In representational theories of mind, concepts are mental units that are bearers of meaning and that can be mapped onto single words such as dog, animal, grow, and die. They are used to form beliefs, such as Fido is a dog or Dogs can die, guide inference and organize knowledge of the world. Conceptual change occurs when there is change in what these mental units are and how they are articulated within larger conceptual systems. Although there are obviously different degrees of conceptual change, ranging from adding a new concept that fits within an existing conceptual system (e.g., learning about a new type of dog) to changing the organizing principle of an entire classification system (e.g., recognizing species as interbreeding populations of individuals that change over time rather than as unchanging ideal types), the term conceptual change has typically been reserved for forms that involve significant restructuring. These changes are notoriously difficult to make because they involve coordinated changes in multiple concepts, rather than simple additions or elaborations. For example, if children say “A lamp is alive,” “The Earth is flat,” or “This piece of clay weighs nothing at all,” they may be expressing beliefs that are true from the perspective of their conceptual systems. Changing their beliefs involves more than telling them they are wrong or providing new information; it involves helping them develop different concepts of the Earth, alive, or weight. An important challenge is to identify cases of learning that involve significant restructuring and to understand how it occurs.
In his influential book, The Structure of Scientific Revolutions (1962), Thomas Kuhn (1922–1996) challenged the view of science as the steady accumulation of knowledge by describing the profound shifts in the meaning of concepts that have occurred when scientists have moved from one dominant paradigm to another. He argued that scientists' work is governed by a paradigm shared by members of that community—a complex structure of theory and laws, preferences about instrumentation, and more general assumptions about what kinds of entities exist and what form laws should take—that helps fix the meaning of common vocabulary, as well as provide the rules for problem solving in a given domain. He distinguished between normal science, periods that involve routine puzzle solving for problems defined by a paradigm, and revolutionary science, periods of crisis during which a dominant paradigm fails to solve key problems and is eventually overthrown by a new paradigm that gains allegiance from the community of scientists. The changes in worldview and commitments about what entities exist and how they interact can be so great when scientists switch paradigms that Kuhn argued that members of the community effectively no longer speak the same language.
In the 1980s George Posner and colleagues (1982) brought these ideas of conceptual change to the science education community. Challenging the prevailing model of learning in which students were seen as gradually accumulating new beliefs based on generalizing from observation, he and his colleagues proposed that in learning current scientific theories, students needed to undergo conceptual revolutions that in many ways were analogous to the paradigm shifts Kuhn had identified in the history of science. They called for a new “conceptual change” model of learning that acknowledged the powerful role children's ideas play in observing, defining problems, and making sense of new information. Indeed, researchers were discovering that adolescents brought a wealth of ideas to the science classroom. Many were strongly held ideas that proved to be extremely resistant to instruction, and some resembled earlier theories in the history of science—for example, the impetus theory in mechanics (McCloskey, 1983) or the source-recipient model of heat (Wiser, 1988).
At the same time Susan Carey (1985) brought the ideas of conceptual change to the developmental psychology community. This group was already keenly aware that development involved qualitative shifts in children's ideas about number, physical quantities, space, time, mind, and life, thanks to the pioneering work of Bärbel Inhelder (1913–1997) and Jean Piaget (1896–1980). However, there was growing dissatisfaction with Piaget and Inhelder's explanations of these changes in terms of the progressive construction of more powerful domaingeneral thinking structures (i.e., shifts from sensori-motor, to pre-operational, concrete operational and finally formal operational thought) in part because children's thinking did not show the kinds of consistency expected on these accounts. Carey proposed that it would be more fruitful to think of children's thought as constrained by domain specific structures (intuitive theories akin to Kuh-nian paradigms). In her view, even preschoolers and elementary school students have implicit intuitive theories in which their everyday explanatory concepts are embedded and play a role, which guide their patterns of inference and problem solving, and which can be fundamentally revised in the face of new information from their culture. Thus, conceptual change came to be seen not only as occurring in mature scientists and students receiving explicit science instruction, but also in younger children in the normal course of their development.
