rosenshine, Meister and Chapman, 1996). These reviews generally support the following claims:

Research also has addressed whether strategy instruction is more effective in teacher-centered versus student-centered learning environments. Neither type of setting appears to increase the effectiveness of the interventions. Specifically, Rosenshine et al. (1996) reported that student-centered approaches such as reciprocal teaching and cooperative learning environments were not more effective than direct instruction approaches.

Researchers also have considered what kinds of strategies are most important to teach. Hattie et al. (1996) compared rank orderings for approximately 25 different learning strategies across three cultures (i.e., Japanese, Japanese-Australian, and Australian). Results indicated that a handful of general learning strategies were rated as most important among all cultures. These included, in order of importance: self-checking, creating a productive physical environment, goal setting and planning, reviewing and organizing information after learning, summarizing during learning, seeking teacher assistance, and seeking peer assistance.


Metacognition refers to knowledge and regulatory skills people have about their own learning (Alexander, Carr & Schwanflugel, 1995; Schraw & Moshman, 1995). Since the term was first coined in the early 1970s, metacognition has been viewed as an essential component of skilled learning because it allows students to control a host of other cognitive skills. Metacognition is like the "mission control" of the cognitive system because it enables students to coordinate the use of extensive knowledge and many separate strategies to accomplish a single goal.

One of the clearest descriptions of metacognition is that of Ann Brown. According to Brown (1987), metacognition includes two related dimensions: knowledge of cognition, and regulation of cognition. Knowledge of cognition usually is assumed to include three components (Brown, 1987; Jacobs & Paris, 1987). Declarative knowledge refers to knowledge about ourselves as learners and what factors influence our performance. For example, most adult learners know the limitations of their memory system and can plan accordingly for a task based on this knowledge. Procedural knowledge refers to knowledge about strategies. For instance, most older students possess a basic repertoire of useful strategies, such as taking notes, slowing down for important information, skimming unimportant information, using mnemonics, summarizing main ideas, and periodic self-testing. Conditional knowledge refers to knowing when or why to use a strategy. One case in point is when you study differently for essay versus multiple-choice tests.

Brown has argued that knowledge of cognition is usually statable and late developing. Research suggests that these assumptions are reasonable when considering the metacognitive activity of older students but do not obtain for preadolescents (Flavell, 1992; Garner & Alexander, 1989). Paris and colleagues (Paris, Cross, & Lipon, 1984; Paris & Jacobs, 1984) found that instructional training programs enhance the development and use of metacognitive knowledge among elementary-age children. These findings indicate that metacognitive knowledge among younger students is not necessarily statable. Studies comparing expert performance among adults, however, are consistent with Brown's assumptions (cf. Glaser & Chi, 1988).

Regulation of cognition typically includes three components: planning, regulation, and evaluation (Jacobs & Paris, 1987; Kluwe, 1987). Planning involves the selection of appropriate strategies and the allocation of resources. Planning frequently includes setting goals, activating relevant background knowledge, and budgeting time. Regulation involves monitoring and self-testing skills necessary to control learning. Activities such as making predictions or pausing while reading, strategy sequencing, and selecting appropriate repair strategies also belong in the category. Evaluation involves appraising the products and regulatory processes of one's learning. Typical examples include reevaluating one's goals, revising predictions, and consolidating intellectual gains.

Brown has argued that regulatory processes, including planning, regulation, and evaluation, may not be conscious or statable in many learning situations. One reason is that many of these processes are highly automated, at least in adults. A second reason is that some of these processes have developed without any conscious reflection and therefore are difficult to report to others. In addition, Brown (1987) draws an important distinction about the relationship of age to metacognitive regulation and abstract reflection. She argues that regulatory mechanisms, such as planning, are independent of age, whereas reflection is not. Thus, like metacognitive knowledge, conscious use of regulatory processes may be related to limitations in one's ability to reflect rather than in one's ability to regulate.


Motivation as used here refers to a number of beliefs and attitudes that affect learning. It is now clear that students do not use existing knowledge and strategies effectively if they do not believe they will improve learning. A number of authors have distinguished recently between will and skill components of learning, where will refers to motivational beliefs that increase engagement and persistence, and skill components include the knowledge and strategies necessary to complete a task. We focus in this section of recent work on self-efficacy, goal orientations, and attributions as examples of will. Knowledge, strategies and metacognition, described earlier, provide examples of skill.

Self-efficacy refers to a judgment of one's ability to perform a task within a specific domain. High efficacy in one setting does not guarantee high efficacy in another. Within a specific domain, however, self-efficacy is linked strongly to a variety of behavioral outcomes such as engagement, persistence, strategy use, task performance, and reduced anxiety (Bandura, 1997; Pajares, 1996; Schunk, 1989).

High self-efficacy is associated with greater cognitive flexibility, resistance to negative feedback, and self-regulation in academic situations even when ability is controlled (Bandura, 1993; Zimmerman & Bandura, 1994). Higher levels of self-efficacy have been linked as well with the way individuals explain their success and failure in a particular situation (i.e., causal attributions)(Schunk, 1996). Self-efficacious individuals are more likely to attribute their failure to low effort rather than low ability, whereas low efficacy individuals attribute their failure to low ability.

