Reprinted by permission from:  Research in Physics Learning: Theoretical Issues and Empirical Studies.  Proceedings of an International Workshop.  R. Duit, F. Goldberg, H. Niederer (Editors)   March 1991  IPN 131,  ISBN 3-89088-062-2

P. H. Scott, H. M. Asoko, R. H. Driver
Children's Learning in Science Research Group University of Leeds, UK


Over the last twenty years an active research programme has been established in the area of children's conceptual understanding in science. Outcomes of this research include detailed information about children's conceptions, at various ages, in a wide range of science domains. Review articles and edited collections of papers are available which provide an overview of this field (Gilbert and Watts, 1983; Carey, 1985; Driver, Guesne and Tiberghien, 1985; West and Pines, 1985). The importance which is accorded to children's existing conceptions about natural phenomena is a common theme running through the research programme. Learning is seen in terms of conceptual development or change rather than the piecemeal accretion of new information. Various models of learning, based upon this viewpoint, have been proposed, some deriving from epistemological literatures (Posner, Strike, Hewson and Gertzog, 1982), others from cognitive psychology (Osborne and Wittrock, 1983). All of this work has strong implications for classroom practice, and approaches to teaching which acknowledge children's alternative conceptions have been researched, developed and tested. These teaching approaches involve a range of different pedagogical strategies, they draw upon various aspects of underlying theory and have tended to be tested and reported for a limited age-range of pupils.

This paper provides a review of those pedagogical strategies reported in the literature which are broadly based on a view of learning as conceptual change. As yet the alternative conceptions research programme has had limited impact on classroom practice. At Leeds we are currently examining ways in which its findings can be used to inform science teaching more generally and this review has been prepared in order to identify the range of strategies being proposed and to analyse the different assumptions on which they are based.

Shuell (1987) has suggested that 'the teacher's task is the non-trivial one of determining which learning tasks are the most appropriate for students to work on' (p. 245). This poses the central question for science educators and teachers - on what basis does the teacher make decisions regarding the selection of learning tasks and strategies? Furthermore, what guidance can the research programme on children's conceptions and conceptual change offer in response to this real and practical problem?

We suggest that there are pedagogical decisions to be made at three levels. Firstly, the teacher needs to foster a learning environment which will be supportive of conceptual change learning. Such an environment would, for example, provide opportunities for discussion and consideration of alternative viewpoints and arguments. A second level of decision-making involves the selection of teaching strategies. We see strategies in terms of overall plans which guide the sequencing of teaching within a particular topic. Finally, consideration must be given to the choice of specific learning tasks. The learning tasks fit into the framework provided by the selected strategy and must address the demands of the particular science domain under consideration.

In making decisions about appropriate teaching strategies, four factors may need to be taken into consideration:

1. Students' prior conceptions and attitudes: students' prior conceptions across a broad range of science domains are extensively documented in the literature, but consideration now needs to be given as to how this literature is to inform teaching.

2. The nature of the intended learning outcomes: learning outcomes and the logical analysis of those outcomes in science terms have traditionally provided a principal focus for planning teaching.

3. An analysis of the intellectual demands involved for learners in developing or changing their conceptions: this analysis focusses upon the nature of the intellectual journey required of the learner in moving from existing conceptions to the intended learning outcome.

4. A consideration of the possible teaching strategies which might be used in helping pupils from their existing viewpoints towards the science view.

This paper addresses the fourth of these factors in reviewing the literature on conceptual change teaching strategies. After presenting an account of various documented strategies we identify and analyse a number of theoretical questions emerging from the review. This is followed by a consideration of practical concerns in teaching for conceptual change.

Review of Strategies to Promote Conceptual Change

We have identified two main groupings of strategies to promote conceptual change. The first grouping is of strategies which are based upon cognitive conflict and the resolution of conflicting perspectives. The second grouping is of strategies which build on learners' existing ideas and extend them, through, for example, metaphor or analogy, to a new domain. Underlying these two groupings are different emphases on where the balance of responsibility for promoting conceptual change in learners may lie. Strategies which emphasise conceptual conflict and the resolution of that conflict by the learner may be seen to derive from a Piagetian view of learning in which the learner's active part in reorganising their knowledge is central. The strategies which build on learners' existing knowledge schemes, extending them to new domains, may be seen to place less emphasis on the role of accommodation by the learner and instead focus on the design of appropriate interventions by the teachers to provide "scaffolding" for new ways of thinking.

