Cultural Studies of Science Education

, 3:397

Discussion, debate and dialog: changing minds about conceptual change research in science education


    • Science & Technology Education GroupKing’s College London

DOI: 10.1007/s11422-008-9093-1

Cite this article as:
Dillon, J. Cult Stud of Sci Educ (2008) 3: 397. doi:10.1007/s11422-008-9093-1


This paper provides a critical commentary on a suite of eight papers, which focus on conceptual change research in science education. Responses by Mercer, Smardon and Wells to a paper by Treagust and Duit are observed to reflect the backgrounds of the three authors with Wells focusing on issues of ontology and the affective domain. Mercer and Smardon focus on issues of identity and the role of dialog. Hewson’s, Vosniadou’s and Tiberghien’s responses to Roth, Lee and Hwang offer robust critique of what appear to be exploratory ideas. To what extent the authors of the response papers enter into dialog with the papers is discussed. How far research into learning in science has progressed since the 1980s is examined.


Conceptual changeTeachingLearningDialogModeling

On framing the review

Our postgraduate pre-service course, aimed at preparing high school science teachers, covers a lot of ground in a short space of time. By government diktat, trainees must spend at least 24 weeks in school or, at least, not in college. To add to the challenge that organising such a course entails, we have decided that this is the year, in line with broader European standardization of qualifications, that we shall make the course more Masters-level than postgraduate certificate level. What this means in practice is a move towards students doing more reading and writing longer assignments that involve Masters-level reflection and analysis.

After you take off study days, induction, tutorials, general educational issues sessions and the like, we’re left with 31 two-hour workshops to teach some science education theory to mixed groups of biologists, chemists and physicists and relate it to their classroom practice. By coincidence, the last session that I organized and taught before reading the eight papers, was on “Vygotsky and Dialogic Teaching.” In reading the papers, I wondered to what extent I could use them as a resource for myself, for colleagues and for our preservice teachers.

To give some indication of the perspective that we take on theories of teaching and learning, I should explain that prior to the session, the students had been asked to read “Social interaction as a means of constructing learning: the impact of Lev Vygotsky’s ideas on teaching and learning” a research summary produced by the General Teaching Council for England (GTCE 2003). The students would have heard about Vygotsky earlier in the course. His work was mentioned in a previous session on “Piaget and Constructivism.” Secondly, Jill Hohenstein, an ex-UC Santa Cruz postdoctoral researcher, now on our faculty, mentioned him during her Educational Issues talk on “Learning.” Also, in a chapter in the course reader, Becoming a Teacher, in which Jill and Heather King, another colleague, explain, inter alia, what is meant by sociocultural theory:

Social-constructivism has been linked to the work of Vygotsky (1978), Bruner (1966) and others. This theory suggests that ideas are first encountered by learners in the social environment, mostly in the form of language. After some experience with these ideas, they become incorporated into children’s habitual knowledge and become “second nature.” Knowledgeable others in the environment can guide learning experiences by supporting children’s experiences through questions and stimulating commentary. Such support has been termed “scaffolding.”

After a general discussion about the main points in the GTC paper, the students discussed questions, based on sections in the paper, including:
  • How can Vygotsky’s work help teachers think about and plan their teaching—the ZPD?

  • How do children construct meaning through social learning?

  • Can the same thinking processes be used by pupils in different contexts?

The dialogic teaching part of the session focuses on the work of Robin Alexander (2006), who has written persuasively about the subject and the work of Phil Scott, Eduardo Mortimer and colleagues who have looked at the tension between authoritative and dialogic discourse in high school science lessons (Scott et al. 2006). All this in two hours, of course. However, some students will, we assume, be able to reflect on the issues raised during the session and during subsequent workshops on “CASE” (the Cognitive Acceleration through Science Education project), on “Ideas and Evidence” and on “Argumentation,” which will draw on Piaget, Vygotsky and dialogic teaching. One of the seven possible titles for one of the major pieces of coursework later in the year is “Thinking,” so students that are interested in the topic can develop a good understanding of the issues and the relevance of the research and theory to classroom practice.