Not all aspects of learning or cognitive development involve conceptual change. Much learning fills in the gaps and elaborates on a given conceptual structure rather than radically restructures it. Change may also involve units other than concepts; for example, beliefs, mental models, strategies, and event scripts. Further, not all aspects of cognitive development involve learning; some profound changes may depend primarily on maturational processes.
However, there are now a number of cases in which change does seem to involve a fundamental restructuring of concepts. In each case, researchers have provided detailed analysis of the changing structure of the concepts across two conceptual systems, empirical evidence that children show the kinds of systematicity expected on these analyses, and evidence that learning shows significant resistance to change. These cases span changes that occur for students during the elementary school years to cases that only rarely occur even among college students. Most of the work has focused on science concepts, but there are some well worked out examples in mathematics and other fields. Indeed, in a couple of cases, relevant conceptual changes in science and mathematics may be linked.
One of the first cases examined was children's concept of the Earth. Joseph Nussbaum (1985) found that how children think about the shape of the Earth has implications for their concepts of space and gravity. Initially children think of the Earth as flat, space as bounded, with an absolute up/down orientation, and objects as falling down to earth. As children come to accept that the Earth is a sphere, they re-conceptualize space as unbounded, extending in all directions from the Earth, with gravity acting towards its center. Stella Vos-niadou and William Brewer (1992, 1994) extended this analysis by showing how children's initial Earth concept was constrained by the presuppositions of a larger framework theory—a näıve physics in which the Earth was seen as a kind of flat, stationary, physical object. They found that children invent many synthetic models of the Earth (such as the flat disc and dual Earth models) in an attempt to reconcile the new information that the Earth is round with their initial presuppositions. Coming to understand the earth as spherical, moving, and unsupported involves constructing and assigning the Earth to a new ontological category—astronomical object—with distinctive properties; this in turn changes the framework they can use for explaining the causes of day/night or the seasons.
The development of children's concepts of matter, weight, and density is another case that may involve restructuring a network of concepts. Carol Smith (2007) found evidence for an initial conceptual system that uses an undifferentiated concept of weight/density (one that unites the elements heavy and heavy for size) and a later system in which weight and density are fundamentally different kinds of quantities that figure in different generalizations. That is, weight is an extensive quantity whose magnitude varies with the amount of matter, whereas density is an intensive quantity whose magnitude is independent of amount of matter. Movement from conceptual system one to two involves coordinated changes in multiple concepts: representing weight and volume as measured quantities mapped to number rather than as perceptual magnitudes, representing matter as a fundamental constituent that occupies space and has weight rather than something that can be seen, felt, and touched; and coordinating weight and volume in a new concept of density. There was evidence of coherency in reasoning patterns among children who failed to differentiate weight and density as well as those who successfully did. In teaching studies, there was both evidence of resistance to change among some children and coordinated patterns of change among multiple concepts for those who were more successful.
The conceptual restructuring that occurs in children's intuitive matter theory may be related to a profound restructuring that occurs in mathematics. During the early years, children develop and entrench a rich concept of counting numbers, positive integers as represented by the integer list and that participate in operations of addition and subtraction. For each positive integer, there is an answer to the question “Which is the next one?” It is the next word in the count list. Later, they are exposed to new entities—fractions and decimals. To develop an understanding that fractions and decimals are numbers, children must restructure their concept of number from count number to rational number and make a host of changes, including developing a mathematical understanding of division. They must also rethink many core assumptions about what numbers are. For example, in the count list, there are no numbers between the integers, but for rational numbers there are. Various reflections of conceptual understanding of fractions develop in parallel across samples of children (Gelman, 1991). Researchers have also found evidence of resistance to change, as children initially try to assimilate fractions to their concept of counting number. Finally, understanding rational number was strongly related to understanding weight as a continuous property of matter (Smith et al., 2005).