Goal orientations refer to beliefs that individuals hold about their purposes for learning. Those with mastery goals, or what Dweck and Leggett (1988) refer to as learning goals, seek to improve their competence. These individuals are characterized by a desire to increase their knowledge and understand a topic better regardless of performance outcomes. Those with performance goals seek to prove their competence. These individuals are characterized by a desire to do better than others and to publicly demonstrate their skills and knowledge, but may have little desire to improve their understanding of a topic otherwise. A substantial body of research indicates that mastery-oriented students select adaptive behavioral responses in the face of frustration such as increased effort, greater persistence, and strategy shifting (Dweck & Leggett, 1988; Elliott & Dweck, 1988). In contrast, performance-oriented students select maladaptive behavioral responses such as self-aggrandizing statements or helplessness (Diener & Dweck, 1978).

Goal orientations are affected by beliefs about the changeability of intelligence (Ames, 1992; Blumenfeld, 1992). Dweck and Leggett (1988) proposed that incremental theorists (i.e., those who believe in malleable ability) develop mastery goals, whereas entity theorists (i.e., those who believe in fixed ability) develop performance goals. Ironically, goal orientation theory suggests that it is one's beliefs about the changeability of intelligence, rather than intelligence per se, that affects one's motivational engagement style.

Several recent studies indicate that goal orientations affect levels of self-efficacy and the type of attributional responses individuals make (see Schunk, 1996, for one such model). Mastery-oriented individuals make more internal attributions to controllable causes such as effort and report greater levels of hope (Roedel, Schraw & Plake, 1994). In contrast, performance-oriented individuals make more external attributions to uncontrollable causes such as ability and report less hope.

Attribution theory is the study of how individuals explain events that take place in their lives (Weiner, 1986). An attribution is a causal explanation of one of those events. Attributions vary along three dimensions that are associated with specific affective responses. The first dimension is locus of control, which defines the cause of an outcome as either internal or external to the individual. Mood, for example, is an internal cause even though it may be affected by external variables. The second dimension is stability, which pertains to the continuation of a perceived cause over time. Some causes (e.g., ability) usually are assumed to be stable, whereas other causes (e.g., luck) are completely unstable. The third dimension is controllability, which pertains to whether an individual has a high or low degree of control over the perceived cause. Some causes of success such as effort and strategy use are highly controllable; others such as ability or interest are not.

The locus of control dimension frequently is linked with the kind of affective responses individuals experience after an outcome. For example, pride and confidence are associated with internal causes of academic success such as ability, expert knowledge, and effort. The stability dimension usually is linked with a person's success expectancy. If success is attributed to a relatively stable trait such as ability or knowledge, it seems reasonable that past success would be repeated. The controllability dimension is related to the amount of effort and persistence an individual allocates to a task. Outcomes that are viewed as uncontrollable often promote anxiety and avoidance strategies, whereas those that are under control lead to increased effort and persistence.

One of the basic tenets of attribution theory is that the interpretation of an outcome (i.e., causal attribution) will determine the kind of behavioral response an individual makes. Attributions in which stability is the critical dimension frequently give rise to higher success expectancies and, in turn, higher levels of task engagement, challenge seeking, and performance. Attributions in which controllability is the critical dimension lead to greater effort and more persistence. Attributions in which internal locus of control is critical lead to feelings of confidence, satisfaction, and pride (Gredler, 1992).

Collectively, self-efficacy, goal orientations and attributions play an important role in learning by helping students engage and persist in difficult tasks. Research suggests that beliefs are changeable, and sensitive to classroom and teacher dynamics. Research also suggests a number of important compensatory relationships among beliefs, knowledge and strategies.

Why the ICML?

The ICML provides a framework for understanding the relationships among cognitive ability, knowledge, strategies, metacognition, motivation and learning. It is an empirically-based model that provides a comprehensive approach to learning. It includes all of the main components known to affect learning. More important, it provides a tentative basis for evaluating the relationships among these components.

The model helps educators better understand the strengths of relationships between each of the components. It provides a basis for understanding and utilizing compensatory relationships among the components, as well. It enables us to think of learning at a broader, systemic level that helps teachers deliver well-integrated instruction.

Compensation Among the Four Components

Compensation is, perhaps, the most important feature of this model. Learners are not set in stone. Having one component that is less than the 'best' or most efficacious can be compensated for by having strength in another component. This section provides a rationale for the proposed relationships among the components in Figure 1.

Figure 1 suggests that cognitive ability, knowledge and regulation, and motivation each affect learning directly. The ICML assumes that cognitive ability and motivation affect learning independently, but to roughly the same extent. Most experts agree that knowledge and regulation exert a strong direct effect on learning that is greater than the effects of either ability or motivational beliefs. Several recent summaries of skill acquisition and expertise highlight the role of deliberate practice in knowledge acquisition and learning (Ackerman, 1988, 1992; Ericsson, 1996). As students acquire knowledge, they construct strategies and metacognitive knowledge that enables them to regulate their learning more effectively.

Of greatest interest in the ICML is our claim that components compensate for one another. Figure 1 suggests the following: a) ability compensates in part for knowledge and regulation, b) knowledge and regulation compensate for cognitive ability and motivation, and c) motivation compensates for ability, knowledge and regulation. Many current theories support the notion of compensatory processes (e.g., Gardner, 1983; Perkins, 1987; Sternberg, 1994). Compensatory processes are important because they enable students to use their strengths in one area to overcome their weaknesses in other areas. One implication is that most students can perform at significantly higher levels of achievement and satisfaction if they use their existing skills in an optimally flexible way.