So, I came to these papers as someone with 10 years experience as a high school science teacher, a background of researching children’s ideas about chemical phenomena, with experience of researching cognitive acceleration with elementary students, and a conviction that science teacher preparation should include developing an understanding of conceptual change, constructivist science teaching, dialogic teaching and inquiry-based science inside and outside the classroom. I begin with David Treagust and Reinders Duit’s paper and the three responses to it by Mercer, Smardon and Wells.

Mercer’s, Smardon’s, and Wells’ responses to Treagust and Duit

As you might expect, the three different backgrounds of the respondents evidence themselves directly and indirectly in their responses. Smardon, a sociologist, begins the “delicate task” of critically reviewing the paper positively by noting that: “All of this research on learning has something to teach sociologists. Some science education researchers may be surprised to learn that the socio-cultural perspective is virtually unheard of in mainstream American sociology.” Tantalizingly, Smardon offers little by way of explanation for the last statement. Are we barking up the wrong tree? Are US sociologists in other silos? Reassuringly, Smardon adds “an itemized list of reasons why sociologists ought to take the time to learn about advances in socio-cultural theory, is not within the purview of my task at hand.” Still, it would be interesting to know what are the main reasons.

Mercer, a “sociocultural researcher,” begins by acknowledging that the paper offered “a much broader and open-minded discussion” than he had expected. Mercer does “not consider a sociocultural account of conceptual change, development and learning incompatible with a recognition of the cognitive aspects of those processes.” This position seems to be echoed by Smardon who, however, refrains “from commenting extensively on the potential for a reconciliation between the sociocultural approach and the cognitive approach.” She notes that others, including Mercer “have commented that both cognitive and socio-cultural research on conceptual change ought to deal with both cognitive and social mechanisms (Mercer 2007).” Mercer describes his contribution as a “collaborative striving for what Treagust and Duit call an adequate ‘multiperspective’ on this important educational topic.” The question is, do the three responses to the Treagust and Duit paper move us towards a “multiperspective” or do we end up with three views of different things?


Wells’ response to the Treagust and Duit paper is broadly philosophical in its tone. Wells asks: “Is the attempt to change students’ concepts the best way to proceed?” Somewhat confusingly, Wells writes: “What is at issue, as I see it, is the ontological status of concepts (or ‘conceptions).” As I see it, Treagust and Duit clearly delineate between concepts and conceptions when they write, “research has shown that students come to science classes with pre-instructional conceptions and ideas about the phenomena and concepts to be learned that are not in harmony with science views.” This distinction is reinforced when Treagust and Duit write: “research has shown that many teachers hold conceptions of science concepts and processes that are not in accordance with the science view and often are similar to students’ pre-instructional conceptions.” My feeling is that when Wells writes “it is useful to make a clear distinction between individuals’ knowing and what, within the scientific community, is taken to be known” that his position is closer to that of Treagust and Duit than he realises.

Nevertheless, Wells’ ontological examination opens up a critical perspective on what is involved in learning and knowing. Students, Wells notes, “are treated as having concepts” which Treagust, Duit and many of the researchers they quote, seem to portray as “relatively stable mental entities” which students possess in the same way that they possess books and pens. To Wells, concepts are “collaboratively produced constructs that constitute the realm of “what is known,” adding, “they are symbolic formulations that can be located in texts of various kinds whose authority is independent of the particular individuals who constructed them. However, although accepted as authoritative at a particular moment, concepts are not unassailable.” Wells refers to Polyani who pointed out that we don’t actually have first-hand evidence for much of what we think we know about the world. Rather, we know only that the scientific community believes that a small number of people’s first-hand evidence points to something being true and then we have to decide whether or not to believe them. Thus few of us have actually carried out research into the relationship between smoking and lung cancer but most of us “know” that most scientists “know” that there is one.

