In H. .J. . Hartman (Ed.) 2001 Metacognition in Learning and Instruction: Theory, Research, and Practice. Chapter 9 Dordrecht, The Netherlands: Kluwer Academic Publishers. 173-201.



Hope J. Hartman

City College of City University of New York

Graduate School and University Center

City University of New York






Recent research on science teaching and learning emphasizes the importance of active, meaningful learning, with metacognitive processing by both teachers and learners. This chapter describes research on students= reading comprehension metacognition in biology and instructional methods I have used to enhance students= metacognition about their science learning and to enhance science professors= metacognition about their teaching.




Students often consider science to be one of the most difficult subjects they take, whether in high school or college. Because of its perceived difficulty, many students even develop science phobias, much like they tend to do with mathematics. Science teachers often reflect on the content they are going to teach, but to what extent do science teachers think reflectively about the pedagogy they use to teach specific scientific concepts and skills? To teach science successfully, teachers can use their metacognitive or high-level thinking about what, why and how they teach in order to manage and regulate their teaching so that it meets the needs of their students. In addition, to help students learn science effectively, teachers can develop their students= use of metacognition so they gain awareness and control over themselves as learners. This chapter reviews some of the literature on science teaching that relates to metacognition in teaching and learning science. Then the chapter describes some of my experiences with metacognition in science teaching and learning.



In a review of the literature on the implications of cognitive science for teaching physics, Redish (1994) identified four broad principles, each with corollaries, which are useful for physics teachers to help them think about their teaching. First is The Construction Principle, which states that people organize their knowledge and experience into mental models and that people must build their own mental models. Second is The Assimilation Principle, which says that mental models control how we incorporate experiences and new information into our minds. Related prior knowledge and experience form mental models into which new knowledge and experience are incorporated. Third is the Accommodation Principle, which emphasizes that sometimes existing mental models must be changed for learning to occur. Fourth is the Individuality Principle, which highlights individual differences in students= mental models as a result of their personal constructions. Students have different mental models for learning and different mental models for physical phenomena. These four principles provide a metacognitive framework for physics and other science teachers to help them plan, monitor and evaluate their instruction, classroom activities, and learning assessments so they can maximize students= understanding of science. For example, Redish suggests that looking at the curriculum from the mental models perspective helps teachers establish the goals of identifying the mental models they want students to develop, stimulates teachers to consider the character and implications of students= pre-existing mental models, and helps teachers realize the benefit of using touchstone problems to analyze and identify critical aspects of the curriculum. According to Redish, one of the implications of the individuality principle is that teachers need to think about how students may arrive at the same answer but for very different reasons, to determine how students reason, teachers should listen to them thinking aloud without guiding them.

Walberg (1991) suggests that in science it is especially useful for students to struggle with interesting, meaningful problems that can stimulate discussion about competing approaches. He recommends using what he calls Acomprehension teaching,@ more commonly called scaffolding, which involves providing students with temporary support until they can perform tasks on their own. Based on Vygotsky's (1978) concept of the zone of proximal development, scaffolding is recommended for teachers to build from what students can do only with temporary guidance from a more competent person, gradually reducing and eventually removing this support as students become independent thinkers and learners who can perform the task or use the skill on their own. Scaffolding has been found to be an excellent method of developing students= higher level thinking skills (Rosenshine & Meister, 1992). However, metacognition is needed for teachers to use scaffolding effectively, as they consider issues such as what types of support to provide, what order to sequence them in, and how to decide when it is time to reduce or withdraw support from students.

Activity-based teaching has been found to be especially beneficial for students with lower achievement records, ability and socioeconomic status. Walberg says the following effective teaching methods are particularly-cost effective: cooperative learning, mastery learning, direct instruction and comprehension teaching (scaffolding). Recent research indicates that extensive knowledge is required for excellent teaching (Walberg, 1991). Extensive knowledge needed for effective teaching includes both subject-area knowledge and pedagogical information, such as strategic metacognitive knowledge about teaching strategies. For example, strategic metacognitive knowledge about the teaching strategy of cooperative learning includes knowing what cooperative learning is, knowing when and why it is useful, and knowing how procedures for implementing cooperative learning (See Chapter 8 for details.) Having this kind of information about teaching strategies helps teachers decide among alternative approaches to use in various situations.

How do science teachers decide which teaching methods to use? For over 20 years now the Learning Cycle (Karplus, 1974) has been used to structure science instruction in order to help students move from concrete experience to formal, abstract thinking about content.This constructivist approach is based on Piaget's theory of intellectual development. Development is structured by three teaching phases: Exploration, Concept Invention/Introduction and Application. Through this sequence students= thinking is expected to progress from concrete thinking about science concepts to being able to deal with these concepts on a formal, abstract level. Application often involves tasks or problems that relate to students' everyday lives (Barman, Benz, Haywood & Houk 1992). However, the learning cycle might not always be the best approach for teachers to use. A study comparing the learning cycle model to modeling in urban, middle school science students found that although both the modeling and learning cycle groups outperformed the control students in their use of integrated science process skills, students who were taught by modeling developed better integrated science process skills than students who were taught by the learning cycle approach (Rubin and Norman,1989). Consequently, rather than just automatically using a commonly accepted approach, such as the learning cycle, teachers should use their metacognition to carefully reflect on the implications of what research has shown about the advantages and disadvantages of a variety instructional methods for their specific students and subject matter. Perhaps a combination of the learning cycle and modeling approaches might be tested to see if it leads to even higher levels of integrated science process skills development.

Increasingly new technologies are supplementing and enhancing the learning process. These technologies can support new views of science teaching. A high school biology course characterized by "model-based reasoning" emphasized both the development of conceptual and strategic knowledge of classical genetics, as well as the development of insights regarding science as an intellectual activity. This nine-week course for seniors involved model revising problem solving in contrast with more common model using problem solving. In the model revising approach, students work in research groups sharing their observations of phenomena, building models, and defending models to groups of students who critique each other=s models. The critiques lead to model revising. The emergence of competing models increases student awareness of the need for models to explain existing data and predict additional observations. Students also get increased awareness that more than one model may be consistent with the data, and may predict and explain. The computer played an important role in the development of this course. Use of computers was guided by the ..."view that science education should allow students to be engaged in many of the activities of science"... "and science teaching as problem posing, problem probing, and persuasion of peers." (p. 334). Software supplemented work with real organisms, and enabled students to learn genetics by engaging in activities like those of geneticists (Stewart, Hafner, Johnson & Finkel, 1992). Thinking metacognitively about teaching includes exploring such alternative approaches to instruction, monitoring their implementation, evaluating their effectiveness, and using feedback to plan future lessons.

Tobias (1986) characterized introductory college science courses by negative features such as failure to motivate student interest, passive learning, emphasis on competitive rather than cooperative learning, and reliance on algorithms rather than understanding. These features sometimes steer students away from careers in the sciences. Recent research suggests that the mismatch between teaching practices and students' learning styles may account for many of these problems. Felder's (1993) model of learning styles is especially appealing because it conceptualizes the dichotomous dimensions (sensing/intuiting, visual/verbal, inductive/deductive, active/reflective, and global/sequential) as a continuum rather than either/or categories. Felder cites research to guide instruction for each of these styles. Felder also recommends systematic use of a few additional teaching methods which overlap learning styles and help meet the needs of all students. These include: give students experience with problems before giving them tools for solving them, balance concrete with conceptual information, liberally use graphic representations, physical analogies and demonstrations, and show students how concepts are connected within and between subjects and to everyday life experience. Teaching science metacognitively can help teachers improve the alignment between their teaching practices and students= learning styles.