Cognitive Ability and Other Variables

Cognitive ability affects learning directly and indirectly through knowledge and regulation. Empirical studies offer a wide variety of estimates regarding the strength of relationship between ability and learning. Brody (1992) provides a number of examples of correlations in the .30 to .40 range. Similar figures are reported in recent books such as The Bell Curve (Herrnstein & Murray, 1994) that suggest a strong role for ability in human learning. In contrast, Brody also reports correlations among social class, father's education and learning in the .30 to .40 range; ability per se does not appear to constrain performance more than social and demographic variables.

The ICML assumes that ability affects learning indirectly via knowledge and strategy development. The main idea is that cognitive ability constrains the amount and sophistication of what individuals can learn and, therefore, how well they perform. Recently, a number of authors (e.g., Alexander et al., 1995) have argued that cognitive ability should be viewed as a threshold variable that weakly limits knowledge and strategy acquisition. Other authors (e.g., Ericsson, 1996) discount the role of cognitive ability altogether in the development of expertise. For example, while estimates of ability have risen more than one-half standard deviation in the last 30 years, measures of achievement such as the ACT have declined during the same period. These conflicting trends clearly suggest that ability, in and of itself, does not guarantee knowledge acquisition and improved performance. The extent to which ability may be necessary is even less clear.

Based on the reviews of Alexander et al. (1995) and Brody (1992), we provide an estimated correlation between ability and knowledge in the r = .20 to .40 range. A correlation of this size suggests that ability plays a modest positive role in knowledge acquisition. Students of higher ability presumably find it easier to learn for two reasons; that is, faster information processing in working memory and less overall effort (Jurden, 1995). However, students of average ability are able to compensate for speed differences through additional effort.

Available evidence suggests that cognitive ability has an important, direct impact on learning. Less clear is the extent to which ability affects learning indirectly by constraining the acquisition of knowledge, strategies and metacognition. These relationships appear to be weak, especially in the case of strategies and metacognition (Alexander et al., 1995: Pressley & Ghatala, 1990). In one study, Swanson found no relationship between ability and metacognition among sixth-graders. In contrast, students in Swanson's study used both strategies and metacognitive knowledge to compensate for differences in ability.

Surprisingly, cognitive ability does not appear to be related strongly to motivational beliefs either. Dweck and Leggett (1988), for example, proposed that goal orientations are independent of measured ability. This hypothesis has been confirmed in a number of recent empirical studies (Roedel et al., 1994). A number of recent studies and review papers also suggest that ability is not strongly related to self-efficacy judgments (Pajares, 1996; Schunk, 1989; Zimmerman & Bandura, 1994), even though ability is related positively to task performance (Stajkovic & Luthans, 1998; Bandura, 1997).

For present purposes, we have excluded an explicit link in Figure 1 between cognitive ability and motivational beliefs to highlight the potentially independent relationship between them. The lack of an explicit link has two important implications. One is that ability does not affect motivation directly. A second implication is that motivation contributes to learning over and above the effect of ability.

Knowledge, Strategies and Metacognition

Many studies indicate a strong relationship among knowledge, strategies and metacognition. Regarding knowledge and strategies, empirical studies consistently show a positive relationship between knowledge, strategy use and deeper cognitive processing. For example, experts in all domains use more strategies and use them much more selectively than novices (Ericsson, 1996). This relationship holds even when initial differences in ability are controlled (Charness, 1991). High-knowledge learners also use more higher-order reasoning skills to construct sophisticated mental representations and solve complex problems (Bereiter & Scardamalia, 1993; Kintsch, 1998).

Regarding the relationship between knowledge and metacognition, a number of studies report a strong, positive relationship (Garner, 1987). For example, Glenberg and Epstein (1987) reported correlations in the .30 to .40 range between domain knowledge in physics and music and monitoring accuracy. Strategies and metacognition also are related quite strongly. In general, high-metacognition students use more strategies compared to low-metacognition students (Garner & Alexander, 1989; Pressley & Ghatala, 1990), use more sophisticated strategies (Schraw et al., 1994), and use them with greater flexibility (Swanson, 1990). A number of strategy intervention programs have found that scaffolded strategy instruction typically improves metacognitive awareness (Paris et al., 1984; Pressley and Wharton-McDonald, 1997). Spontaneous strategy construction also is linked to greater metacognitive awareness (Siegler, 1996).

Empirical findings generally suggest a strong compensatory relationship among knowledge, strategies and metacognition. Strategy instruction almost certainly improves metacognitive awareness and greatly facilitates the acquisition of knowledge (Pressley et al., 1987). Increasing one's knowledge within a domain also facilitates the revision of old strategies and the construction of new strategies (Siegler, 1996).

Knowledge, Regulation and Motivation

A strong relationship appears to exist between knowledge, regulation and motivational beliefs. For example, explicit strategy instruction increases self-efficacy (Schunk, 1996; Zimmerman, Greenberg & Weinstein 1994) and attributions to internal, controllable causes such as effort (Graham & Weiner, 1996). Students with strong mastery orientations report more strategy use and use more sophisticated strategies (Schraw et al., 1994).

Current research indicates that motivational beliefs such as self-efficacy and mastery goals facilitates knowledge acquisition because students are more apt to engage and persist in a challenging task (Bruning, Schraw & Ronning, 1999). There also is evidence that increased knowledge enhances self-efficacy (Bandura, 1997; Schunk, 1989) and attributions to controllable causes such as effort (Graham & Weiner, 19xx). For these reasons, the arrow connecting knowledge and regulation with motivational beliefs in Figure 1 is bidirectional. This suggests that motivational beliefs have an important compensatory effect on knowledge and regulation, as well as the reverse.