Teaching strategies based upon cognitive conflict and its resolution

Cognitive conflict has been used as the basis for developing a number of approaches to teaching for conceptual change. These approaches involve promoting situations where the student's existing ideas about some phenomenon are made explicit and are then directly challenged in order to create a state of cognitive conflict. Attempts to resolve this conflict provide the first steps to any subsequent learning. Various approaches to conflict based teaching are reviewed in the following paragraphs.

a) Discrepant events:

Nussbaum and Novick (1982 a,b) suggest a teaching sequence which draws upon the Piagetian notion of accommodation (Piaget 1964) and includes four main elements:

- initial exposure of students' preconceptions through their responses to an exposing event.

- sharpening student awareness of their own and other students' frameworks.

- creating conceptual conflict by attempting to explain a discrepant event.

- encouraging and guiding cognitive accommodation and the invention of a new conceptual model consistent with the accepted science view.

This sequence was used as a basis for teaching aspects of the structure of gases (including the notions of vacuum between particles and of particle motion) to students of age 11 to 13. In evaluating the teaching approach, the authors comment upon its success 'in creating cognitive challenge and motivation for learning' but acknowledge that the instruction, 'did not lead to the desired total conceptual change in all students.' In conclusion, they make comparisons with the history of science in suggesting that 'a major conceptual change does not occur .... through revolution, but is by nature an evolutionary process'.

b) Conflict between ideas:

Stavy and Berkovitz (1980, p.679) draw attention to two types of training by conflict. They make a distinction between where "a conflict is produced between a child's cognitive structure related to a certain physical reality and the actual physical reality" and where "a conflict is produced between two different cognitive structures related to the same reality".

They made use of the second type of conflict in developing a teaching strategy which is aimed at advancing children's understanding of the concept of temperature. In particular they explored the conflict between two different representational systems that the child uses to describe temperature: the qualitative-intuitive system and the quantitative-numerical system. The conflict training took advantage of the fact that children's qualitative knowledge regarding certain aspects of the concept of temperature is correct at a certain age and can be used to encourage them to apply their knowledge to solve a problem in numerical form. For example, at the age of 9/10 years a significant proportion of children will assert that warm water added to warm water will still produce warm water and yet they will maintain that water at 300C plus more water at 300C will produce water at 600C.

The teaching strategy used a combination of worksheets and practical work designed to make children become aware of the conflict existing within their thinking. It was tested with pupils with an average age of 10 years, who worked individually and in class groups. In conclusion the authors write, 'our findings indicate that training by conflict did improve children's understanding of the concept of temperature both in individual and in classroom training situations' (p.689). They note the particular success of the strategy in helping to develop children's understanding of the 'intensivity' of temperature.

Cosgrove and Osborne (1985), Champagne, Gunstone and Klopfer (1985) and Rowell and Dawson (1985) have developed teaching approaches which require students explicitly to resolve differences between ideas from a range of different sources (e.g., other students, the teacher, the science text). Cosgrove and Osborne (1985) have proposed a 'Generative Learning Model of Teaching' which is organised into four phases:

Preliminary Phase: teacher needs to understand the scientist's view, the children's view, his/her own view.

Focus Phase: Opportunity for pupils to explore the context of the concept, preferably within a 'real' everyday situation. Learners to engage in clarification of own views.

Challenge Phase: learners debate the pros and cons of their current views with each other and the teacher introduces the science view (where necessary).

Application Phase: opportunities for application of new ideas across a range of contexts.

The authors emphasise the point that an alternative science view may not be 'received with much enthusiasm until it can be rendered intelligible and plausible by experimentation, demonstration or reference to analogy' (p.107). They also stress the importance of the preliminary phase in preparing for teaching.

The model has been applied to teaching about current flow in electric circuits and includes, within the Challenge Phase, a 'critical test' which involves measuring the electric current on either side of a bulb. The electric current unit has been used on many occasions (11 to 14 age range) and the authors report its success in helping pupils move from a view of electricity "being used up" to adopting a view of current being conserved around the electric circuit.

They draw attention to problems of stability of new ideas with regard to both context and time and suggest that, 'where ideas are counter-intuitive and not reinforced by other learning situations, then it would appear highly desirable for further experiences with the topic to occur at some later date' (p. 122).

Champagne, Gunstone and Klopfer (1985) have proposed a dialogue-based strategy which they call Ideational Confrontation and which is specifically designed to alter students' declarative Imowledge within a particular domain (e.g., the motion of objects). It involves the following steps:

-students make explicit the' notions they use to explain, or make predictions about, a common physical situation, (e.g., the motion of a deflating balloon).

-each student develops an analysis that supports his/her predictions and presents it to the class.

-students attempt to convince each other of the validity of their ideas; discussion and argument results in each student becoming explicitly aware of his/her ideas about motion in that context.

-the instructor demonstrates the physical situation (e.g., releases the balloon) and presents a theoretical explanation using science concepts.

-further discussions allow students to compare their analyses with the scientific one.