There are other cases in which the current scientific theories are deemed even more wildly counter-intuitive and fail to be understood by most adults. Coming to understand Newtonian dynamics, thermal physics, and Darwinian theory of natural selection are prime examples. For example, Andrew Shtluman (2006) identified two contrasting frameworks for understanding evolution: a transformationist framework (that characterized most students) and the variational framework of Charles Darwin (1809–1882). Students think of species as having essences, leading them to downplay variation within species and to think of evolution as a directed process of transforming the species essence over time. In contrast, Darwin thought of species as populations of inter-breeding individuals, making individual variation more important and salient, with evolution involving the two-step process of production of new variations followed by selection. Most individuals reasoned about six diverse phenomena in ways that were consistent with a given framework. Further, there was evidence of resistance to change, as students were actually asked to answer as they thought Darwin would.
Researchers have also characterized the contrasting frameworks for understanding knowledge acquisition in science, ranging from simpler knowledge-unproblematic epistemologies in which knowledge is seen as accumulating through observation, to more complex knowledge-problematic epistemologies in which conceptual frameworks guide inquiry. Movement from one framework to the other involves making fundamental differentiations (e.g., differentiating ideas from evidence, theories from hypotheses) as well as reanalyzing the meaning of concepts (e.g., scientific truth; scientific model). Studies have shown some consistency in student reasoning across tasks, as well as considerable resistance to change. Most college students still embrace a limited knowledge-unproblematic epistemology, although significant conceptual restructuring can occur with unusual science teaching experiences in elementary school (Smith et al., 2000).
Most theorists see conceptual change as involving change at multiple interacting levels. First, there are changes in the internal structure of concepts: changes in the attributes represented in a concept and the weighting of those attributes. For example, movement to a differentiated weight concept involves adding a representation of weight as a measured quantity mapped to number, making this new representation central to the meaning, moving felt weight from core to periphery, and removing heavy for size as a relevant attribute of weight at all. Second, there are changes in the external structure of concepts: the collection of beliefs stated using the concept, including the relations formulated between concepts and ways that concepts are used in explanations. For example, as children differentiate weight and density, they can explicitly state their relation and use these concepts in separate generalizations.
In addition, many conceptual change researchers and developmental psychologists argue that there are multiple interacting levels of external structure. More specifically, some have proposed a distinction between larger framework theories and the more specific intuitive theories that can be formulated within a given framework theory (Wellman & Gelman, 1992), inspired by a similar distinction made by philosophers and historians of science. Framework theories delimit different domains of human reasoning by identifying the important classes of entities that need to be considered in the domain (its ontology) and the kinds of causal relations that are expected to occur among those entities. For example, preschool children may have at least three framework theories: a näıve physics (centered on physical objects that interact via mechanical causality), a näıve psychology (centered on people whose actions are caused by their goals, intentions and desires), and the beginnings of a näıve biology (centered on animals whose growth and development is governed by innate potential and whose movement and health is powered by vital forces). The existence of framework theories explains how children are able to make certain classes of inferences readily, often in the face of limited data, and why new beliefs, concepts, mental models, and even specific theories can be formed relatively quickly and fluidly when consistent with a framework theory. In contrast, more radical forms of conceptual restructuring involve changes in the larger framework theory. These changes are difficult because they involve changing basic ontological and epistemological assumptions, as well as expectations about types of causal patterns that will occur.
Not all conceptual change researchers embrace the notion that children's concepts are embedded in intuitive theories. Micheline Chi (2008) acknowledges the critical importance of ontological categories in organizing concepts, but describes them more generally as schemas rather than as part of theories. Andrea diSessa (1993) acknowledges the explanatory component to children's concepts, but suggests that children's explanatory ideas are too shallow and uncoordinated with each other to be well described as a systematic theory, at least in their näıve mechanics. The framework theory proposal, however, may capture the abstract level at which there is consistency in children's thought while leaving room for children to be ignorant about the details needed for a well worked out specific theory, and for much fluidity in children's thought.
Researchers have also considered what stays the same in conceptual change. Carey (2008) notes that not all concepts are reworked even in conceptual revolutions; some survive and serve as sources of stability. DiSessa identifies subconceptual elements, or phenomenological primitives, such as “balancing” or “overcoming,” that he thinks are as close as one gets to stable elements of thought and that are the explanatory bedrock of intuitive thought. He sees conceptual change as shifts in salience and cuing priorities among these elements, not replacing or eliminating these elements. Indeed a main contribution of his research program is to call attention to the productive resources within intuitive thought and the ways they continue to be used in later, more well structured theories.