Classroom Applications for College Teachers

Effective learning includes a number of autonomous components that compensate for each other. Each component is malleable and subject to change in the classroom. This section describes a number of strategies for improving each of the components.

Improving Cognitive Ability

Debate continues as to the extent to with teachers or environmental factors such as peers or cooperative groups can change cognitive ability. Some experts, but particularly psychometric theorists, believe that cognitive ability is unitary (e.g., g) and cannot be changed in any significant way (Herrnstein & Murray, 1994; Jensen, 1992). This view is the exception rather than the rule, however. Most educators and researchers believe that ability is somewhat changeable. For example, in Cattell's (1987) two-factor model of intelligence, crystalized ability is assumed to increase as a function of formal education. Fluid ability, which is usually considered less changeable, is thought by many to be at least somewhat changeable. For example, Lohman (1993) has argued that fluid ability can be improved substantially by teaching critical thinking and reasoning skills across the full educational curriculum.

Componential theories strongly support the notion of incremental change in cognitive ability. Sternberg (1986) emphasizes the role of adaption in which individuals select and shape their environments as strategically as possible. A variety of programs have been developed and tested over the past 20 years to do so. These include the Productive Thinking Program (Covington, Crutchfield, Davies, & Olton, 1974); the IDEAL Problem Solver (Bransford & Stein, 1984); the CoRT Thinking Materials (de Bono, 1973); and the Feuerstein Instrumental Enrichment (FIE) Program (Feuerstein, Rand, Hoffman, & Miller, 1980).

Feuerstein's Instrumental Enrichment (FIE) system centers on what Feuerstein (Feuerstein et al., 1980) calls "mediated-learning experiences" (MLEs). Mediated-learning experiences provide activities that teach learners to interpret their experience. MLEs are deliberate interventions by teachers, parents, or others designed to help learners interpret and organize events. The basic task of the MLE is to teach the child to play an active role in critical thinking and, ultimately, to think and solve problems independently.

FIE affords a series of exercises (called "instruments" by Feuerstein) that provide the context for learning. At present, fourteen or fifteen instruments, arranged in order of increasing complexity, are available for ten- to eighteen-year-old students. The program is designed to be taught three to five times per week for two or three years. The exercises are paper-and-pencil activities designed to help the student identify problem-solving procedures and permit the teacher to "bridge" from the activities (problems) to subject matter of interest to student and teacher. Most FIE lessons provide "practice" exercises carried out under teacher supervision to provide feedback to students in their attempts to identify and evaluate the strategies used in solution attempts. FIE also provides a language for teaching problem-solving concepts such as planning, strategy choice, evaluation, and the like. Each instrument is designed to have wide generality. A special feature of this program is the deliberate focus on instruction of special populations. Thus, it has been used with youngsters who have mental retardation, learning disabilities, behavioral disorders, and hearing impairment.

Bransford, Arbitman-Smith, Stein, and Vye (1985) and Savalle, Twohig, and Rachford (1986) evaluated the effectiveness of Feuerstein's Instrumental Enrichment (FIE). On the basis of a wide range of evaluation studies conducted in Israel, Venezuela, the United States, and Canada, students exposed to the FIE program performed better than control groups on tests such as the Raven's Progressive Matrices and some achievement subtests, such as mathematics. The effects were found with a wide variety of student types. However, a number of studies also have shown no significant difference as a result of FIE training. In general, features of successful studies were the presence of well-trained FIE teachers and student instruction that lasted eighty hours or more.

Modular theories such as Howard Gardner's (1983) multiple intelligences also emphasize the flexible nature of intelligences. Garner suggests that each intelligence can be improved through systematic instruction and embedded classroom activities. Less clear in Gardner's model is the extent to which gains in one area transfer or compensate for weaknesses in other areas. Recent reviews of the transfer literature suggest that there is some reason for optimism in that transfer across domains is observed provided that teachers actively promote such transfer (Cox, 1997; Mayer & Wittrock, 1996).

Improving Knowledge

Classroom instruction increases knowledge. The debate is not whether knowledge is changeable, but how to change knowledge most effectively in a limited amount of time and in ways that promote deeper conceptual understanding. Three general questions arise with respect to changing knowledge. What kind of knowledge does one want to develop? Factual declarative knowledge can be changed in a matter of weeks with well organized instruction. A broad conceptual understanding of a domain may take years to develop, however. Most educators emphasize the importance of helping students construct integrated schemata that enable them to understand the big picture in a domain. To become knowledgeable within a domain, students must possess a broad array of declarative, procedural and conceptual knowledge (Shuell, 1996).

What type of instruction best facilitates knowledge development? Most educational debates focus on three main approaches: direct, socially mediated, and autonomous learning (Greeno, Collins & Resnick, 1996). Direct instruction includes teacher led classrooms as well as learning environments such as laboratories in which declarative and procedural knowledge are modelled for students. Socially mediated learning includes a variety of student-centered approaches such as cooperative learning in which a small group of students work together with minimal assistance from teachers. Autonomous learning enables students to work independent of teachers and other students.

Research suggests that all three approaches can be effective (Shuell, 1996). Direct instruction enables teachers to convey a large body of knowledge quickly and efficiently to students. Socially mediated learning helps students use that knowledge to solve problems, and to modify that knowledge to fit new problems. Autonomous learning helps students reflect on new ways to apply that knowledge.