Science educators (e. g., Baird & White, 1984) suggest that self-questioning and think- aloud processes are effective strategies to promote scientific thinking. Baird and White conducted a study designed to improve metacognition in ninth grade students learning science and eleventh graders learning biology. They identified seven learner objectives:1) increased knowledge of metacognition, 2) enhanced awareness of their learning styles, 3) greater awareness of tasks= purposes and natures, 4) more control over learning through better decision-making, 5) more positive attitudes toward learning, 6) higher standards for understanding and performance set by the students themselves, with more precise self-evaluation of their achievements, and 7) greater effectiveness as independent learners, planning thoughtfully, diagnosing learning difficulties and overcoming them, and using time more productively. Instructional materials included a question-asking checklist; an evaluation of learning behaviors, an outcomes notebook, and a techniques workbook, where students tried out concept mapping. This extensive study went through four phases and involved 15 methods of collecting data, including video and audiotapes, classroom observations, questionnaires and tests. The results showed increased student control over learning and understanding of content.

Stress on Analytical Reasoning (SOAR) is a program at Xavier University in New Orleans with a record of success teaching science to minority students interested in health sciences, physics, engineering or mathematics (Carmichael, Ryan, Jones, Hunter & Vincent, 1981). One of the teaching strategies it has found especially useful is Pair Problem Solving. Whimbey and Lochhead (1982) describe this technique as a thinker and listener pair working on problems and rotating roles.. Students take turns serving as thinkers (problem solvers), who externalize their thought processes by thinking aloud while analytical listeners track and guide the problem solving process as needed. This method makes problems more engaging and promotes self-monitoring and self-evaluating, giving students feedback on what is understood and what is still unclear. It encourages skills of reflecting on beginning and later thoughts. It also teaches communication skills, fosters cooperation, and encourages the formation of study and support groups. Finally, pair problem solving exposes teachers and students to various solution approaches (Heiman, Narode, Slomianko, & Lochhead, J. 1987). By listening to one=s own thoughts, the student gains awareness and control over problem solving. Externalizing thoughts enables them to be seen from a fresh perspective. Together, the students can discover errors, misconceptions, organizational problems, and other impediments to academic performance. The teacher needs to observe each pair, monitor progress, and provide feedback on the process. This approach has been demonstrated to be an effective approach for helping students learn science and math (Whimbey & Lochhead, 1982.) The findings about pair-problem solving provide strategic metacognitive knowledge science teachers can use to decide when and how to use the pair problem-solving method in their classes.

To what extent do teachers think about their assessment techniques and how well they measure important instructional goals? Research on assessing hands-on science suggests that there should be symmetry between curriculum and assessment, that assessment should be continuous, and that performance measures are needed to supplement traditional multiple choice-type assessments in order to get a comprehensive picture of student achievement. Performance measures should emphasize science process skills, such as observing and inferring, not just getting the right answer. Four performance assessments that can be used to assess science achievement are: 1.lab notebooks recording students= procedures and conclusions; 2. computer simulations of hands-on investigations; 3. short answer paper-and-pencil problems in planning, analyzing and/or interpreting experiments; and 4) multiple choice items developed from observations of students conducting hands-on investigations. Effective science performance assessment requires multiple iterations to revise assessments based on students= experiences and feedback. To shortcut this process often leads to poor assessment and low quality classroom instruction (Shavelson & Baxter, 1992). Feedback is important for students in several ways: it helps them assess their mastery of course material, helps them assess their use of thinking and learning strategies, and helps them connect their efforts and strategies to their academic outcomes. The primary benefit of feedback is the identification of errors of knowledge and understanding and assistance with correcting those errors. Feedback generally improves subsequent performance on similar items. Research suggests that feedback can guide students in their use of learning strategies, and that adults who try different strategies and are tested on their learning can generally identify effective strategies (Crooks, 1988). Mestre (1994) found that problem posing, when followed by an interview, is a powerful assessment strategy for evaluating the development and understanding of physics concepts in high-performing university physics students. Mestre found that these Agood novices@ were able to pose appropriate, solvable problems when responding to a problem situation or concept scenario, but they also had major flaws in their conceptual understanding. The flaws suggested that students were deficient in how their knowledge was organized in memory and how it was connected with procedures and problems. Most teachers could benefit from having such information about assessment and feedback and using it metacognitively to improve their teaching and evaluation practices as well as to improve students= performance.


Mental representations of information to be learned or used in problem solving are important determinants of whether and how learning will occur. Representations can be internal, like mental images, or external, like charts or tables, as metacognitive aids. McIntosh (1986) found that teaching ninth grade physical science students to generate visual images helped them remember rules in science (e.g., Boyle=s Law). Sternberg (1985) characterized intelligent performance by the use of multiple representations. Similarly, in his review of the learning strategies literature, Dansereau (1978) reported that multiple encodings have a more facilitative effect on retrieval than do single encodings. For example, it is better to use both visual imagery and mnemonics to remember than to use either encoding strategy alone.

In a review of research on using concrete, visual models to facilitate understanding of scientific information, Mayer, (1989) found that such models consistently helped lower aptitude learners think systematically about scientific material. According to Mayer, concrete models, which consist of words and/or diagrams, help students construct representations of the major objects, actions and their causal relations in the scientific content being studied. He identified seven characteristics of effective models in this review. The good models he found were: complete, concise, coherent, concrete, conceptual, correct and "considerate" (i.e.using vocabulary and organization appropriate for the learner). In short, models are "good" with respect to certain learners and certain instructional goals. He also identified some guidelines for application of concrete models, including when and where they should be used, why to use them.


Concept maps and vee diagrams

Concept maps and Vee diagrams help people learn how to learn. Procedures for creating them are in appendices of Novak=s (1998) new book, Learning, Creating and Using Knowledge, which describes his human constructivist theory . Concept maps, developed by Novak in 1972, are graphic representations of knowledge with the most general concept at the top, hierarchically leading to more specific concepts. Concepts are in boxes or circles, with labeled connecting lines identifying relationships. The labels are words that link one concept to another, and the label is placed in the middle of the linking line. For example, .as shown in Figure 1, the concept Abody fluid compartments@ is in a box, with a line drawn from it to linking words, such as Aare@ which has a line drawn from it to the more specific concepts Aintracellular fluid@ and extracellular fluid @ which are also in a box. The linking line coming from intracellular fluid, with the word Ahas@, leads to a more specific box containing the concepts Alow concentration of sodium ions@ and Ahigh concentration of potassium ions@ and so forth. According to Novak, concept maps help students become empowered, and reduce the need for rote learning, and they help teachers negotiate meaning with students and design better instruction. Concept mapping can be used successfully individually and in teams; with concepts, events, and social relationships; with young children and adults; in schools and corporations with researchers, teachers/managers and students/workers; and in everyday life. They are used in teacher education, curriculum development, and assessment (Hartman, 1999). One example is from a handbook for college chemistry workshop leaders illustrating the integrated components of their Workshop Model (Roth, Strozak, Cracolice & Gosser, 1997). The concept map consists of four circled concepts arranged in a diamond. The concepts are students (top circle), learning specialists (bottom circle), workshop leaders (left circle), and faculty (right circle). The circles are connected by four diagonal lines labeled Workshop, connecting circles of workshop leaders and students; Lectures & Laboratory, connecting circles of students and faculty; Program Direction, connecting circles of faculty and learning specialists; and Leader Training, connecting circles of Learning Specialists and Workshop Leaders. The leadership training includes intensive experience learning how to construct and use concept maps, so the workshop leaders can teach the chemistry students how to construct and use them..