If student learning is constrained by the assumptions of their guiding framework theory, how can new framework theories be learned? Instructional texts often focus on refuting specific beliefs. But if the faulty student belief stems from a faulty ontology, then refutation at that level will not be that effective. Instead, students will discount the data in numerous ways as they reinterpret it within their existing views rather than perceive it as anomalous (Chinn & Brewer, 1993). Even if they recognized an anomaly, they would not know what they needed to change to resolve it.
Chi argues that instruction in these cases would be more effective if the student were directly taught a new (more appropriate) ontological category. For example, she argues that many of the mistaken beliefs students develop about forces, heat, electricity, light, and magnetism stem from their assigning these concepts to the ontological category of “substance-kind entities” or “direct causal processes,” and that these scientific concepts can be better understood if thought of as “emergent causal processes.” Because students do not initially have such a category, she advocates directly teaching it by telling students the ways that emergent causal processes are different from direct causal processes. She found that college students who had a brief (1-hour) introduction to the category of emergent causal process, explained with some computer animations and several worked examples, had improved conceptual understanding of electricity in a follow-up unit compared to controls (Slotta & Chi, 2006).
Carey (2008) argues that conceptual change often rests on multi-step, iterative bootstrapping processes. One step involves the (culture's) introduction of a network of external symbols that function as (at best) partially interpreted placeholders (e.g., a drawing of a number line, or children being told “You need to eat to grow and stay alive”). These new networks of symbols may initially be learned by rote, where their meaning is captured more by their relations to each other than to prior knowledge. A second step involves coming to interpret the placeholders, using a variety of non-deductive modeling processes (e.g., analogies, thought experiments, inference to best explanation) that allow children to build connections with prior knowledge and observations in the world.
Both John Clement (2008) and Nancy Nersessian (1992) have studied how modeling processes work, both in students and in scientists, and how they can lead to the construction of fundamentally new representations. Crucially, models draw on people's everyday capacity to run mental simulations of events, in which they can both simplify their representations of events (by selectively abstracting away from surface details) and enrich them (by adding in depictions of unseen entities and processes, such as moving and colliding atoms, that help explain how things work). In creating these representations, individuals combine elements from diverse analogies, visual images and sensori-motor schemes in novel ways. Overall, Clement sees engaging students in incremental cycles of generating, evaluating, and revising explanatory models as at the heart of conceptual change.
Reasoning about models, of course, does not occur in a vacuum, but depends critically on social discourse and argumentation (e.g., dialogue about whether the model makes sense and explains the phenomena in question) and on using a wide variety of external symbols, inscriptions, and cultural artifacts. For this reason, many have extended analyses to include the study of how different discourse practices, participant structures, and cultural tools contribute to learning in conceptual change.
Finally, there is increasing recognition of the active role of students in regulating their own learning (Sinatra & Mason, 2008), guided by personal goals, interests, values, metacognitive understandings, and epistemic ideals.
Successful instruction calls for the careful orchestration of multiple practices that actively engage students' initial ideas and help them restructure them in order to build a better understanding of their world. There is much art in this orchestration, as well as much that is domain specific. There are a number of examples of instructional sequences that have led to conceptual change in different domains.
Jim Minstrell, a high school physics teacher, is one of the pioneers in crafting a high school physics curriculum that helps students to build a deeper understanding of the core concepts in Newtonian mechanics (Minstrell, 1984). His instruction combines diagnostic quizzes that engage students in making predictions about everyday events and explaining their reasoning, classroom discussion of these predictions that leads to outlining a range of views, classroom demonstrations and experiments that often produce discrepant events, followed by further discussion to sort out and interpret findings in terms of a network of concepts. It also involves the careful sequencing of benchmark lessons and revisiting ideas throughout instruction. More accessible ideas are built up first so they can be used as resources for building other ideas later through logical arguments.