What instructional practices best promote deeper learning and expertise? Research suggests it takes about five to ten years, or 10,000 hours, to become an 'expert' (Ericsson, 1996; Ericsson & Smith, 1991)! One reason why expertise develops so slowly is that much of the declarative and procedural knowledge needed to master a domain is acquired tacitly over a long period of time (Bereiter & Scardamalia, 1993; Wagner & Sternberg, 1985). Evidence suggests that much of our knowledge is acquired tacitly even when we receive a great deal of formal training in a domain. Because of this, even highly skilled experts often find it difficult to describe what it is they know about a body of knowledge and, as a consequence, may be poor decision makers when forced to reflect on their knowledge (Johnson, 1988).

In our opinion, the best way to address this problem is to employ all three general instructional methods. It is essential for skilled experts to codify their knowledge into useable schemata for their students. In addition, it is necessary to model expert reasoning skills (Kuhn, 1991). Students will need time to reflect on what they are learning with other students. Cooperative activities provide an opportunity for students to integrate new knowledge and practice new procedural skills until they are automatized. Equally important is the opportunity to work alone to modify new knowledge and fine-tune difficult procedural skills.

Improving Strategies and Metacognition

Many educators advocate direct strategy instruction as a means to improve learning and help students compensate for other weaknesses. Strategy instruction is ideally suited for the classroom because it can be done quickly and efficiently compared to the time and effort needed to change ability and expert knowledge (Perkins, Faraday & Bushey, 1991; Pressley et al., 1987). Although strategies do not substitute for cognitive ability and knowledge, they often are able to compensate for lower levels of ability and knowledge.

College teachers should make explicit strategy instruction a priority. Research indicates that it is the strategic use, rather than the mere possession of knowledge that improves learning. Teaching students strategies not only improves their learning but also empowers them by increasing self-efficacy. We recommend that teachers target "age and content-appropriate" strategies by comparing the most and least successful students in each class. The chances are good that highly successful students rely on strategies that struggling students do not use. Modeled instruction with these strategies should enhance students' ability to regulate their learning.

Pressley and Wharton-McDonald (1997) recommend strategy instruction that addresses learning needs including those needed before, during, and after the main learning episode. Strategies that occur before learning include setting goals, determining how much information to learn, deciding how this information relates to what one already knows, and anticipating how the to-be-learned information will be used. Strategies needed during learning include identifying important information, predicting, monitoring, analyzing, and interpreting. Strategies used after learning include reviewing, organizing, and reflecting. Good strategy users possess some degree of competence in each of these areas.

Teachers should consider how to sequence strategy instruction as well. The following draws heavily on the work of Palincsar and Brown (1984), Pressley et al. (1992), and Poplin (1988).

Improving Motivation

A large number of studies suggest that motivational beliefs are subject to change (Pintrich & Schunk, 1996; Stipek, 1993). Changes occur for several reasons including student awareness, teacher modelling of adaptive beliefs, and changes in classroom processes such as evaluation and grading. Here are some ways to change self-efficacy, goal orientations, and attributions.

According to Bandura (1997), self-efficacy is affected by self-assessments, behavioral feedback, and environmental cues. Increase students' awareness of the self-efficacy concept. Emphasize the positive consequences of high efficacy by describing how efficacy develops and deteriorates, and promote positive efficacy messages in the classroom. Peer and teacher modeling also increases self-efficacy. In many settings, peer models are most effective because they are judged to be most similar to students. Provide feedback; behavioral and environmental feedback are two of the most important influences on self-efficacy. Feedback is most effective when it relates performance outcomes to activities that cause those outcomes.

Dweck and Leggett's (1988) theory suggests that goal orientations are due to attitudes about the changeability of cognitive ability. Those who believe that ability is changeable report more satisfaction with school and persist longer on a difficult task without succumbing to frustration. Promote the view that intellectual development is controllable. Emphasize the process rather than the products of learning. Focusing on the process of learning highlights its incremental nature, whereas focusing on the products emphasizes the outcome of that process. Reward effort and improvement; deemphasize native ability to help students develop a stronger mastery orientation. Stress that mistakes are a normal (and healthy) part of learning. Everyone makes mistakes when learning a new skill. How teachers respond to these mistakes sends a powerful message to students. When mistakes are viewed positively, receive corrective attention, and are used to provide feedback to students, students learn more than when mistakes are viewed in a negative light (Poplin, 1988). Encourage individual rather than group evaluative standards. Much of the evaluation that occurs in education, especially among high school and college students, is norm referenced (each student's performance is compared with the group's average performance). Group-based grading may lead to the adoption of a performance orientation, given that each student is compared directly with the group norm. In contrast, encouraging individual standards (e.g., portfolio evaluation) is more likely to promote the development of a learning orientation.

Help students better understand the learning process by explaining the role of attributions. Some students struggle unnecessarily because they incorrectly attribute failure to low ability rather than to either lack of effort or undirected effort. Helping students understand their attributions is a valuable first step in changing their beliefs. One way to do so is through attributional retraining. A review by Försteling (1985) found that the majority of attributional retraining programs are quite successful. The general sequence is as follows: (1) individuals are taught how to identify undesirable behaviors, such as task avoidance, (2) attributions underlying avoidant behavior are evaluated, (3) alternative attributions are explored, and (4) favorable attributional patterns are implemented.

Försteling reported that most programs emphasize a shift in unfavorable attributions based on ability to favorable attributions based on effort. Shifting attributions from ability to effort appears to be effective because effort is a controllable variable, whereas ability is not. Programs adopting this strategy frequently report an increase in task persistence and achievement levels.

In a series of studies by Schunk and colleagues (Schunk, 1983, 1987; Schunk & Cox, 1986), attributional feedback provided to students while they were engaged in a task increased self-efficacy and performance. Feedback about effort frequently improved task persistence, especially when it was given early in the learning cycle. To be effective, however, attributional feedback must be credible.