(Insert Figure 1, concept map Abody fluid compartments@)

Extensive work has been done using concept maps in schools and corporations. Many benefits of concept maps cited include promoting meaningful learning (especially in science), understanding superordinate and subordinate relationships, improving peer relationships and trust, resolving conflicts, and improving understanding of one=s role in and contributions to team projects. They also help students and teachers differentiate misconceptions from valid conceptions, decrease anxiety, improve self-confidence, and more. Junior high school science students taught to use concept maps and Vee diagrams outperformed students who were not taught these strategies on tests of novel problem solving (Novak, Gowin, & Johansen 1983). Research by Okebukola on using concept maps with high school biology students in Nigeria showed that students using concept maps had significantly better content mastery, better attitudes toward biology and less anxiety than students who did not use concept maps (Novak, 1990). Chemistry students, ages 16-18 in a technical school, were taught concept mapping to aid their visualization of knowledge structures and to document and explore changes in their knowledge structures as a result of learning (Regis & Albertazzi, 1996). After four years of experience Regis and Albertazzi found, A...we have grown more and more impressed by the potential of this metacognitive tool to help chemistry teachers and learners to improve teaching and learning@ (p. 1088). They found that concept maps help teachers know what students know and how they relate concepts in their knowledge base, as well as highlight what misconceptions students have and let teachers see how students reorganize their cognitive structures after a specific learning activity. They found that concept maps benefit learners by making learning of new subject matter meaningful. Support was also found for using a concept map to design an artificial intelligence training program for diagnosing coronary problems (Ford, Canas, Jones , Stahl, Novak, & Adams-Weber 1991).

Vee diagrams (The Knowledge Vee) were developed by Novak=s retired colleague Gowin in 1977 to help students understand research. They are Vee-shaped graphic organizers that help learners systematically observe and measure all the relevant variables by focusing on the specific principles and concepts that are involved in the event and the focus question. They consist of four basic sections: top center - focus question, bottom center - event, left side - thinking, conceptual/theoretical, and right side - doing, methodological . The left side consists of the learner=s world view, philosophy/epistemology, theory, principles, constructs and concepts. The right side consists of value and knowledge claims, transformations and records. All Vee components interact to create new knowledge.

Thinking metacognitively about conducting science lessons often includes selecting which representations or models to present to a class (e.g, flow charts, diagrams, concept maps), determining when to present them (e.g. order in the instructional sequence), and deciding how to present them (e.g. blackboard, transparency, CD-ROM). It also includes the teacher=s self-assessment of the effectiveness of the representations selected, the timing of their implementation, the method of presentation, and a lesson improvement plan for more effective use of representations.


Students often complain about reading their science texts. Even otherwise competent readers aren=t aware of the top-level structures underlying scientific texts (Cook and Mayer,1988). Top-down structures are important because they trigger higher-order ideas that activate schemata which allow details to be inferred and attention to be allocated effectively (Pressley and McCormick, 1995). Research by Cook and Mayer suggests that college students who don=t understand the structure of scientific texts have problems representing the material, thereby impeding comprehension and retention. One study found that students had difficulty sorting text into the text-structure categories of classification, comparison/contrast, enumeration, sequence and generalization. In another study, after receiving eight hours of training in analyzing, recognizing, and organizing relevant information in scientific texts, experimental junior college chemistry students outperformed controls on measures of comprehension. Text structure instruction included modeling reading strategies, explicitly explaining how to identify a sequence (for example, how to put a sequence into one=s own words), how to identify the key words signaling a sequence, and how to identify supporting evidence. Thus developing students= metacognition about how to read scientific texts can improve their comprehension by helping them focus on relevant information and use it to create internal connections and representations.

In a related study, Speigel and Barufaldi (1994) focused on four of the same common science text structures as Cook and Mayer, classification, enumeration, sequence and generalization, and one different one, cause and effect. Community college students in anatomy and physiology were taught to recognize these text structures and to construct graphic organizers of them after reading (postorganizers). Students who constructed postorganizers demonstrated superior memory on immediate and delayed posttests when compared to students who used rereading, highlighting or underlining. Spiegel (1996) emphasizes the importance of providing students with strategic metacognitive information (what, when, why and how) on the use of learning strategies such as graphic organizers. To what extent do science teachers regularly provide students with the metacognitive knowledge needed to effectively and efficiently learn to use graphic organizers and other text digestion strategies?

When reading scientific texts students often try to rotely learn big words, facts and details instead of trying to understand ideas. They learn so that they can "report back" information but not apply it (Roth, 1991). Roth reported that some students, "conceptual change readers", tried to understand and accommodate their beliefs to the information in the text. They activated their prior knowledge and recognized when it was somewhat inconsistent with the meaning described in the text. Conceptual change readers thought about the meaning and worked to resolve the discrepancy to refine their own thinking. This effort to clarify the misconception was described as a "conceptual change strategy." These students exhibited the self- awareness and self-regulation that are the essence of metacognition in learning. How did the conceptual change readers learn to use these strategies?

Scientific textbooks sometimes contain misconceptions and alternative conceptions about science, so reading them can interfere with learning unless the teacher filters conceptual problems before students read them and treat them as valid knowledge (Abimbola and Baba 1996). Abimbola and Baba developed a procedure for teachers to use to analyze textbooks and identify misconceptions. In their analysis of one textbook, STAN Biology, they found 117 misconceptions and 37 alternative conceptions and were distributed in 18 of the 22 chapters. One type of misconception is using wrong or out-of-date words to represent concepts: e.g. Asemi-permeable membrane@ has been replaced with Aselectively permeable@ or Adifferentially permeable,@ so it is not misunderstood as partially permeable or partially impermeable. Another type of misconception is statements that are wrong:e.g., AOxygen is produced as a waste product@, is erroneous because in the context of nutrition and photosynthesis, where it appeared in the text, oxygen is really a useful end product of photosynthesis because it oxidizes food to release energy. An example of an alternative conception they found is defining dentition by teeth, without including that dentition also includes the arrangement of teeth. Abibola and Baba recommend that teachers consider the number of misconceptions and alternative conceptions when selecting among science textbooks and select the one with the fewest. This research suggests that effective science teaching metacognition includes awareness of how commonplace misconceptions are in standard science texts, and control over textbook selection to avoid those with false statements.



Students are far from "tabula rasa" who simply acquire information teachers and books provide. They usually come to courses with at least some prior knowledge, beliefs, values, attitudes and experiences that influence what and how they think and learn. Background information provides a foundation to build on. Some of what they bring is an emerging foundation, parts of which can be built on, parts of which must be revised, and parts of which must be discarded. Some of what they bring creates obstacles to, inhibits or prevents learning. Finally, some of their misconceptions don't really matter. What are misconceptions? Misconceptions are faulty ideas that are based on false or incomplete information, limited experience, incorrect generalizations or misinterpretations and are consistent with the student's basic understanding. Some misconceptions result from cultural myths or scientifically out-of-date information. Others may arise from vague, ambiguous, or discrepant information. Some researchers view and refer to misconceptions as "alternative frameworks" or "preconceptions" which emphasize the emergent nature of structures of knowledge. Anderson, Sheldon & Dubay=s (1990) study of college biology students looked at concepts of respiration and photosynthesis. Sample misconceptions include the simplistic definition that respiration is exhaling CO2 , and not understanding that plants manufacture their own food but thinking that plants get their food from nutrients in the soil. Textbooks are not the only source of students= misconceptions. Teaching science metacognitively includes teacher awareness of the sources and characteristics of students= misconceptions, selection of strategies to overcome students= misconceptions, and monitoring/evaluating the extent to which important misconceptions have been replaced with accurate conceptions.

Research shows that misconceptions are deeply entrenched and enduring, even after students learn new information that is inconsistent with their prior knowledge. Learners must have extensive and deep, meaningful learning for the new, correct knowledge to come to mind and be applied instead of the old misconceptions (Pressley & McCormick , 1995). According to Duit (1991), prior knowledge affects students= observations, guiding them to information that is consistent with their own perspectives. Students selectively attend to information, seeking to confirm what they already "know." Sometimes students' prior knowledge is so strong that they won't even believe what they see. A videotape AA Perfect Universe@ (Schneps, 1994) shows that even students and professors at Harvard suffer from deeply entrenched scientific misconceptions.