Jim Stewart collaborated with high school teachers to develop modules in genetics, evolution, and observational astronomy that are attentive to student starting ideas and that develop student understanding through engaging them in model-based inquiry (Stewart, Cartier, & Passmore, 2005). All modules start with activities that promote explicit understanding of a model as a conceptual structure used to explain data and guide inquiry, because they found students enter thinking of models only as representations and need a better framework for later discussions. In the evolution module, students contrast three models of species adaptation and diversity: the model of intelligent design, created by William Paley (1743–1805) the model of acquired characteristics, developed by Jean-Baptiste Lamarck (1744–1829), and the model of natural selection, created by Darwin. Contrasting these models was an effective way of helping them sort out what Darwin was claiming from what he was not, thus avoiding many common misconceptions about natural selection. Students then extend their understanding of natural selection through applying this model to make sense of real data sets. The other modules also develop and inter-relate multiple models in a carefully sequenced manner.
Others have successfully engaged elementary and middle school students in model-based inquiry that simultaneously promotes conceptual change and greater metacognitive understanding of science. For example, Barbara White's ThinkerTools curriculum (1993) involved sixth grade students in constructing, testing, and revising models of force and motion, through designing and carrying out investigations with computer micro-worlds and real-world experiments. As students moved through the curriculum, the microworlds became more complex (e.g., progressively adding in effects of friction, mass, and gravity), allowing them to sequentially build up more complex conceptual models. Microworlds have been used productively as additional resources in promoting conceptual change in a variety of domains (Snir & Smith, 1995; Wiser & Amin, 2001). Finally, some educators have redesigned their whole approach to teaching grades 1 to 6 science (Hennessey, 2003) or science and math (Lehrer, Schauble, Strom, & Pligge, 2001) with a focus on engaging children with model building in the service of developing and testing their own ideas. These complex curricula not only have successfully built domain specific understandings, but showed that elementary school students can develop insights about the conceptual change process and the role of frameworks in inquiry that elude many adults!
Carey, S. (1985). Conceptual change in childhood. Cambridge, MA: MIT Press.
Carey, S. (2008). The origin of concepts. New York: Oxford University Press.
Clement, J. (2008). Creative model construction in scientists and students: The role of analogy, imagery, and mental simulation. Dordrecht, The Netherlands: Springer.
Chi, M. (2008). Three types of conceptual change: Belief revision, mental model transformation, and categorical shift. In S. Vosniadou (Ed.), International Handbook of Research on Conceptual Change. Boca Raton, FL: Routledge.
Chinn, C. A., & Brewer, W. F. (1993). The role of anomalous data in knowledge acquisition: A theoretical framework and implications for science instruction, Review of Educational Research, 63(1), 1–49.
diSessa, A. A. (1993) Toward an epistemology of physics. Cognition and Instruction, 10(2/3), 105–225.
Gelman, R. (1991). Epigenetic foundations of knowledge structures: Initial and transcendent constructions. In S. Carey and R. Gelman (Eds.), The epigenesis of mind: Essays on biology and cognition (pp. 293–322). Hillsdale, NJ: Erlbaum.
Hennessey, M.G. (2003). Probing the dimensions of metacognition: Implications for conceptual change teaching-learning. In G. Sinatra & P. Pintrich (Eds.), Intentional conceptual change (pp. 103–132). Mahwah, N.J.: Lawrence Erlbaum Associates.
Kuhn, T. (1962). The structure of scientific revolutions. Chicago: University of Chicago Press.
Lehrer, R., Schauble, L., Strom, D., & Pligge, M. (2001) Similarity of form and substance: Modeling material kind. In S. Carver & D. Klahr (Eds.), Cognition and instruction: Twenty-five years of progress (pp. 39–74). Mahwah, NJ: Lawrence Erlbaum Associates.
McCloskey, M. (1983). Näıve theories of motion. In D. Gentner & A.L. Stevens (Eds.), Mental models (pp. 299–324). Hillsdale, NJ: Erlbaum.
Minstrell, J. (1984). Teaching for the development of understanding of ideas: Forces on moving objects. In C. Anderson (Ed.), AETS Yearbook: Observing Classrooms: Conceptual Change Perspectives from Research and Practice (pp. 55–73). Columbus: Ohio State University.