Use of the strategy is reported with two age-groups of students: middle school and graduate teacher trainees. The authors suggest that discussion, considering the views of others, and relating a situation under consideration to other real-world phenomena are significant in promoting change of views. They also make the point that students must be motivated and that the quality of arguments improved over the course of instruction.

Rather than attempting to promote conceptual change by inducing conflict with students' prior conceptions at the beginning of a teaching sequence, Rowell and Dawson (1985) propose a strategy in which resolution between students' prior ideas and new conceptions occurs after new conceptions have been introduced. Their approach, which draws upon a perspective from the history and philosophy of science and equilibration theory (Piaget 1977), is based upon the following premises:

-a theory is only replaced by a better theory and not discarded on the basis of contradictory evidence alone.

-the construction of a better theory need not involve an immediate confrontation with the knowledge that an individual spontaneously considers relevant.

-although cognitive change involves both strategic and metastrategic knowledge (Kuhn 1983), they need not be constructed together.

The teaching approach involves six steps:

-the ideas which students consider relevant to the problem situation are established.

discussion and are retained in a 'paper memory' for subsequent consideration.

-students are told that a theory is going to be taught to them which may solve the problem and that their help will be required both in its construction and later, its evaluation against the alternatives they have proposed.

-the new theory is presented by linking it to basic knowledge already available to the class.

-students are asked to apply the new theory to problem solution, in order to indicate its construction by individuals. Written work must form a part of this procedure to provide a second paper memory for each student.

-each student compares paper memories from steps 1 and 5 and the quality of the ideas is examined. Initial examination is directed to the stimulus problems used in the tests, to which the paper memories relate. Subsequent examination is broadened to cover as many relevant situations as possible. That is, the student is involved in gaining metastrategic knowledge.

Dawson (1990) reviews the use of this approach in the context of introducing chemical change to novices.

Teaching strategies based upon the development of ideas consistent with the science point of view

In contrast to the strategies which promote conflicts and require students to resolve them, a second group of teaching strategies can be identified which builds on pupils' existing ideas. Subsequent teaching and learning involves the pupil in developing and extending these existing ideas towards the science viewpoint.

Clement et al. (1987) have developed and tested an analogical teaching strategy, within the field of mechanics, which aims to 'increase the range of application of the useful intuitions and decrease the range of application of the detrimental intuitions' (Brown and Clement, 1989, p.239). The strategy assumes that conceptual change can be encouraged by providing opportunities for students to build up qualitative-intuitive understandings of phenomena before mastering quantitative principles. Such understandings are developed by forming analogy relations between a misunderstood target case and an 'anchoring example', which draws upon intuitive knowledge held by the student The use of a bridging strategy' has been found to be useful in developing this relationship.

As described by Brown and Clement (1989) the bridging strategy consists of four steps:

- the students' misconceptions relating to the topic under consideration are made explicit by using a target question. For example, a question which draws out a misconception for a majority of introductory physics students concerns the existence of an upward force on a book resting on a table. Students typically view the table as passive and unable to exert an upward force.

- the instructor suggests a case which she/he views as analogous (such as a hand holding up a book) and which will appeal to the students' intuitions. This case is termed an 'anchoring example' or simply an 'anchor'. (M anchoring intuition is defined as being a belief held by a naive student which is roughly compatible with accepted physical theory. This belief may be articulated or tacit, Clement et al., 1987).

- the instructor asks the student to make an explicit comparison between the anchor and target cases in an attempt to establish the analogy relation.

- if the student does not accept the analogy, the instructor then attempts to find a bridging analogy' (or series of bridging analogies), conceptually intermediate between the target and anchor. In the book on table' example, such a bridging analogy might be a book resting on a spring.

Experimental use of such strategies to overcome misconceptions about static forces, frictional forces and Newton's Third Law for moving objects have been reported as producing significantly greater pre-post test gains than for control groups. Recent work (Clement, Brown and Zietsman, 1989) has involved further research on anchoring conceptions.

Stavy (1991) also reports work which aims to use students' intuitive perceptual knowledge, in this case to understand that matter is conserved on evaporation. Stavy suggests that the use of an analogical relation between the known and the unknown can help students learn new information and discard or modify misconceptions. In the study reported, students from grades 5 and 6 were divided into two groups. One group completed a task involving iodine evaporation where the gaseous iodine is visible as a coloured gas before attempting a similar task using acetone, an invisible gas. The second group used acetone first, followed by iodine. It was found that performance in the acetone task was significantly higher when it followed the iodine task. The intuitively under-stood, perceptually supported iodine task apparently served as an analogical example for the misunderstood acetone case ('the acetone ......... it's no longer there').