Research suggests that usually multiple knowledge levels (or domains) of misunderstandings are involved in any given misconception. Teachers should pay attention to each level and the relationships between them. Each domain or level contains a variety of kinds of knowledge. According to Perkins and Simmons' (1988) integrative model of misconceptions, deep understanding involves four interlocked levels of knowledge and teachers need to address all four: 1. Content: e.g. recalling facts, using vocabulary; 2. Problem Solving: e.g. strategies, self-regulation; 3. Epistemic: e.g. explaining rationales, providing evidence; and 4. Inquiry: critical thinking - extending and challenging domain-specific knowledge. To apply teaching metacognition to this problem, science teachers could develop plans to identify the types of misconceptions their students have and select or develop procedures to overcome them.

Many researchers believe that students overcome misconceptions by recognizing and replacing them. Nussbaum & Novick (1982) proposed that awareness of beliefs is necessary before students can overcome misconceptions. Awareness creates cognitive conflict which motivates conflict resolution to accommodate current beliefs or cognitive structures. Accommodation may lead to modifying existing structures and/or creating new ones. Minstrell (1989) claims that earlier ideas are seldom pulled out and replaced. He believes it is more effective for teachers to help students differentiate between their present ideas and those of scientists and to help them integrate their ideas into conceptual beliefs more like those of scientists.

Research on teaching for conceptual change suggests that students can be taught active processing strategies (e.g. predict, explain) to help them notice and correct their misconceptions, thereby deepening scientific understanding. Students can learn to distinguish similar concepts from each other (e.g., force, impulse, work) and from properties of systems or objects (McDermott, 1984). Direct hands-on experiences can be used to help students develop a model of a concept based on their own observations enabling them to make more accurate predictions and explanations (McDermott, 1991). Posner, Strike, Hewson and Gertzo (1982) highlight conditions for conceptual change: dissatisfaction with a current concept, perceived plausibility of a new concept, and perceived usefulness of a new concept. They also emphasize some aspects of the learner's "conceptual ecology," e.g. epistemological commitments about the nature of evidence, the importance of parsimony, and metaphysical beliefs, such as faith in nature's orderliness. Their conceptual change model emphasizes confronting existing concepts and facts, pointing out contradictions, asking for consistency, and making theory intelligible, plausible and fruitful.

Several aspects of this literature have been applied to my work with science teachers and learners. The next section describes some of this work.



Fortune smiled upon me in 1987 when Professor Joseph Griswold, of CCNY=s biology department wanted to work with me on Introductory Biology, a high risk course primarily taken by Nursing, physician assistant and physical education majors. Through funding from the Aaron Diamond Foundation we restructured the course and provided out-of-class academic support using the Supplemental Instruction model (Blanc, De Buhr, & Martin, 1983). Additional support from CUNY Exemplary Programs allowed us to integrate multimedia technology - laser videodiscs and video microscopes- into classroom instruction to motivate both faculty and students . We began working on Anatomy and Physiology in 1989, with a Title 111 grant. These experiences, like all of my work were guided by the BACEIS model of improving thinking (Hartman & Sternberg, 1993) which emphasizes the internal world of the student (cognition and affect) and the environmental context (academic and nonacademic) as interacting systems affecting students' academic performance (See Chapter 12 for more details).

Eylon and Linn (1988) report that cognitively, students respond better to systematic in-depth treatment of a few topics than they do to conventional "in-breath" treatment of many topics. Increasingly it is recommended that science teachers streamline the curriculum and focus more on a limited set of ideas. Students' misconceptions and lack of understanding of science basics reflect limitations of mental processing and memory. Thinking metacognitively about teaching led first to Professor Griswold streamlining the curriculum of the introductory biology course after examining the knowledge required for the advanced courses in anatomy and physiology these same students would take. This comparative analysis enabled him to identify fundamental concepts that students needed to learn well in the introductory course so they could be built upon in the advanced courses. Selecting critical content enabled the professor to identify less fundamental concepts that could be cut from the course so that the critical concepts could be treated in greater depth. Teaching metacognitively included creating advance organizer outlines to be put on the blackboard at the beginning of each lecture and structuring more active student involvement in the lecture through different types and levels of questioning. In addition, Professor Griswold started incorporating vocabulary building skills (e.g. recognizing prefixes, suffixes and root words) into the beginning of the course. The first lecture exam was changed to include assessment of critical vocabulary and word attack skills.

Professor Griswold used several metacognitive teaching techniques, including graphic organizers such as flow charts and concept maps, to help his students organize, understand and remember what they had learned. Lectures and laboratories were designed to begin with the concrete and lead to the practice of formal operational functions such as problem solving, using symbols and verbal reasoning based on a Learning Cycle type approach. For example, initially small groups of students brainstorm together about how smoking could interfere with gas exchange in the lungs. Sharing ideas activates prior knowledge, reveals misconceptions, and identifies information gaps that need to be filled. This exploratory phase is followed by a dissection of the respiratory system and a study of function using demonstrations and videomicroscopy. Next, students analyze the issue of smoking using their new knowledge. Finally they apply what they have learned to explaining new situations such as a change in function with physical exertion or pathology (Griswold & Hartman, 1991). During application, students return to the original question on the impact of smoking, with a new understanding of the mechanics of interference.


Developing Scientific Thinking Skills

One of our goals for the Supplemental Instruction component of our program was to help students develop the thinking and reasoning skills they needed to master science. This goal is based on the assumption that students will master science content more effectively if they use well- developed intellectual processes when thinking and learning about science. Research has indicated that if we want students to be able to apply their knowledge and skills to academic and professional tasks on a long term basis, then instruction should address both cognitive and metacognitive aspects of students' performance. We targeted cognitive skills from Sternberg's (1985) triarchic theory of intelligence (e.g., comparing, applying, justifying) and provided them with strategic metacognitive knowledge about each skill (Hartman & Griswold, 1991) to facilitate their effective use, long-term retention and transfer. Metacognitive knowledge includes what the skills are (declarative), when/why to use them (contextual or conditional) and how to use them (procedural). There are several teaching strategies that can be used to develop intellectual skills, such as activation of prior knowledge, self- questioning, imagery, and graphic organizers such as concept maps, flow charts and, matrices (Narode, Heiman, Lochhead and Slomianko, 1987; Hartman, 1993). We used scaffolding strategies to help students develop their cognitive skills so that they could ultimately self-regulate their use of these skills; this meant providing students with models, guided practice with feedback in groups, guided individual practice with feedback, and unguided individual practice with feedback. Many activities during Supplemental Instruction were designed to develop students= metacognitive strategies for mastering biology. These included: comprehension monitoring, graphic organizers, self- questioning, imagery, thinking aloud, time management, testwiseness, and error analysis. Table 1 is an example of the strategic metacognitive knowledge we gave students about the cognitive skill of justifying.


Table 1

Strategic Knowledge about Justifying

KNOWING WHAT: Justifying is explaining reasoning, providing evidence underlying conclusions/reasoning; comparing obtained outcome with achieved outcome and evaluating degree of difference. For example, if a professor asks A What makes you think that the small intestine is involved in protein digestion?@, the student must provide supporting factual knowledge, such as AYou can see from the professor's concept map that the small intestine has intestinal glands which produce peptidase. Peptidase acts on polypeptides, which are broken into amino acids. The small intestine receives acid chyme, which includes proteins, which are broken into polypetides, which are broken into amino acids. Amino acids are the final products of protein digestion.A

KNOWING WHEN/WHY: Use justifying to establish a sound basis of support for beliefs, decisions, and/or actions.