In contrast, Schunk (1984) found that sometimes feedback about ability has a stronger effect than feedback about effort. One group of students in this study received effort-only feedback, a second group received ability-only feedback, and two other groups received either effort-ability or ability-effort feedback. Schunk found that those who received ability feedback before effort feedback performed better and reported higher self- efficacy than those who received effort-only feedback or those who received effort feedback prior to ability feedback. Apparently, information regarding ability is linked more closely with one's sense of efficacy than is information about effort.

Overall, the attributional retraining literature provides clear evidence that increasing students' awareness about the attributions they make and helping them make more favorable attributional responses improve self-efficacy and learning while reducing achievement-related anxiety. For this reason, we believe that teachers should discuss the role of attributions in learning and provide some degree of retraining for students who make inappropriate attributions.

We recommend four general strategies for promoting motivational change. One is to increase student awareness about the importance of motivational beliefs and classroom influences on these beliefs. A second is to encourage teacher and student modeling of skills and beliefs. Modeling skills such as strategies helps students construct positive attitudes as their skills improve. Modeling positive beliefs also helps students better understand the factors that enable successful students to reach challenging academic goals. A third strategy is to emphasize controllable factors such as strategy use, effort, and help seeking. Students who focus on controllable factors are more apt to engage and persist in challenging tasks (Graham & Weiner, 1996 ). Last, teachers should keep classroom mistakes in proper perspective. Mistakes can be viewed in one of two ways; as indices of incompetence, or as an opportunity for further improvement. Focusing on improvement (i.e., mastery) has many positive benefits for students.


The purpose of this paper was to describe the Interactive Compensatory Model of Learning (ICML) and discuss its implication for classroom instruction. The model includes cognitive ability, knowledge, strategy, metacognition and motivational components. Our main claim is that each component contributes directly to classroom learning. In addition, each of the components potentially compensates for weaknesses in other components.

We emphasized the following key points throughout our discussion: a) effective learning depends on the dynamic interrelationship among a variety of learning skills, b) no single skill can totally support or interfere with self-regulated classroom learning, c) each skill makes an important contribution to learning, d) it is possible for most learners to compensate for weaknesses in one area using strengths in other areas, and e) it is possible to improve all of these skills via classroom instruction.

Despite claims to the contrary, it is foolish to believe that academic success hinges exclusively on one of the five skill areas described in the ICML. Some authors such as Jensen (1992) and Herrnstein and Murray (1994) have attempted to argue that cognitive ability plays a prohibitive role in academic achievement. Nevertheless, others have argued persuasively against this view. Most evidence suggests that ability plays an important role in achievement that is augmented by knowledge, regulatory skills and motivational beliefs. Our view on the matter is that students who possess a moderate degree of skill in all areas and use these skills flexibly will have little difficulty succeeding in even challenging academic situations.

No single skill can do all the brain's work for it. We encourage college instructors to improve as many skills in their students as they can. Achieving some degree of expertise in each of the five skill areas is probably more helpful to students than achieving a very high level of expertise in any one of them. Each of the five areas is necessary to succeed. Each contributes some unique aspect to learning.

We feel quite strongly that it is possible for most learners to compensate for weaknesses in one area using strengths in other areas. Currently, there are a number of studies supporting this view and none that we know of that refute it. As Pressley et al. (1987) emphasized over a decade ago, self-regulated learners use a wide variety of skills in a versatile way.

We also believe that it is possible to improve all of these skills via classroom instruction. Some skills no doubt are more amenable to instruction than others. Nevertheless, all the skills described in this paper, including cognitive ability, can be changed given a supportive environment and the will to do so on the part of the learner.


Ackerman, P. L. (1988). Determinants of individual differences during skill acquisition: Cognitive abilities and information processing. Journal of Experimental Psychology: General, 117, 288-318.

Alexander, J. M., Carr, M., & Schwanenflugel, P. J. (1995). Development of metacognition in gifted children: Directions for future research. Developmental Review, 15, 1-37.

Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261-271.

Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28, 117-148.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman.

Bereiter, C., & Scardamalia, M. (1993). Surpassing ourselves: An inquiry into the nature and implications of expertise. Chicago: Open Court.

Blumenfeld, P. C. (1992). Classroom learning and motivation: Clarifying and expanding goal theory. Journal of Educational Psychology, 84, 272-281.

Bransford, J. D., & Stein, B. S. (1984). The IDEAL problem solver. San Francisco: Freeman.

Bransford, J. D., Arbitman-Smith, R., Stein, B. S., & Vye, N. J. (1985). Improving thinking and learning skills: An analysis of three approaches. In J. W. Segal, S. F. Chipman, & R. Glaser (Eds.), Thinking and learning skills: Relating instruction to basic research (Vol. 1, pp. 133-206). Hillsdale, NJ: Erlbaum.

Brody, N. (1992). Intelligence. (2nd ed.). San Diego: Academic Press.

Brown, A. (1987). Metacognition, executive control, self-regulation, and other more mysterious mechanisms. In F. Weinert & R. Kluwe (Eds.), Metacognition, motivation, and understanding (pp. 65-116). Hillsdale, NJ.: Lawrence Erlbaum.

Brown, R., Pressley, M., Van Meter, P., & Schuder, T. (1996). A quasi-experimental validation of transactional strategies instruction with low-achieving second-grade readers. Journal of Educational Psychology, 88, 18-37.