An example might help. I could tell you that I’m right-handed. If you didn’t already know that then I’ve just added to your knowledge, if, and only if, you believe me. So, belief plays a key role in knowing and so, as a consequence your attitudes towards whoever is offering you some knowledge is critical. Of course, your attitudes towards the veracity of a concept is colored by a range of cultural, historical, political and social factors. There is a reason why so many people in the world believe that the earth is about 5,000 years old and it’s not because they’re unintelligent, it’s because they choose not to believe the scientific consensus. If I now tell you I’m actually left-handed then, you have to choose whether or not to believe this new piece of knowledge. If you accept it as true then you’ve just undergone a brief moment of cognitive conflict and undergone conceptual change. I don’t need to be there to teach you something and, indeed, dead folk can still teach. Is it possible to believe that I’m left-handed and that I’m right handed simultaneously? Well, it’s possible to believe that both options are possible but you can’t believe that they’re both true at the same time. Now, depending on whether you think I’m right or left-handed, you’re either right or wrong. You either have the correct concept or you don’t. The trouble is I’m not going to tell you so you’ll probably never “know” for sure. One of the issues here is that probably all that’s happened in your head is that a couple of links have been made—nothing’s actually gone into your brain—as Wells argues: “Concepts are not mental objects, nor do they belong to individuals.”

Is “misconception” misconceived?

The thorny issue of the use of the word “misconceptions” gets another going over from Wells. I don’t know when this issue was first raised, but the debate doesn’t seem to have moved on in recent years. One of my colleagues sent round an email only the other day that uses the term in the context of a task in which our preservice teachers need to assess the effectiveness of a sequence of lessons:

If [students choose to write] tests, then [we] need to see if test gives them info on misconceptions, pre-knowledge of terms, understanding, etc. and decide whether to ask to include more questions in test or replace some to make it more useful.

Wells writes that: “it is inappropriate to categorize students’ use of particular concepts as correct and others as misconceptions independent of the activity and problem to be solved (Wertsch 1991)“ and I have to agree with him. But how do I convince my colleague for whom misconceptions is ingrained and part of the warp and weft of her daily language?

Cognitive and affective

Moving now to the cognitive and affective “domains.” Wells notes that: “Treagust and Duit briefly discuss the influence of the ‘affective domain,’ remarking that the influence of students’ interests and motivation has had ‘limited attention from CC researchers.’” The key word here is “briefly,” indicating that the “affective domain” is not a high priority with cold conceptual change researchers. Smardon, is more elaborate in her observations on this point, noting that whereas she thinks that: “descriptions of relationships, membership and social interaction are key to understanding the role of collective emotion as well as individual affect,” it seems to her that Treagust and Duit assume that: “The notion of social context is essentially a theoretical appendage to the project and is therefore secondary to data analysis.” Her evidence for this point is that because they nominate formal interviews as the “best way to itemize a student’s conceptual change,” Treagust and Duit thus privilege “individual cognition theoretically.”

All three respondents examine the empirical data provided by Treagust and Duit. In particular, they look at the extracts from interviews with Dana and Jane. What they see, from their different perspectives, is different, though related. Although, as Smardon writes, Treagust and Duit “note that different interpretations of the data presented on Dana’s conceptual change process are possible,” she rather uncharitably dismisses this acknowledgement as “somewhat cursory” which is pretty close to saying that they ignore other interpretations.

Wells’ interpretation of the differences between Jane and Dana’s responses is that Dana had been taught more effectively, she had “potentially, a more complex set of experiences to draw on than did Jane.” Secondly, Wells hypothesizes that: “the interview was experienced quite differently by the two girls” and in Dana’s case “when she initially failed to provide an explanation, the interviewer was much more interactive, prompting Dana to recall the analogy of the wheels and then co-constructing the point of the analogy with her.” Wells expands on this idea to argue: “the interviewer changed the affective tone of the interview and, by working with her, increased her engagement with the problem.” This point is important and highly convincing given the data that Treagust and Duit provide. Wells expands on his position:

These differences, and particularly the different nature of the two interviews, strongly suggest that a student’s affective state – her interest as well as her self-confidence and comfort in the interview situation – makes a significant contribution to her mode of engagement with a task and thus with the possibility of learning from the experience.