A rather different approach to the problem of producing conceptual change has been reported by Niedderer (1987), working with 16 to 19 year-olds. This approach is acknowledged by the author as being based in the 'New Philosophy of Science' as outlined by Brown (1977) and aims not to replace students' theories (related to everyday-life thinking) by the scientific theory but to allow them to arrive at a conscious knowledge of both and to learn scientific concepts by learning the differences between everyday-life thinking and scientific thinking, a position which has also been argued by Solomon (1983). In outline form, the strategy consists of six stages:

- Preparation: the teaching process which precedes the intervention, and which may contain tools and concepts that may be drawn on.

- Initiation: an open-ended problem is posed.

- Performance: this comprises parts of the following sequence: formulating questions or hypotheses, planning and performing experiments, making observations, theoretical discussions, formulation of findings.

- Discussion of findings: in a class forum.

- Comparison with science: class findings are compared with similar historical theories or modern ideas. Differences are stated and possible reasons for those differences are discussed.

- Reflection: students are encouraged to look back on the process of performance and to consider particular questions or difficulties which have arisen.

An illustration of this sequence is given for a teaching unit on 'force', whereby, following a preparation phase in which students learn concepts such as distance, time, velocity, acceleration, they are given the general question, 'what does acceleration depend upon?' Working in small groups, students formulate questions or hypotheses, carry out experiments and report back. The teacher then explains the power of general theories relevant to a range of circumstances and considers the formula F=ma in relation to the specific instances investigated by the groups.

The author notes that in their investigations students generally arrived at solutions to their particular problems but not at general relationships. There appeared to be some success in introducing students to fundamental ideas about the nature of scientific enquiry. The author also claims that, 'It seems plausible that this teaching strategy has started a far reaching learning process by letting students come to own results and by comparing those results systematically with the results of scientific research' (p. 365).

Theoretical Issues in Conceptual Change Teaching

Two broad approaches to teaching for conceptual change ('cognitive conflict and its resolution' and 'development of ideas') have been identified in the literature. We now consider a number of theoretical issues, some of which arise directly from the review and others which are of more general concern.

Acknowledging Students' Ideas

The fundamental principle underlying all of the approaches reported is one which stresses the importance of acknowledging the learner's existing ideas and understandings in any teaching/learning event. This process has been put into operation in various ways:

- through explicit elicitation of students' ideas in class.

- in informing selection of starting points to teaching.

- in informing curriculum design.

All of the conflict strategies (and Nieddererts work in addition) involve phases where pupils have the opportunity to make explicit, and to clarify, their own views. Differences, both between the pupil ideas and with the science view-point, are then identified. Niedderer works from the students' ideas in an alter-native way in using them as a basis for developing generalizations within the science perspective. In all of these instances, pupils' ideas are explicitly brought out into the open and used as a basis for subsequent teaching.

Clement et al. (1987) take as their starting point a target question designed to reveal pupil 'misconceptions' about some phenomenon but then shift the focus to consideration of an 'anchoring case' which the teacher views as being analogous to that phenomenon. Here we have a situation where knowledge of children's ideas and understandings is used to inform the choice of starting points to teaching; the initial teaching is not designed to respond to pupil ideas explicitly raised in class. In a similar way, Stavy (1991) uses awareness of children's understandings (their intuitive perceptual knowledge) to provide a starting point for teaching on conservation of matter.

Other researchers have used knowledge of children's conceptions to inform the design and sequencing of parts of the science curriculum. Thus Schollum, Hill and Osborne (1982), in teaching mechanics to 11 to 13 year old children, base their instruction on the conception held by many children that objects move forward because, 'there is something in them that keeps them moving.' They start by introducing the idea of momentum for this "something". Teachers using the approach report that the children, 'seem to have the ideas already.' In a similar way, Eisen and Stavy (1987) have developed a unit for teaching photosynthesis which works from material cycles in nature which they consider to represent a form of natural order for pupils.

The Nature and Role of Conflict

It could be argued, from the point of view of the teacher, that all the teaching approaches reviewed above, including those which are not considered to be based in cognitive conflict, have an element of actual or potential conflict in them. This conflict is between accepted scientific theory and those ideas which students either bring to the learning situation or might construct as a result of it. In Stavy's work, for example, knowledge of students' ideas about conservation of mass on evaporation leads the teacher to expect that the likely response to the question involving acetone evaporation will be in conflict with the science view. Such a situation, in this case, is avoided by using the iodine evaporation task first. However, while the teacher may be aware of conflict situa tions, the student may be entirely unaware. Indeed, even if the conflict is highlighted by some means, there is no guarantee that the student will recognise either its existence or its significance.

Teaching strategies which deliberately aim to use a conflict-based approach are likely, at some point, to make such conflicts between the students' ideas and the science perspective explicit (e.g., Rowell and Dawson, 1985) but may also utilise discrepancies between:

- two sets of ideas already available to the learner, e.g., Stavy and Berkovitz (1980) qualitative-intuitive and quantitative-numerical representations of temperature.