Bruning, R. H., Schraw, G. J., & Ronning, R. R. (1999). Cognitive psychology and instruction. (3rd ed.). Englewood Cliffs, NJ: Merrill.

Butler, D., & Winne, P. (1995). Feedback and self-regulated learning: A theoretical synthesis. Review of Educational Research, 65, 245-281.

Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. Cambridge, England: Cambridge University Press.

Cattell, R. B. (1987). Intelligence: Its structure, growth, and action. Amsterdam: North-Holland.

Charness, N. (1991). Expertise in chess: The balance between knowledge and search. In K. A. Ericsson & J. Smith (Eds.), Toward a general theory of expertise. Cambridge, UK: Cambridge University Press.

Covington, M. C., Crutchfield, R. S., Davies, L. B., & Olton, R. M. (1974). The productive thinking program: A course in learning to think. New York: Merrill/Macmillan.

de Bono, E. (1973). CoRT thinking materials. London: Direct Education Services.

Diener, C. I., & Dweck, C. S. (1978). An analysis of learned helplessness: Continuous changes in performance, strategy, and achievement cognitions after failure. Journal of Personality and Social Psychology, 36, 451-462.

Dweck, C. S., & Leggett, E. S. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256-273.

Elliott, E. S., & Dweck, C. S. (1988). An approach to motivation and achievement. Journal of Personality and Social Psychology, 54, 5-12.

Ericsson, K. A. (1996). The acquisition of expert performance. In K. A. Ericssonn (Ed.), The road to excellence: The acquisition of expert performance in the arts, science and games (pp. 1-50). Mahwah, NJ: Lawrence Erlbaum.

Ericsson, K. A., & Smith, J. (1991). Toward a general theory of expertise: Prospects and limits. Cambridge, England: Cambridge University Press.

Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The Role of Deliberate Practice in the Acquisition of Expert Performance. Psychological Review, 100(3), 363-406.

Feuerstein, R., Rand, Y., Hoffman, M. B., & Miller, R. (1980). Instrumental enrichment: An intervention program for cognitive modifiability. Baltimore: University Park Press.

Flavell, J. H. (1992). Perspectives on perspective taking. In H. Beilin & P. Pufall (Eds.), Piaget's theory: Prospects and possibilities (pp. 107-139). Hillsdale, NJ: Erlbaum.

Försteling, F. (1985). Attributional retraining: A review. Psychological Bulletin, 98, 495-512.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic Books.

Garner, R. (1987). Metacognition and reading comprehension. Norwood, NJ: Ablex Publishing.

Garner, R., & Alexander, P. A. (1989). Metacognition: Answered and unanswered questions. Educational Psychologist, 24, 143-158.

Glaser, R., & Chi, M. T. (1988). Overview. In M. Chi, R. Glaser, & M. Farr (Eds.), The nature of expertise (pp. xv-xxviii). Hillsdale, NJ: Erlbaum.

Glenberg, A. M., & Epstein, W. (1987). Inexpert calibration of comprehension. Memory & Cognition, 15, 84-93.

Graham, S., & Weiner, B. (1996). Theories and principles of motivation. In D. C. Berliner & R. C. Calfee (Eds.), The handbook of educational psychology (pp. 63-84). New York, NY: Macmillan.

Gredler, M. E. (1992). Learning and instruction: Theory into practice. (2nd ed.). New York: Macmillan.

Greeno, J. G., Collins, A. M., & Resnick, L. (1996). Cognition and learning. In D. C. Berliner & R. C. Calfee (Eds.), The handbook of educational psychology (pp. 726-764). New York, NY: Macmillan.

Gustafsson, J. E., & Undheim, J. O. (1996). Individual differences in cognitive functions. In D. C. Berliner & R. C. Calfee (Eds.), The handbook of educational psychology (pp. 186-242). New York, NY: Macmillan.

Hattie, J., Biggs, J., & Purdie, N. (1996). Effects of learning skills interventions on student learning: A meta-analysis. Review of Educational Research, 66, 99-136.

Herrnstein, R., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life. New York: Fress Press.

Horn, C. A., & Bruning, R. H. (1993). Achievement in Biology: An exploration of the rolesof ability, motivation, and strategy use in the construction of knowledge and course achievement. Paper presented at the American Educational Research Association.

Jacobs, J. E., & Paris, S. G. (1987). Children's metacognition about reading: Issues in definition, measurement, and instruction. Educational Psychologist, 22, 255-278.

Jensen, A. R. (1992). Understanding g in terms of information processing. Educational Psychology Review, 4, 271-308.

Johnson, E. J. (1988). Expertise and decision under uncertainty: Performance and process. In M. Chi, R. Glaser, & M. Farr (Eds.), The nature of expertise (pp. 209-228). Hillsdale, NJ: Erlbaum.

Jurden, F. H. (1995). Individual differences in working memory and complex cognition. Journal of Educational Psychology, 87, 93-102.

Kintsch, W. (1998). Comprehension: A paradigm for cognition. Cambridge, England: Cambridge University Press.

Kluwe, R. H. (1987). Executive decisions and regulation of problem solving. In F. Weinert & R. Kluwe (Eds.), Metacognition, motivation, and understanding (pp. 31-64). Hillsdale, NJ.: Lawrence Erlbaum.

Kuhn, D. (1991). The skills of argument. New York: Cambridge University Press.

Lohman, D. F. (1993). Teaching and testing to develop fluid abilities. Educational Researcher, 22, 12-22.