KNOWING HOW: To justify one must understand the concept of evidence or support and its value in making decisions about what one does, knows, or believes. Procedures include finding evidence, weighing and comparing evidence to a standard or set of criteria, evaluating its strengths, weaknesses and degrees of difference from the criteria, and looking at all possibilities. When there is strong support for one answer, interpretation, approach etc., and weak support for all the others, justifying includes judging that one to be the best explanation, approach, answer-choice, etc. under the circumstances.

Another way we tried to improve students= scientific thinking skills was to teach them the AI DREAM of A@(Hartman, 1996a) method to help them think systematically about how they plan, monitor and evaluate their approaches to solving problems, thereby using their metacognition for self-management or self-regulation. I DREAM of A is an approach to developing metacognitive aspects of scientific and mathematical problem solving skills by using thinking aloud and questioning strategies, and is derived from Bransford and Stein's (1984) IDEAL problem solver. Each capitalized letter stands for a component of the problem solving process, so the acronym represents a systematic guide to problem solving. These components involve executive management skills for planning, monitoring and evaluating the problem solving process. The first four letters are all planning steps (identify and define, diagram, recall, explore alternatives) which may be performed in different sequences. The next two letters (AM) focus on applying and monitoring the plan, and the final A stands for assessment, where students evaluate their solutions to the problem, both before and after getting someone=s feedback. The I DREAM of A approach is not a rigid, cookbook, rote formula; rather it is a method of remembering to plan, monitor and evaluate one=s problem solving. For example, problem solving often begins with "D", diagraming the problem, which sets the stage for "I,@ identifying the problem. The method addresses both cognitive and affective aspects of students= performance and it must be personally adapted by the problem solver to fit the specific needs of each problem-solving situation.

The teacher can serve as an expert model, demonstrating how to use I DREAM of A by playing the role of questioner, who guides the problem-solving process by questioning the other student and by having the problem solver think aloud periodically while problem solving . The problem solver answers questions thinking aloud and self-questions while solving problems. Then two work together, as in the pair-problem solving method described previously. Questions are asked for each of seven components of " I DREAM of A.@ Although most questions focus on knowledge and strategies needed to solve a problem, the questioner occasionally asks about the problem solver's feelings to establish and maintain a positive attitude. The questioner decides what questions to ask, when to ask the problem solver to think aloud, and when to ask about the problem solver's attitudes. Finally, the student can use the model alone to stimulate self-management of problem solving, asking self-questions and thinking aloud, always adapting it to the specific context.

Restructuring introductory biology involved multidimensional aspects of teaching and learning metacognition. Restructuring involved modifying both the curriculum and instruction, implementing new teaching strategies, and developing students= scientific thinking and learning skills and attitudes. The results suggest that the restructuring efforts were beneficial: student achievement increased and course failure and attrition decreased over seven semesters. In Spring 1986, 34.7% of the class earned grades of C or above. By Spring 1989, 64.6% of the class earned grades of C or above. In Spring 1986, 40.3% of the class failed or withdrew from the course; in Spring 1989, only 17% failed or withdrew from the course (Hartman & Griswold, 1994).


Anatomy and Physiology

In 1994 NSF grants with Daniel Lemons and Joseph Griswold focused on revising curriculum, instruction and assessment in two advanced anatomy and physiology courses taken by the same set of students whose introductory biology we had worked on previously. Students taking these two courses were primarily ethnic minorities whose native language was not English. Most of these students (75%) were nursing majors who were required to take these courses . For many years students had a history of a high failure rate (defined as D or below) in both courses. This project was based on the assumption that there is a substantial discrepancy between how traditional Anatomy and Physiology courses are taught and how non-traditional students learn. Traditionally, learning in anatomy and physiology courses has emphasized extensive memorization of facts about human systems rather than understanding how they work. Thinking metacognitively about what, why and how to teach anatomy and physiology to nontraditional students led the biologists to the following curriculum innovations: 1) in the context of a broad, organizing theme, sequence instruction so that function is presented first, followed by the anatomical and physiological details which explain that function; 2) start with explorations using hands-on physical models, then progress to higher-level activities, such as applications of models to new situations; 3) to help students learn to solve problems, place the highest priority on critical thinking; 4) use computer-based activities (e.g., simulations , CD-ROM images) to support learning at all phases; 5) provide students with structured, out-of-class academic support; and 6) use assessment not just as an endpoint, but as a integral component of the learning process (Griswold, Lemons & Hartman, 1995).

Using their teaching metacognition to revise the curriculum for the two anatomy and physiology course led Lemons and Griswold to develop a curriculum planning model to serve as a framework for managing the content, pedagogy and assessment of their courses. The curriculum model, a modified learning cycle, is summarized in Figure 2.

(Insert fig. 2 Modified Learning Cycle)


Another example of their teaching metacognition was their awareness that their students tended to think concretely instead of abstractly (confirmed by students= pretest scores on the Group Abstract Logical Thinking (Roadrangka, Yoany, & Padilla,(1983). Therefore, a learning cycle model framework is appropriate because it is specifically designed to help students progress from concrete to abstract thinking about content. However, again using their teaching metacognition, Lemons and Griswold determined that the original learning cycle model needed to be adapted for the needs of their students, goals and activities. Hence, they modified the original learning cycle model, systematically integrating into it their computer-based learning activities and simulations and their formative and summative evaluations. Thus their metacognitive awareness of the learners= and learning cycle=s characteristics led to their metacognitive control of modifying the model for the specific context.

Another example of their teaching metacognition is their reflection on how to design a curriculum to achieve the targeted objectives. This goal led Lemons and Griswold to design specialized materials and activities to support their curriculum. These include a CD-ROM of computer simulations and instructional activities, physical models and manipulatives, and A Laboratory Guide for Human Anatomy and Physiology (Lemons, 1994), a student workbook of laboratory activities. Each curriculum unit has explicit objectives identified and disseminated to students to establish a shared framework from which to view the material and activities. Explorations introducing students to the concepts to be studied in the unit involved students activating their prior knowledge about a topic and relating their prior knowledge (and/or experience) to the material to be learned. In each unit activities were designed specifically to stimulate students= critical thinking, and therefore required students to think metacognitively about their interpretations, explanations and solutions.

Because learning biology requires extensive reading of technical text, students with limited-English proficiency may encounter more difficulty processing the text than their professors and the text authors realize. Research on the influences on reading achievement of language minority children found oral language proficiency to be the best predictor across eight ethnolinguistic groups (DeAvila & Duncan, 1985). Again illustrating Professor Lemon=s metacognition, he realized that his students were reading the textbook incorrectly, if at all, so he decided to model for them how to read the text. Serving as an expert he demonstrated how to use the figures to improve their understanding of the text and their ability to identify important information in the text. Thus he used his teaching metacognition to develop their science reading metacognition. The next section describes two studies on these students= thinking about biology.


Metacognition, Reading Comprehension and Misconceptions in Biology

Two exploratory studies were designed to help us learn about factors affecting students= performance in their biology courses on anatomy and physiology. The first study focused on students= reading of their textbook; the second focused on their misconceptions about biology.

Due to the high failure rate of students in the two course anatomy and physiology sequence, we wanted to identify students likely to have difficulties learning the material so we could both help them and identify variables that predict success. This research examined students' biology reading comprehension and metacognition to identify effective correlates of success in college anatomy and physiology, The Biology Reading Test was developed using students' actual text as the source of material in order to assess their comprehension and reading-test metacognition (Hartman , 1996b).1

The test consists of four reading selections from the first two chapters of the text, which are reviews of information students are expected to bring to the course. Each passage has four reading comprehension items. Each of the 16 comprehension items is followed by a metacognitive question requiring students to evaluate their answer to the preceding comprehension item, judging whether they thought their answer was right, wrong, or uncertain, comprising a 32- item test.