- an explanatory model held by the learner and an event which cannot be explained by this model, e.g., Nussbaum and Novick (1982a) continuous model for the structure of a gas versus evidence that a gas can be compressed.

- ideas which a student holds and the ideas of his/her classmates, e.g., Champagne et al. (1985) various student ideas about the motion of objects.

The success of any conflict-based strategy depends upon the willingness and ability of the learner to recognise and resolve the conflict. For example, different student ideas within a class, or a student expectation and a physical event, cannot be brought into conflict unless the learner is prepared to construct an understanding of such ideas and events and then attempt to relate them to each other. In addition, Dreyfus et al. (1990, p.567) make the point that, 'even meaningful conflicts are not always successful, in the sense that they do not always ensure the construction of the required knowledge and/or only of the desired knowledge.'

In the teaching strategies reviewed in this paper, four positions about the role of conflict and its resolution can be identified:

i) Strategies where the conflict must be recognised by the student in the early stages of teaching if learning is to occur.

Nussbaum and Novick (1982a) describe their rationale for introducing conflict at the beginning of the teaching Sequence in terms of the need for, 'recognition by the learner of a problem and his inability to solve it with his existing conceptions' and assume that human beings have, 'an innate need to reduce dissonance, incongruity and conflict between two cognitions' (p.186). The presence of conflict is thus seen as a motivating factor in the search for a better explanation. This is also true for Stavy and Berkovitz's (1980) work on representations of temperature.

ii) Strategies where an alternative 'way of looking' is introduced and the conflict highlighted later.

Rowell and Dawson (1983) suggest that conflict should only be invoked after pupils have had the opportunity to clarify their own ideas and the science viewpoint has been introduced. They consider that this approach is more effective and reduces threat to the learner in that, 'no challenge is raised to the child's old way of thinking until a new way is already available to him/her which can take its place' (p.124).

iii) Strategies where conflicts 'nay occur but are not seen as being essential for promoting learning.

Clement (1987) sees conflict as a potentially useful motivator for learning. He suggests that, in the case of tensions between a misconception and a correct conception in the same student, an appropriate approach is to 'draw out both ........ and play them off against one another' (p.94). Although Clement's basic strategy is one of using analogy to develop 'useful' conceptions, rather than challenging misconceptions, he notes the potential for utilising conflicts between students who hold different points of view. He suggests that, 'skillfully led classroom discussions appear to be one effective vehicle for fostering dissonance, internal motivation and conceptual restructuring' (p.94).

iv) Strategies which aim to avoid conflict for the student.

Stavy (1991), expresses the opinion that conflict strategies may cause a loss of self confidence in students and, on occasions, regression from correct to incorrect judgements. In her approach to teaching by analogy (1991) she claims that students, 'are not made aware of the conflict or of the learning process the learning takes place without the student's awareness. From the student's point of view, there were no misconceptions and no learning took place.'

The Construction of scientific conceptions

Students are not going to adopt a new conception unless they can first represent it to themselves. In other words, they must find it intelligible. Where such new conceptions originate from and how they are made intelligible to students varies across different strategies. In some strategies, for example those emphasising conflict, there tends to be an assumption that alternative conceptions will be generated from among the students; that if some students suggest viable alternatives, these are then available in a 'pool' of ideas to be consider ed, both through discussion and experiment, by all students. Such a strategy may be effective in some cases where viable alternatives are readily put for-ward by students.

This is not always the case. Some researchers in the field acknowledge that the teacher may need to intervene to suggest the 'accepted view' alongside other students' suggestions. This draws attention to a fundamental theoretical point about the process of construction of scientific conceptions. We would argue that scientific conceptions are not simply individual constructions which are developed to make sense of experience. They are 'ways of seeing' which have been developed within the social community of science. In this sense they may need to be transmitted through the culture of science rather than be 'discovered' from personal experience. Helping students in constructing scientific conceptions thus involves a process of induction into a scientist's culture and the teacher has an important role to play in guiding that induction.

In reviewing the different strategies for conceptual change we identify different ways in which this process of construction of science conceptions is facilitated. At one extreme there are the strategies where students, through brainstorming and reflection, are encouraged to generate more viable conceptions themselves. The way the teacher then encourages some ideas and discourages others may give differential status to certain ideas (see Edwards and Mercer, 1987) and in this indirect way induct students into a scientific perspective.

Other strategies are much more direct in supporting the knowledge construction process. The careful and disciplined choice of analogies to support the construction of new conceptions (as in the work of Clement et al., 1987) is a case in point. The strategy proposed by Rowell and Dawson explicitly gives time for the construction of a new model but without saying how. In other strategies, (for example, that described by Niedderer) the teacher, drawing on the range of experiences provided for by students, offers an integrative 'way of seeing' and negotiates its use with students. We see this process as one in which the concepts and theories of the scientific community are being made available to students and their meanings are being negotiated with them. Work undertaken in the tradition of Vygotsky emphasises this process whereby learners are supported in their development of new capabilities through an explicit phase of 'scaffolding'.