Mercer’s view is somewhat different. To him it’s all about the process of dialogue, “Dana was led through a classic piece of Socratic dialogue.” Smardon is more in tune with Wells when she writes, in tones reminiscent of Carrie Bradshaw in “Sex and the City,” “Do emotions block learning or facilitate it?” Answering her own question, using the Dana interview, Smardon notes: “Dana is initially described as, ‘unenthusiastic.’ After an analogy is introduced the authors describe her as enthusiastic and talkative.” Smardon, echoes Wells’ point when she writes:

While [Dana] initially says she doesn’t know the answer she later offers an answer that pleases her interviewer. But the reader is left wondering … What is the relationship between affective change and conceptual capture for Dana? Are we meant to surmise that analogies trigger emotional reactions? Could it be that the interviewer smiles at Dana thus increasing her sense of confidence and enthusiasm through a sense of social membership? Could the interviewer be backchanneling (perhaps nodding) thus giving Dana a sign of deference that makes her feel more powerful? The authors do not provide these details nor do they provide evidence that they have considered alternative interpretations.

Reinforcing the point, and drawing on Vygotsky’s insistence of the interrelationship between cognition and affect, Wells argues that: “In his view, tests (or interviews) that are impersonal and non-interactive provide a very limited estimate of a person’s understanding.” Treagust and Duit are thus left damned by the vagueness “in terms of how affect is related to conceptual change.” Smardon is unable to detect from Treagust and Duit’s paper how affect influences conceptual change at all. Affect is indeed a rather vague term compared with cognition. In retrospect, Treagust and Duit might wish they had spent more time clarifying how affect is related to conceptual change.

One way forward for researchers comes in Smardon’s comment that she prefers to talk about “emotions or moods” and, in a footnote, writes: “Rather affect and motivation are theorized as essentially individualistic traits rather than collective accomplishments. Affect is measured in terms of goals, values, self-efficacy, and control beliefs. These variables are essentially individualistic as well as tending to be, ironically, rather cognitive, cold, static and disembodied in nature.”

The role of dialog

Despite thinking that Treagust and Duit would end up at the shared destination Mercer notes that: “conceptual change cannot be understood without considering the role of dialogue,” in the end, they never deliver, lost in cognition. Treagust and Duit are criticized for maintaining “a conception of conceptual change which does not recognize the dynamic motor of dialogue.” Using their own data to illustrate his point, he continues, “They acknowledge that conceptual understanding may be expressed through talk, but not (despite the illustrative evidence of their own data) that talk can affect conceptual understanding.” Worse still, “Even their ‘multiperspective’ does not include any consideration of the social, rhetorical processes of thinking collectively, which now constitutes a multi-disciplinary field of study (as discussed in Mercer 2000).” It’s not as if tools don’t exist to investigate such processes “systematically and rigorously.”

Wells’ reference to Lemke’s comment that:

fluency in science requires practice at speaking, not just listening. It is when we have to put words together and make sense, when we have to formulate questions, argue, reason and generalize, that we learn the thematics of science. (2001, p. 24)

reminded me of the saying “I know what I think when I hear myself say it.” Many of the examples of interviews with students about phenomena seem to involve lots of puzzling out aloud, lots of “erms” and “umms” as people listen to their own voices and realize their own confusions and you need someone else to be listening (and asking) in order to go through that process. Indeed, a characteristic human response when faced with a difficult problem, even when you’re on your own, is to verbalize the problem, “Oh, what’s that?”— it’s our brain’s way of solving problems, to generate dialog with ourselves.

So, what might teaching for learning look like?

I begin this section by pointing out that classrooms and laboratories can be dangerous places for students because of the nature of adolescence and because of the structural and cultural forces that enact themselves during schooling. As Smardon points out, “In some communities students may be taking a much larger emotional risk than scientists typically do and this is a matter of both status and power.”