Metacognition was studied by examining students' evaluations of the accuracy of their answers to biology reading comprehension items. Correct metacognitive assessments (knowing you got the answer right and knowing that you didn't ) were expected to positively correlate with achievement; incorrect assessments (thinking you know when you don't and thinking you don't know when you do) were expected to negatively correlate with achievement. Being unsure whether you know the answer was expected to have a small positive correlation with achievement. Comprehension was defined as the total number of items correct on this reading test and achievement was measured by final grades. Data were collected on 75 students enrolled in Anatomy and Physiology 1, for non-science majors, at an ethnically diverse urban college. Most were majoring in nursing (75%) and other health professions; 74% were female and 26% were male. Most students were ethnic minorities: 30% African American, 20% Latino, 13% Asian, 9% Caucasian and 27% Other. Many of these students do not have English as their native language.

The mean number of correct comprehension items was only 10 out of 16 ( 62.5%). Comprehension showed a low but significant positive Pearson correlation with achievement (r=.30, p<.01). Students' judgements about the accuracy of their answers were compared to the actual accuracy of their answers, resulting in six metacognitive measures. Two were accurate metacognitions: knowing you know the answer (++) and knowing you don't know the answer (- -), two were inaccurate metacognitions, thinking you know the answer when you don't (+ -) and thinking you don't know the answer when you do (-+), and two were unsure - knowing the answer, but not being sure about it (+?), and not knowing the answer and not being sure about it (-?). These results are summarized in Table 2. The most common form of metacognition was the correct metacognitive assessment of knowing they knew the answer (Meta 1 ++), while the least common form was the incorrect metacognitive variable of knowing the answer, but thinking they didn't (Meta 2 +-) . The other incorrect metacognition, not knowing the answer but thinking that they did (Meta 4 - +) was the second most common occurrence. The other correct metacognition, not knowing the answer and knowing they didn't know it, occurred very infrequently.

Table 2

Summary of Results: Reading Comprehension, Metacognition and Course Achievement


Type of




1st Symbol

Answer to


2nd Symbol


of Answer


Rank Order

& (means)




Correl. with





+ +


- -

got right


got wrong

















+ -


- +

got right


got wrong




















+ ?


- ?

got right


got wrong

not sure


not sure











* p < .05

** p < .01

*** p < .001

Students most commonly thought they had the right answer to the comprehension questions, occasionally were unsure about the accuracy of their answer and rarely thought their answer was wrong. Three of the metacognitive measures showed significant Pearson correlations with comprehension and three correlated significantly with final grades. Knowing that they knew the answer had a high, positive correlation with comprehension (r=.86, p <.001), and a positive correlation with achievement (r=.37, p<.001). Getting the answer wrong but thinking it was right had a high negative correlation with comprehension (r= - .74, p>001), but no correlation with achievement. Two metacognitive measures that correlated negatively with final grades were: got the answer right but thought it was wrong (r=-.27, (p<.05) and got the answer wrong and was not sure if it was right/wrong (r=-.35, p<.01). Only two of the metacognitive measures correlated significantly with both achievement and comprehension. Getting the answer wrong and not being sure about it (-?) had negative correlations with achievement and comprehension and getting the answer right and knowing it was right (++) had positive correlations with achievement and comprehension.

Metacognition often differentiates successful from unsuccessful readers and students in general. Results of this study partially supported the prediction that correct metacognitive assessments would positively correlate with achievement. However, because some of the n=s were small for these different types of metacognition, the results are subject to chance and should be considered exploratory. The important points are that a metacognitive variable (knowing that you know) correlated more highly with achievement than the traditional cognitive reading variable (comprehension) did, and that the direction of the correlations is as predicted, so they suggest a real underlying relationship between reading metacognition and achievement, except for the prediction for the correct metacognitive assessment of "knowing that you don't know.@ Uncertain metacognition (doubting whether the answer is right or wrong) had stronger negative correlations with achievement than inaccurate metacognition. These variables should be studied with larger numbers of items to verify their stability. Also, reliability and validity studies should be conducted on the Biology Reading Test.

Many students in this study had difficulty understanding word/phrase meanings, main ideas and generalizations. It is somewhat astounding that the comprehension mean was only 10 out of 16, considering that students supposedly had learned the material they were reading in a previous, prerequisite course. How and to what extent do misunderstandings of the meaning of specific words or phrases inhibit further learning from text when reading biology? What are the implications of misunderstanding main ideas and generalizations? Reading difficulty appeared to be topic-dependent to some extent. Why? Were selections written at different levels, which Zook and Mayer (1994) might call an instructional variable? Were differences related to students' prior knowledge about the material, and/or a function of most these students being ethnic minorities, many of whom are not native speakers of English, which Zook and Mayer might call learner variables? Applying Perkins and Simmons' (1988) model, these students appear to have misunderstandings at three levels or in three domains: content, such as using vocabulary, problem solving, such as self-regulation of their test taking strategies, and epistemic, such as explaining the author's main idea.

Because mastery of reading comprehension and other academic objectives often is determined by multiple-choice tests, assessing whether one has selected the right answer is important for maximizing success. Self-regulating students learn with specific goals in mind, observe their performance as they work, evaluate progress in attaining their goals and react by continuing or changing their approach as needed, depending upon the value of the task and upon perceived self-efficacy (Schunk, 1991). How can students detect when they are making errors so they have the opportunity to self-correct before turning in their tests? The metacognitive assessment technique used here forces students to self-evaluate their answers to reading comprehension test items. When taking a test, students who evaluate their performance accurately are more likely to react appropriately by keeping and/or changing their answers to maximize their test score if they feel it is important.

We intended to administer the Biology Reading Test to students before they enrolled in the first anatomy and physiology course so we could advise students who might need substantial improvement in their reading comprehension before taking the course. We also planned to implement methods in the course to enhance students= reading and test-taking strategies so that their knowledge, understanding and performance in anatomy and physiology would improve.

Misconceptions in Biology

The second exploratory study identified misconceptions students have about anatomy and physiology and examined whether they were overcome after the two-course sequence in anatomy and physiology. The Biology Knowledge Test (Griswold and Lemons, 1995) was developed to assess students= prior knowledge of biology before taking anatomy and physiology. It was administered over a two-semester anatomy and physiology sequence, at the beginning of the first A&P course as a pretest and at the end of the second A&P course as a posttest. Content of the test spanned three levels. We knew that many of our students had reading comprehension problems, especially because many are not native speakers of English, and we wondered about other types of problems that might be interfering with our students= ability to master the material in Anatomy and Physiology 1 and 2. Mindful of the literature on scientific misconceptions we wondered whether these were affecting our students, and if so, how we might help them. This study, funded by a City College of New York President=s Fund for Innovation Grant, is a pilot project to identify and examine potential misconceptions and other faulty ideas in anatomy and physiology and see what happens to them over the two-semester sequence of anatomy and physiology courses. Which are overcome and which remain? Why? This pilot study focused on potential misconceptions at the most basic level. The results showed that students had many erroneous conceptions of biology. Of the numerous problematic conceptions and misconceptions identified at the beginning of Anatomy and Physiology 1, some were overcome and some remained by the end of Anatomy and Physiology 2. A few misconceptions were actually more common at posttesting than at pretesting. Problematic conceptions appeared to range from no conception to problematic conception to misconception. In this pilot study two different levels of faulty ideas were classified preliminarily as misconceptions (not all wrong answers are misconceptions): those at the level of the test question itself were called item misconceptions and those at the level of a wrong answer choice were called distractor misconceptions (Hartman, 1996c).