Mayer, R. E., & Wittrock, M. C. (1996). Problem solving transfer. In D. C. Berliner & R. C. Calfee (Eds.), The handbook of educational psychology (pp. 47-62). New York: Macmillan.

Palincsar, A. S., & Brown, A. (1984). Reciprocal teaching of comprehension fostering and monitoring activities. Cognition and Instruction, 1, 117-175.

Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66, 543-578.

Paris, S. G., Cross, D. R., & Lipon, M. Y. (1984). Informal strategies for learning: A program to improve children's reading awareness and comprehension. Journal of Educational Psychology, 76, 1239-1252.

Paris, S. G., & Jacobs, J. E. (1984). The benefits of informed instruction for children's reading and comprehension. Child Development, 55, 2083-2093.

Perkins, D. N. (1995). Outsmarting IQ: The emerging science of learnable intelligence. New York, NY: Free Press.

Perkins, D. N., Faraday, M., & Bushey, B. (1991). Everyday reasoning and the roots of intelligence. In J. F. Voss, D. N. Perkins, & J. W. Segal (Eds.), Informal reasoning and education (pp. 83-106). Hillsdale, NJ: Erlbaum.

Pintrich, P. R., & Schunk, D. H. (1996). Motivation in education: Theory, research, and applications. Englewood Cliffs, NJ: Prentice Hall.

Poplin, M. S. (1988). Holistic/constructivist principles of the teaching/learning process: Implications for the field of learning disabilities. Journal of Learning Disabilities, 21, 401-416.

Pressley, M., Harris, K. R., & Marks, M. B. (1992). But good strategy instructors are constructivists! Educational Psychology Review, 4, 3-31.

Pressley, M., Borkowski, J. G., & Schneider, W. (1987). Cognitive strategies: Good strategy users coordinate metacognition and knowledge. In R. Vasta & G. Whitehurst (Eds.), Annals of Child Development (Vol. 5, pp. 89- 129). Greenwich, CT: JAI Press.

Pressley, M., & Ghatala, E. S. (1990). Self-regulated learning: Monitoring learning from text. Educational Psychologist, 25, 19-33.

Pressley, M., & Wharton-McDonald, R. (1997). Skilled comprehension and its development through instruction. School Psychology Review, 26, 448-466.

Roedel, T. D., Schraw, G., & Plake, B. S. (1994). Validation of a measure of learning and performance goal orientations. Educational and Psychological Measurement, 54, 1013-1021.

Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching students to generate questions: A review of the intervention studies. Review of Educational Research, 66, 181-221.

Savalle, J. M., Twohig, P. T., & Rachford, D. L. (1986). Empirical status of Feuerstein's "instrumental enrichments" (FIE) technique as a method of teaching thinking skills. Review of Educational Research, 56, 381-409.

Schraw, G., & Moshman, D. (1995). Metacognitive theories. Educational Psychology Review, 7, 351-373.

Schunk, D. H. (1983). Ability versus effort attributional feedback: Differential effects on self-efficacy and achievement. Journal of Educational Psychology, 75, 848-856.

Schunk, D. H. (1984). Sequential attributional feedback and children's achievement behaviors. Journal of Educational Psychology, 76, 1156-1169.

Schunk, D. H. (1987). Peer models and children's behavioral change. Review of Educational Research, 57, 149-174.

Schunk, D. H. (1989). Self-efficacy and achievement behaviors. Educational Psychology Review, 1, 173-208.

Schunk, D. H., & Cox, P. D. (1986). Strategy training and attributional feedback with learning-disabled students. Journal of Educational Psychology, 78, 201-209.

Schunk, D. H., & Zimmerman, B. J. (1994). Self-regulation of learning and performance: Issues and educational applications. Hillsdale, NJ: Lawrence Erlbaum.

Schunk, D. H. (1996). Goal and self-evaluative influences during children's cognitive skill learning. American Educational Research Journal, 33, 359-382.

Seger, C. A. (1994). Implicit learning. Psychological Bulletin, 115, 163-196.

Shuell, T. J. (1996). Teaching and learning in a classroom context. In D. C. Berliner & R. C. Calfee (Eds.), The handbook of educational psychology (pp. 726-764). New York, NY: Macmillan.

Sternberg, R. J. (1988). The Triarchic Mind : A New Theory of Human Intelligence. New York, NY: Viking.

Sternberg, R. J., & Wagner, R. K. (Eds.). (1994). Mind in context: Interactionist perspectives on human intelligence. Cambridge, UK: Cambridge University Press.

Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work related performance: A meta-analysis. Psychological Bulletin, 124, 240-261.

Stipek, D. J. (1993). Motivation to Learn. Boston: Allyn and Bacon.

Swanson, H. L. (1990). Influence of metacognitive knowledge and aptitude on problem solving. Journal of Educational Psychology, 82, 306-314.

Wagner, R. K., & Sternberg, R. J. (1985). Practical intelligence in real-world pursuits: The role of tacit knowledge. Journal of Personality and Social Psychology, 52, 1236-1247.

Weiner, B. (1986). An attributional theory of motivation and emotion. New York: Springer Verlag.

Zimmerman, B. J., & Bandura, A. (1994). Impact of self-regulatory influences on writing course attainment. American Educational Research Journal, 31, 845-862.

Zimmerman, B. J., Greenberg, D., & Weinstein, C. E. (1994). Self-regulated academic study time: A strategy approach. In D. Schunk & B. J. Zimmerman (Eds.), Self-regulation of learning and performance: Issues and
educational applications (pp. 181-201). Hillsdale, NJ: Erlbaum.