In terms of the relationship between conceptual change researchers and science teachers, Smardon describes the overall tone of Treagust and Duit’s paper and many other conceptual change papers as being “characterized by an exasperated tone.” She hypothesises that this exasperation “is focused on a deep disappointment with the degree to which teachers have ignored the lessons offered by conceptual change literature.” Smardon, in turn, expresses dissatisfaction with this perspective, attributing the lack of engagement to time constraints/lack of access and the literature failing to address teachers’ priorities (such as classroom management), and she makes the important point that: “theories of learning that do not acknowledge the emotional drain associated with power struggles do not address the lived experience of teachers.”

It is when Treagust and Duit turn to the issue of disseminating useful advice to teachers that Smardon is able to identify “the wedge issue” that “drives socio-cultural and cognitive approaches to conceptual change apart” and that is “the way in which context is defined” (or, rather, conceptualized). Smardon is perplexed by Treagust and Duit’s statement that:

The ability to select intelligible, plausible and fruitful representations or conceptions for a specific context is itself a measure of expertise; however researchers need to be aware that apparent conceptual changes may in fact be context-driven choices rather than conceptual status changes. In learning settings, Mortimer (1995) proposed the use of conceptual profiles to help differentiate conceptual changes from contextual choices.

The phrase “context-driven” is problematic and she wonders why a context-driven choice would not represent authentic conceptual change. Refining the point, Smardon notes that: “the authors appear to accept without question the notion that social context is a variable that can be controlled for, thus isolating true cognitive conceptual change.” She is “not denying the reality of conceptual change but rather challenging the idea that researchers can control for a variable that they do not fully understand.”

The complexity that sociocultural researchers perceive is encapsulated by Wells who argues that identifying the upper and lower limits of a child’s ZPD is a type of formative assessment which is “an integral aspect of instruction, which itself is seen as involving collaboration between the learner and the teacher or peer(s) that involves affective and interpersonal as well as cognitive dimensions.”

The aims of education

Wells, making another philosophical point, notes that Treagust and Duit do not “consider the possibility that the attempt to bring about conceptual change may be, in itself, an inappropriate way of working toward the achievement of the PISA goals of scientific literacy that they quote.” Wells, like Mercer, sees Treagust and Duit as heading off in a different direction to himself. Wells’ destination, which he expands on in depth in his 1999 book, Dialogic Inquiry: Towards a Sociocultural Practice and Theory of Education, and which begins with the delightful story of the boy, his mother and the baking tin, is that:

rather than thinking of students as possessing concepts that need to be changed, it is more appropriate to think of them as learning to use the powerful scientific concepts that have been developed by others, and which form part of the cultural resources that are already available for their appropriation, in tackling problems that they find interesting and relevant.

The issue here is what does Wells think “learning to use” looks like? At this point, the classroom hoves into view—Wells quotes Brown et al. who argue that a teacher’s job is to devise learning environments:

that enable group participation and dialogic interaction, which support reflection, argumentation and refutation …. Students operate as researchers who are free to select a topic of inquiry, free to do research on whatever they like within the confines on the targeted topic for their grade. (Brown et al. 1996, p. 159)

Wells argues that this freedom of choice is critical if students are to be enabled “to feel a sense of ownership and self-direction in their work.” Ownership and self-direction are important because they fuel students’ “motivation to work at achieving understanding.” Wells lays his cards on the table when he writes, “nothing generates dialogue as effectively as students having results to report from their own inquiries.” Teachers of English and the social sciences might disagree with that point and it’s a relatively untested hypothesis. And I still think Wells isn’t 100% clear about where he stands on how students develop their scientific understanding.

When Wells relates the work of Herrenkohl and Guerra (1998) “in which different students in a grade four class were assigned the roles corresponding to [Hatano and Inagaki’s three discourse practices—clarification, disputation and coordinating evidence with theory]” noting that by the end “students were able to manage them without assistance,” I was reminded of work that I carried out with Jenny Frost ten years ago in which we allocated roles to group members carrying out investigations. Our assumption was that pupils would gain more from practical work in science if teachers could encourage more thinking as well as doing. We set out to give each student in a group of three a distinct role, such as monitoring performance and questioning (things that a teacher normally does). As with Herrenkohl and Guerra, we found that: “combining guidance on the social as well as intellectual level proved to be a more effective way to encourage student engagement than providing the cognitive piece alone” (p. 467).