Why were some misconceptions overcome while others were not? The concepts the professors emphasize in course work tended to become clarified so misconceptions were overcome. However, just reading valid conceptions in the text did not appear sufficient for overcoming misconceptions. Some misconceptions remained entrenched and even worsened despite coursework emphasis. The results suggest there are many possible sources of faulty conceptions students have about anatomy and physiology. Some of these are:

1. genuine misconception: wrong idea about something as part of a working system of beliefs.

2.some conception, but incomplete understanding- good guess

3. no conception lack any idea - pick answer at random - wild guess.

4. confused by wording of question.

5. confused by labeling of diagram.

6. forget important concepts from other courses that need to be applied to the problem/question.

7. confused by similarity of concepts, diagrams or problems.

8. remember concepts but not sequence.

9. trouble with part/whole relationships, focusing on specific details instead of the general point or big picture or focusing on the big picture/general point and missing important details.

10. problems with both content knowledge & problem solving.

11. problems with whole/whole relations.

What were we to do with the information about students= misconceptions? Teaching science metacognitively includes awareness of such misconceptions as well as planning, implementing, monitoring and evaluating strategies to help students overcome them. Our grant contained funds to provide one-on-one tutoring for students to help them succeed in the course. After reviewing the literature on conceptual change models, Driver's (1987) model was adapted as follows:

1. Orientation: introduction to the topic and motivation.

2. Awareness: Recognition of misconceptions may occur through:

a. independent recognition: without feedback from an external source, by the student using her/his own knowledge and reasoning which interacts with the context to discover there is a misunderstanding.

b.disconfirmational feedback: students are exposed to information from an external source (e.g. lab experiment/ professor/tutor/book/) that directly contradicts their conceptions; change through cognitive conflict.

c. relational recognition: students are exposed to information that is related to their conceptions, and this information helps them discover that their conceptions are inadequate.

d. induced recognition: students are directly told their conception is invalid; they are confronted with conflicting concepts and facts (Hartman, 1981).

3. Elicitation: explication of student ideas and misconceptions. dissemble concepts into component parts - deconstruct. (Like Gagne's 1965 task analysis- break component knowledge and skills into a learning hierarchy).

4. Restructuring: students are receptive to changing their conceptions. New and revised conceptions are integrated. Students exchange and clarify ideas after exposure to conflicting meanings, recursively expanding and reworking information.

5.Application: consolidation of new or restructured ideas by using them to solve problems or answer questions.

6 Review: reflection on concepts, what they are, when, why and how they are used; what they are related to - how they fit into the big picture.

The plan was for the tutor to apply this model when working individually with students. However, we were unable to implement it because the School of Nursing was retrenched so at least 75% of our students were gone and science majors, who quickly filled their seats, did not have the same conceptual difficulties (i.e., reading comprehension and misconceptions). Nevertheless, the course restructuring efforts were not done in vain because the changes led to these Anatomy and Physiology courses becoming popular with science majors for the first time.

This project shows several examples of metacognition in teaching. First, we became aware of the need to consider students= misconceptions about anatomy and physiology, because the high failure rate of students in the courses made us wonder whether misconceptions might be contributing to their difficulties. Second, we developed a strategy for identifying students= misconceptions. Third , we developed a plan to help students overcome their misconceptions. Fourth, we realized that the students= reading problems were even more fundamental than their misconceptions about biology. Finally, we realized that the new students taking Anatomy and Physiology did not have the same types of misconceptions as the original group so Professors Lemons and Grisowld adapted instruction to the needs of the new population.

To summarize, my work on metacognition in science teaching and learning has included faculty and curriculum development to promote and analyze metacognition in teaching biology, and research and academic support to promote and analyze metacognition in learning biology.



Science teaching and learning are complex processes, both because of the content and thinking skills required to understand science at a deep enough level to be meaningful and useful. Metacognition helps science teachers think about how they manage curriculum, instruction and assessment, as well as systematically reflect on what they teach, why and how. Metacognition helps science learners develop and use effective and efficient strategies for acquiring, understanding, applying and retaining extensive and difficult concepts and skills. Good science teaching requires teaching both with with own metacognition and for the development of their students= metacognition.


Abimbola, I.Ol, & Baba, S. Misconceptions and Alternative Conceptions in Science Textbooks: The Role of Teachers as Filters. The American Biology Teacher. 58(1). 14-19

Anderson, R.C. and Kuhlavy, R.W. (1972). Imagery and prose learning. Journal of Educational Psychology, 63, 242-243.

Anderson, C., Sheldon, T. & Dubay, K. (1990). The effects of instruction on college nonmajors conceptions of respiration and photosynthesis. Journal of Research in Science Teaching. 27, 761-776.

Baird, J. & White, R. (1984). Improving Learning through Enhanced Metacognition: A Classroom Study. Paper presented at the Annual Meeting of the American Educational Research Association. New Orleans.

Barman, C., Benz, R. Haywood, J. & Houk, G. (1992). Science and the Learning Cycle. Perspectives in Education and Deafness. 11(1) 18-21.

Bransford, J. & Stein, B.. (1984 ) The IDEAL Problem Solver: New York: Freeman.

Blanc, R., De Buhr, L., & Martin, D. (1983). Breaking the Attrition Cycle. Journal of Higher Education. 54(1). 80-90.

Carmichael, Jr. J., Ryan, M. A., Jones, L., Hunter, K & Vincent, H. (1981). Project SOAR.

New Orleans: Xavier University of Louisiana

Cook, L. & Mayer, R. E. (1988). Teaching Readers about the structure of scientific text.

Journal of Educational Psychology. 80(4), 448-456.

Crooks, T. 1988. Impact of classroom evaluation practices on students. Review of Educational Research. 58(4). 438-481

De Avila, E & Duncan, S.. 1985. The language-minority child: A psychological, linguistic, and social analysis. in S. Chipman, J. Segal, and R.Glaser Thinking and Learning Skills: Vol. 2. Research and Open Questions. Hillsdale, NJ, Lawrence Erlbaum Associates.

Driver, R. (1987). cited by Skane and Graeber (1993). A conceptual change model implemented with college students: Distributive law misconceptions. The Proceedings of the Third International Seminar on Misconceptions and Educational Strategies in Science and Mathematics. Ithaca, NY: Misconceptions Trust

Duit, R. (1991). Students' conceptual frameworks: Consequences for learning science. In S. M.

Glynn, R. Yeany, and B. Britton (Eds.) The Psychology of Learning Science. (pp. 65-85), Hillsdale, NJ: Erlbaum & Associates.

Eylon, B. & Linn, M. (1988) Learning and Instruction: An examination of four research perspectives in science education. Review of Educational Research. 58, 251-301

Felder, R. 1993. Reaching the second tier - Learning and teaching styles in college science education. Journal of College Science Teaching. .March/April 286-290.

Ford, K., Canas, A., Jones, J., Stahl, H., Novak, J.. & Adams-Weber, J. (1991). ICONKAT: An integrated constructivist knowledge acquisition tool. Knowledge Acquisition. 3, 215-236.

Gagne, R. M. (1965). The Conditions of Learning. New York: Holt.

Glynn, R. Yeany, and B. Britton (Eds.) The Psychology of Learning Science. (pp. 65-85), Hillsdale, NJ: Erlbaum & Associates

Griswold, J. & Hartman, H. (1991).Restructuring Introductory Biology for Success with Minority Students: A Developmental Approach. Invited participation in City University of New York Colloquium On Introductory Science Courses: New Approaches to Teaching and Learning. City University of New York, New York

Griswold, J. & Lemons, D. (1995/1996) Biology 332/333 Pretest. City College of City University of New York. NY.