In our own work in mechanics at needs (Twigger et al., 1991) we are explicitly introducing representational devices (such as arrows for force vectors) at appropriate points in the students' work with computer microworlds. We do not expect students to 'discover' such devices for themselves, although they do

need opportunities to use such notation in thinking about and interpreting a range of situations so that it becomes appropriated by them.

A problematic feature of the knowledge construction process is that in some domains the construction of a scientific model requires the establishment of a quite extensive series of concepts and inter-relationships. This is the case in students' construction of Newtonian mechanics (with the links between concepts such as force, velocity, acceleration, momentum), and current electricity (where concepts of current, voltage resistance, power need to be established and related). In these cases it takes time to establish a new theoretical model and during this time students may, even without being aware of it, be trying to interpret experiences they are given in terms of their prior ideas in that domain. Although they can only be constructed in bits, such complex ways of seeing may only make sense as a whole - thus presenting a complex problem for pedagogy.

The Evaluation of Scientific Conceptions

Students may be able to construct a representation of a scientific idea. It may be intelligible to them. However, there is still a further aspect of the process of conceptual change and that is what status the learner gives the scientific conception. There are various possibilities here. The conception may be under-stood (i.e. the student can create an internal representation of it) but the student does not believe that it represents the way the world is. We have experience of secondary students responding to the particle theory in this way.

Rather than evaluating a new conception ontologically, i.e. as representing how things are in the world, students may evaluate them in a utilitarian way. In this case the conception is evaluated in terms of its usefulness in particular contexts (both social and phenomenological). What may be important here is the point that Solomon (1983) made that, rather than instruction being orientated to changing students' conceptions, what is necessary is helping them to appreciate the appropriateness of particular 'ways of seeing' to particular contexts.

A further way in which a new conception may be evaluated is in terms of its generality. Here the feature is not whether a new conception better 'fits' experience nor whether in particular instances it is judged to be more useful or appropriate. This way of evaluating introduces an epistemological criterion - that of consistency. Although there is probably some internal cognitive pressure towards consistent as opposed to inconsistent mental representations, studies of children's and adults' reasoning indicates that this is far from a powerful influence. Inducting students Into science however involves adopting this criterion as a major one. Just as we argued earlier that teaching in science needs to involve the explicit "scaffolding" of scientific conceptions so we would argue that it also needs to introduce explicitly the epistemological assumptions underlying the 'language game' of science. It is a criterion such as parsimony that also introduces a discontinuity between students' everyday conceptions and scientific conceptions. It is not only that their everyday conceptions are in-commensurable with those of science but also that the criteria for evaluation differs significantly in emphasis. Whereas for everyday conceptions ontological and perhaps utilitarian criteria probably dominate, in evaluating scientific conceptions parsimony has much greater status.

There are a number of comments to be made in the light of this about specific conceptual change strategies. Many include opportunities for students to evaluate competing conceptions, either explicitly as in the strategy proposed by Rowell and Dawson (1985) or by discussion and the general interplay of ideas as in the strategies suggested by Champagne et al. (1985), Cosgrove and Osborne (1985) and Niedderer (1987).

Such an evaluation process involves a comparison of two or more competing conceptions on the basis of Various criteria including internal consistency and their extension. In practical terms what this means is that in order to evaluate a scientific conception, students need to have opportunities to consider not just single well chosen phenomena but a wide range of instances. An important part of adopting a scientific conception is appreciating the range of situations to which it relates.

Practical Concerns in Conceptual Change Teaching

Students and their teachers are clearly the central players in all of the class-room activity which has been described in this paper. What then are the special demands which 'conceptual change teaching' might place upon them?

Demands Upon Students

One feature common to many of the teaching approaches reviewed is the extent to which students are involved in discussion, both in small groups and in whole class situations. Discussion provides the means by which students become more aware of their own, and other students', ideas and understandings. It makes various demands upon students including the need to listen to, construct meaning for and evaluate, the points of view of others. Having considered these different points of view, the student (especially in the conflict based approaches) is then often confronted with a further perspective which may carry the additional authority of being introduced by the teacher. A range of perspectives is thus placed on offer and the student is expected to consider the relative merits of them all.