Nersessian, N. (1992). Constructing and instructing: The role of ‘abstraction techniques’ in developing and teaching scientific theories. In R. Duschl & R. Hamilton (Eds.), Philosophy of Science, Cognitive Science, & Educational Theory and Practice (pp. 48–68). Albany, NY: SUNY Press.
Nussbaum, J. (1985). The Earth as a cosmic body. In R. Driver, E. Guesne, & A. Tiberghien (Eds.), Children's ideas in science (pp. 170–192). Philadelphia: Open University Press.
Posner, G. J., Strike, K. A., Hewson, P. W., & Gertzog, W.A. (1982). Accommodation of a scientific conception: Toward a theory of conceptual change. Science Education, 66(2), 211–227.
Shtulman, A. (2006) Qualitative differences between näıve and scientific theories of evolution. Cognitive Psychology, 52, 170–194.
Sinatra, G., and Mason, L. (2008) Beyond knowledge: Learner characteristics influencing conceptual change. In S. Vosniadou (Ed.), International Handbook of Research on Conceptual Change. Boca Raton, FL: Routledge.
Smith, C., Maclin, D., Houghton, C. & Hennessey, M.G. (2000). Sixth graders' epistemologies of science: The impact of school science experiences on epistemological development. Cognition and Instruction, 15(3), 317–393.
Smith, C., Solomon, G., & Carey, S. (2005). Never getting to zero: Elementary school students' understanding of the infinite divisibility of matter and number. Cognitive Psychology 51, 101–140.
Smith, C. (2007). Bootstrapping processes in the development of students' commonsense matter theories: Using analogical mappings, thought experiments, and learning to measure to promote conceptual restructuring. Cognition and Instruction, 25(4), 337–98.
Snir, J., & Smith, C. (1995) Constructing understanding in the classroom: Integrating laboratory experiments, student and computer models, and class discussion in learning scientific concepts. In D. Perkins, J. Schwartz, M. West, & S. Wiske (Eds.), Software goes to school (pp. 233–254). Oxford, England: Oxford University Press.
Stewart, J., Cartier, J. & Passmore, C. (2005). Developing understanding through model-based inquiry. In M. S. Donovan and J. D. Bransford (Eds.), How Students Learn: Science in the Classroom (pp. 515–565). Washington, DC: The National Academies Press.
Thagard, P. (1992). Conceptual revolutions. Princeton, NJ: Princeton University Press.
Vosniadou, S., & Brewer, W. (1992). Mental models of the Earth: A study of conceptual change in childhood. Cognitive Psychology, 24, 535–585.
Vosniadou, S., & Brewer, W. (1994). Mental models of the day/ night cycle. Cognitive Science, 18, 123–183.
Welman, H., & Gelman, S. (1992). Cognitive development: Foundational theories of core domains. Annual Review of Psychology, 43, 337–375.
White, B. (1993). Thinkertools: Causal models, conceptual change, and science instruction. Cognition and Instruction, 10, 1–100.
White, B., & Frederickson, J. (1998). Inquiry, modeling, and metacognition. Making science accessible to all students. Cognition and Instruction, 16(1), 3–117.
Wiser, M. (1988). The differentiation of heat and temperature: History of science and the novice-expert shift. In S. Strauss (Ed.), Ontogeny, phylogeny, and historical development (pp. 28–48). Norwood, NJ: Ablex Publishing Company.
Wiser, M., & Amin, T. (2001). Is heat hot? Inducing conceptual change by integrating everyday and scientific perspectives on thermal phenomena. Learning & Instruction, 11, 331–355.
- Coats and Car Seats: A Lethal Combination?
- Kindergarten Sight Words List
- Child Development Theories
- Signs Your Child Might Have Asperger's Syndrome
- 10 Fun Activities for Children with Autism
- Why is Play Important? Social and Emotional Development, Physical Development, Creative Development
- The Homework Debate
- Social Cognitive Theory
- GED Math Practice Test 1
- First Grade Sight Words List