Griswold, J., Lemons, D. & Hartman, H. (1995). Developing critical thinking and content mastery in the Anatomy and Physiology Laboratory: A Workshop. Presented at the Annual Meeting of the National Science Teaching Association: Philadelphia PA

Hartman, H. (1999). A Blueprint for Reform? Learning, Creating,and Using Knowledge: Concept Maps as Facilitative Tools in Schools and Corporations. Contemporary Psychology: APA Review of Books. 44(1) 111-113.

Hartman, H. (1996a). Cooperative Learning Approaches to Mathematical Problem Solving. In: The Art of Problem Solving: A Resource for the Mathematics Teacher. Alfred S. Posamentier & Wolfgang Schulz (Eds.). Corwin Press, Inc. Thousand Oaks CA (1996) pp.401-430

Hartman, H.(1996b) Reading Comprehension College Reading and Learning Association, Albuquerque, NM

Hartman, H. (1996c) Misconceptions in Anatomy and Physiology. Human Anatomy and Physiology Society. Portland, OR

Hartman, H. And Griswold, J. (1994). Restructuring Introductory Biology. Paper presented at the Annual Meeting of the American Educational Research Association. New Orleans.

Hartman, H. & Sternberg, R.J. (1993). A broad BACEIS for improving thinking. Instructional Science. 21 (5) 400-425

Hartman, H. & Griswold, J. (1991) Metacognition and Cognition in Learning Science. Invited participation in City University of New York Colloquium On Introductory Science Courses: New Approaches to Teaching and Learning. City University of New York, New York

Hartman-Haas, H. (1981). Story Clarification: Comprehension-Directed Problem Solving.

Dissertation Abstracts International 41 (10)

Herman, J. 1992. What research tells us about good assessment. Educational Leadership. 49(8) 74- 78.

Hooper, S. & Hannafin, M. 1991. Psychological Perspectives on Emerging Instructional Technologies: A Critical Analysis. Educational Psychologist 6(1), 69-95.

Karplus, R. 1974. The Science Curriculum Improvement Study. (SCIS) Lawrence Hall of Science, University of California, Berekely

Lemons, D. (1994). A Laboratory Guide for Human Anatomy and Physiology. New York: City College of City University of New York

Mayer, R. (1989). Models for Understanding. Review of Educational Research. 59(1). 43-64.

Mcintosh, W.J. (1986) The effect of imagery generation on science rule learning. Journal of Research on Science Teaching. 23(1) 1-9.

McDermott, L. (1991). Millikan Lecture 1990: What we teach and what is learned - Closing the gap. American Journal of Physics, 59 (4) 301-315.

McDermott, L. (1984). Research on conceptual understanding in mechanics. Physics Today. 37, 24-32.

McKeachie, W., Pintrich, P., Lin, Y.G., & Smith, D. (1986). Teaching and Learning in the College Classroom: A Review of the Research Literature. Ann Arbor: University of Michigan

Mestre, J. (1994). Problem Posing as a Tool for Probing Conceptual Development and Understanding in Physics. Paper presented at the Annual Meeting of the American Educational Research Association, New Orleans

Minstrell, J. (1989). Teaching science for understanding. In. L. Resnick & L. Klopfer (Eds.) Toward the Thinking Curriculum: Current Cognitive Research. Association for Supervision and Curriculum Development, Alexandria, VA.

Narode, R., Heiman, M. Lochhead, J. & Slomianko, J. ( 1987) Teaching Thinking Skills: Science. Washington, D.C. : National Educational Association.

National Research Council (1996). National Science Education Standards. Washington D.C.: National Academy Press

Novak, J. (1998). Learning, Creating and Using Knowledge.Lawrence Erlbaum Associates:Mahwah NJ

Novak, J. (1990). Concept maps and Vee diagrams: two metacognitive tools to facilitate meaningful learning. Instructional Science. 19(1). 29-52.

Novak, J. D., Gowin, D. B., & Johansen, G. T. (1983). The use of concept mapping and knowledge Vee mapping with junior high school science students. Science Education, 67(5)625- 645.

Nussbaum, J. & Novick, S. (1982). Alternative frameworks, conceptual conflicts and accommodation: Toward a principled teaching strategy. Instructional Science. 11, 183- 200.

Perkins, D. and Simmons, R. (1988). An integrative model of misconceptions. Review of Educational Research. 58(3) 303-326.

Pressley, M. and McCormack, C. (1995). Advanced Educational Psychology. Harper Collins: NY

Posner, G., Strike, K., Hewson, P, & Gertzog, W. (1982). Accommodation of a Scientific Conception: Toward a Theory of Conceptual Change. Science Education, 66, 211-228.

Redish, E. F. (1994). The Implications of Cognitive Science for Teaching Physics. American

Journal of Physics, 62(6). 796-803.

Regis, A. & Albertazzi, P. G. (1996). Concept Maps in Chemistry Education. Journal of Chemical Education. 73 (11). 1084-1088

Roadrangka, V., Yoany, R., & Padilla, M. (1983). The Construction and Validation of Group Assessment of Logical Thinking (GALT). Paper presented at the annual meeting of the National Association for Research in Science Teaching. Dallas TX

Rosenshine, B. & Meister, C. 1992 The use of scaffolds for teaching higher level cognitive strategies. Educational Leadership.April. 26-33.

Roth, K. (1991). Reading science for conceptual change. In C. M. Santa & D. E. Alvermann (Eds.) Science Learning: Processes and Applications. (pp. 48-63). Newark DE: International Reading Association.

Roth, V., Strozak, C., Cracolice, M. & Gosser, D. (1997). The Workshop Model: Peer Leadership and Learning. A Guidebook. City College of City University of New York

Rubin, Rochelle L. and Norman, John T. (1989) A comparison of the effect of a systematic modeling approach and the learning cycle approach on the achievement of integrated science process skills. Paper presented at the annual meeting of the National Association for Research in Science Teaching San Francisco CA ED 305268

Schneps, M. (1994). A Private Universe. San Francisco: The Astronomical Society of the Pacific

Schunk, D., 1991. Self-efficacy and academic motivation. Educational Psychologist, 26(3 & 4), 199-205.

Shavelson R. & Baxter, G. 1992. What we've learned about assessing hands-on science. Educational Leadership 49(8). 20-25

Spiegel, G. (1996). Text structure awareness and graphic postorganizers of reading. Workshop presented at the Tenth Annual Human Anatomy and Physiology. Portland OR

Spiegel, G. and Barufaldi, J. (1994). The effects of a combination of text structure awareness and graphic postorganizers on recall and retention of science knowledge. Journal of Research in Science Teaching. 31(9) 913-932.

Sternberg, R. (1985). Beyond IQ: A Triarchic Theory of Intelligence. New York: Cambridge University Press.

Stewart, J., Hafner, R., Johnson, S. & Finkel, E. 1992. Science as model building: Computers and high-school genetics. Educational Psychologist, 27(3) 317-336.

Tobias, S. (1986). Peer Perspectives on the Teaching of Science. Change. March/April. 36-41

Walberg, H. 1991. Improving School Science in Advanced and Developing Countries..Review of Educational Research. 61(1) 25-69.

Whimbey, A. & Lochhead, J. (1982) Problem Solving and Comprehension. Franklin Institute Press: Philadelphia





Work on restructuring Anatomy and Physiology was supported by NSF grants with Daniel Lemons and Joseph Griswold (DUE-93544477 and DUE-9451852).

I am grateful to Joseph Griswold and Daniel Lemons for their assistance in developing and administering the Biology Reading Test.

Figure 1

Sample Concept Map

from Hope Hartman=s Chapter 1

AMetacognition in Science Teaching and Learning@

Figure 2

Modified Learning Cycle Used to Design

the New A & P Curriculum

from Hope Hartman=s Chapter 1

AMetacognition in Science Teaching and Learning@