A fundamental point which might be made about such a learning environment, in which a plurality of viewpoints is encouraged, is that it follows from particular views of science and of learning and these views may or may not be shared by the students. This can lead to problems. If, for example, students have views of learning which are essentially transmissive in nature and they adhere to positivist views of science then asking them to consider their own, and other, perspectives about a phenomenon may not appear sensible to them. This dilemma is encapsulated in a comment made during a practical science lesson; a fourteen year-old girl, on being asked for her ideas about a phenomenon, replied, 'why ask me? Just tell us the right answer.' The girl's response is perfectly valid in the context of her own thinking about science and learning. If students are to engage usefully in the kinds of approaches to teaching and learning which are described in this paper then they must be introduced to, and encouraged to reflect upon, the underlying assumptions.

Through these teaching approaches the student is clearly placed in an intellectually challenging position and that, of course, is the point of the exercise. Dreyfus, Jungwirth and Eliovitch (1990), however, remind us that students bring more than alternative conceptions with them to science lessons, they also import attitudes which will influence subsequent learning. In particular, they found that 'bright successful students reacted enthusiastically to cognitive conflicts.' They liked the 'flabbergasting effect' of the method and the confrontation with new problems.' In contrast, 'unsuccessful students .... have been shown to develop negative self-images, negative attitudes towards school and school tasks and high levels of anxiety.' As a result, 'they tried to avoid the conflicts. They were most characteristically apologetical when confronted with a conflict which, to them, seemed to represent just another failure' (pp. 565-566). Stavy (1991) makes a similar point when she argues that conflict training can, result in students' loss of self-confidence and can sometimes cause regression.'

Whatever the strategy that is adopted, however, a central feature from the student's point of view is that knowledge is not provided for them 'ready made'. They need to take ultimate responsibility for making sense of learning activities.

Demands upon Teachers

All of the approaches reported require the teacher to be responsive to the ideas and understandings of pupils. This responsivity is achieved in different ways with different approaches and the associated demands upon teachers vary accordingly.

In some situations the teacher is required to take on a neutral 'consultative role.' Aspects of this role may be unfamiliar to the teacher in, for example, acting as a sounding heard for ideas and declining to express opinions, or sup-porting students as they develop their own programmes of questions for future investigation.

At other times the teacher will need to respond directly to student ideas in helping them towards the science viewpoint. In such situations, knowledge of the science domain in which the teaching is located, of the conceptions that students tend to use in that domain and of the conceptual pathways they tend to follow once teaching is under way are all essential. Knowledge of conceptual pathways provides insights into the dynamic processes and routes of class-room learning within any particular domain. By their very nature conceptual pathways can not be defined, for individual learners, in advance of teaching (through, for example, a theoretical comparison of a student's conceptual starting points and the intended learning outcomes). Knowledge of common conceptual pathways can only be gained through practice and the development of such knowledge is likely to contribute greatly to the expertise and confidence of the teacher.

A further and fundamental demand upon the teacher lies in creating a classroom environment within which students feel confident and able to express and discuss their views openly. Such an environment can only be created through the teacher both being sensitive to students' needs, feelings and ideas and being an effective manager of class groups.

We would contend that, for many teachers, demands such as these are likely to represent a significant change from existing practice. The teacher is required to:

- be aware of students' ideas and understandings relating to the topic under consideration.

- be aware of likely conceptual pathways for that topic.

- be sensitive to students' progress in learning.

- be able to generate learning tasks to support and encourage that progress in learning.

- be sufficiently confident in his/her own understanding of the subject topic to be able to appreciate, and respond to, differing points of view.

- be able to organise and manage a classroom which will allow for all of this to happen.

Final Comments

This paper has reviewed a range of teaching strategies designed to promote conceptual change in students. The principal aim of all the approaches is to help students towards a more scientific view of the world. There are differences, however, in how this general aim is realised. Students have variously been encouraged: to exchange their existing ideas for entirely new conceptions (Nussbaum and Novick, 1982); extend or develop existing views and apply them in new situations (Brown and Clement, 1989); develop a scientific understanding which may be held in parallel with existing notions (Niedderer, 1987); recognise the appropriateness and/or applicability of models in different situations (Stavy and Berkovitz, 1980).

It is clear that different strategies make differing cognitive demands upon students. This point brings us back to the fundamental issue raised in the introduction to this article, that of selection of teaching strategies. A comparison of a student's existing conceptions with intended learning outcomes provides an overview of the desired conceptual change and gives some indication of the extent and nature of the intellectual journey which the learner must make. Additionally, any strategy which is selected to promote that change will introduce, for the learner, its own cognitive demands and these need to be taken into account during planning along with other relevant factors.


Brown, H. J. (1977) Perception. Theory and Commitment - The New philosophy of Science, Precedent, Chicago

Brown, D. E. and Clement, J. (1989) Overcoming misconceptions by analogical reasoning: abstract transfer versus explanatory model construction Instructional Science 18: 237-261

Carey, S. (1985) Conceptual Change in Childhood, MIT Press, Massachusetts

Champagne, A. B., Gunstone, R. F. and Klopfer, L. E. (1985) Effecting changes in cognitive structures among physics students in Cognitive Structure and Conceptual Change, West L. and Pines A. (Eds.). Academic Press

Clement, J. et al. (1987) Overcoming students' misconceptions in physics: the role of anchoring intuitions and analogical validity. Proceedings of the Second International Seminar. Misconceptions and Educational Strategies in Science and Mathematics. 3: 84-97

Clement, J., Brown. 0. and Zietsman, A. (1989) Not all preconceptions are misconceptions: finding 'anchoring conceptions' for grounding instruction on students' intuitions. International Journal of Science Education 11(5): 554-565

Cosgrove, M. and Osborne, R. (1985) Lesson Frameworks for Changing Children's Ideas. In: Learning in Science: The implications of children's science, Osborne R. and Freyberg P. Heinemann

Dawson, C. (1990) Dealing with students' intuitive conceptions: some research implications for chemistry teachers. Chemeda: Australian Journal of ChemIcal Education

Dreyfus, A., Jungwirth, E. and Eliovitch, R. (1990) Applying the 'Cognitive Conflict' strategy for conceptual change - some implications, difficulties and problems. Science Education 74 (5): 555-569

Driver, R., Guesne, F. and Tiberghien, A. (1985) Children's Ideas in Science, Open University Press

Edwards, 0. and Mercer, N (1987) Common Knowledge. Methuen

Eisen, Y. and Stavy, R. (1987) A different approach to the teaching of photosynthesis. Proceedings of the international seminar on adolescent development and school science. Kings College, London

Gilbert, J. K. and Watts, 0. M. C1983) Concepts, misconceptions and alternative conceptions: changing perspectives in science education. Studies in Science Education lO: 61-98

Kuhn, 0. (1983) On the dual executive and its significance in the development of developmental psychology. Contributions to Human Development 8: 81-110

Niedderer, H. (1987) A teaching strategy based on students' alternative frameworks -theoretical conceptions and examples. In: Proceedings of the Second International Seminar. Misconceptions and Educational Strategies in Science and Mathematics 2: 360-367 Cornell University

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

Nussbaum, J. and Novick, S. (1982b) A study of conceptual change in the classroom. Paper presented at the annual meeting of the National Association for Research in Science Teaching, Lake Geneva, Chicago

Osborne, R. J. and Wittrock, M. C. (1983) Learning science; a generative process. Science Education 67(4): 489-5 OS

Piaget, J. (1964) Development and Learning. Journal of Research in Science Teaching 2:176-186

Piaget, J. (1977) The development of thought. Translated by A. Rosin, Blackwell, Oxford

Posner, C. J., Strike, K. A., Hewson, P. W. and Gertzog, W. A. (1982) Accommodation of a scientific conception: toward a theory of conceptual change. Science Education 66(2): 211-227

Rowell, J. A. and Dawson, C. J. (1983) Laboratory counter-examples and the growth of understanding in science. European Journal of Science Education 5 (2): 203-215

Rowell, J. A. and Dawson, C. J. (1985) Equilibration1 conflict and instruction: A new class-oriented perspective. European Journal of Science Education 4 (4): 331-344

Schollum, B. W., Hill, C. and Osborne, R. (1982) Teaching about force. Working Paper No. 34, Learning in Science Project. Hamilton, New Zealand; SERU, University of Waikato

Shuell, T. J. (1987) Cognitive psychology and conceptual change: Implications for teaching science. Science Education 71 (2): 239-250

Solomon, J. (1983) Learning about energy: how pupils think in two domains. European Journal of Science Education 5(1): 49-59

Stavy, R. (1991) Using analogy to overcome misconceptions about conservation of matter. Journal of Research in Science Teaching 28 (4): 305-313

Stavy, R. and Berkovits, B. (1980) Cognitive conflict as a basis for teaching quantitative aspects of the concept of temperature. Science Education 64: 679-692

Twigger, D. et al. (1991) The 'Conceptual Change in science1 project. Journal of Computer Assisted Learning (in press).

West, L. and Pines, A. (Eds.) (1985) Cognitive structure and conceptual change. Academic Press


Reprinted by permission from:  Research in Physics Learning: Theoretical Issues and Empirical Studies.  Proceedings of an International Workshop.  R. Duit, F. Goldberg, H. Niederer (Editors)   March 1991
IPN 131,  ISBN 3-89088-062-2

Section C5,  Teaching for conceptual change: a review of strategies  from: Connecting Research in Physics Education with Teacher Education
An I.C.P.E. Book © International Commission on Physics Education 1997,1998
All rights reserved under International and Pan-American Copyright Conventions
Return to the Table of Contents