Without directly mentioning Lave and Wenger, Wells notes that: “[a] further advantage of adopting a community of learners/inquiry approach, is the greater opportunity that this provides for “multiple zones of proximal development” (Brown et al. 1996, p. 161).” Wells moves on to bring in the notion of identity: “Learning is also centrally involved in the construction of personal identity.” He argues that:

Since “science” is rapidly changing humans’ relationship with other people as well as the planet that supports us, it is essential that the learning and teaching of science look beyond the classroom learning of scientific concepts to the uses outside the classroom to which students will put the knowledgeable skills they are mastering. Science cannot be separated from the other activities in which people engage or from the ethical, ecological and political issues that the use of scientific knowledge inevitably raises. Nothing less than this sort of engagement in scientific inquiry, I would suggest, will make possible the achievement of the PISA goals of enabling students “to understand and help to make decisions about the natural world and the changes made to it through human activity.”

Recently, Jonathan Osborne and I have argued that:

Today, many of the political and moral dilemmas confronting society are posed by the advance of science and technology and require a solution which, whilst rooted in science and technology, involve a combination of the assessment of risk and uncertainty, a consideration of the economic benefits and values, and some understanding of both the strengths and limits of science […] To understand the role of science in such deliberations, all students, including future scientists, need to be educated to be critical consumers of scientific knowledge. Improving the public’s ability to engage with such socio-scientific issues requires, therefore, not only a knowledge of the content of science but also a knowledge of “how science works” – an element which should be an essential component of any school science curriculum. (Osborne and Dillon 2008, p. 8).

Our point, which fits well with Wells’ position, is that students need to engage with scientific ideas and with the ways in which scientists build knowledge.

For many years, Piagetian ideas were seen as being outdated and, by some people, just plain wrong. That situation has clearly changed and the value of both Piagetian ideas and Vygotskian ideas now seems to be broadly accepted, as Mercer points out:

There is thus a sizeable of research on the effects of the expression of different conceptions amongst students on their learning and conceptual understanding in science, stimulated by the Piagetian approach, and producing broadly consistent results. The results are also consistent with a Vygotskian/sociocultural perspective (as explicated by Wertsch 1991), as change seems to depend upon activity in the “social plane” which permits the expression of contrasting perspectives.

Mercer explains why this should be the case:

it is not simply that sociocultural theory implies the developmental importance of children’s engagement with different views; it is also that Piaget’s account of such engagement seems to depend upon children establishing mutual goals and a shared understanding of the task at hand.

It is not only necessary for students to interact and exchange ideas—“they also have to ‘interthink’ (Mercer 2000).” For Mercer, this process “has some of the characteristics of what has been called exploratory talk (Barnes 1976).” Mercer and Littleton (2007) see this form of dialog as involving:

partners in a purposeful, critical and constructive engagement with each other’s ideas. Statements and suggestions are offered for joint consideration. These may be challenged and counter-challenged, but challenges are justified and alternative hypotheses are offered. Partners all actively participate, and opinions are sought and considered before decisions are jointly made.

The point is that it’s not just about expressing and listening to other ideas, it’s about resolving differences between points of view. “Through resolution they may even find themselves converging upon ideas that go beyond what any of them were capable of achieving individually,” an idea that fits with the Vygotsky’s idea of a ZPD. However, “groups that do not resolve their differences have been found to achieve learning gains as well as groups that do resolve (Howe et al. 1990).” The impact of this sort of dialogue is not just short term (Howe et al. 1992). For Mercer, “The key factor is that children experience dialogue resembling exploratory talk.” Mercer introduces what he calls a “more sophisticated mapping of scientific concepts”:
  1. (a)

    gaining scientific understanding involves taking on new conceptual frameworks and ways of evaluating knowledge;

  2. (b)

    the taking up of a scientific perspective quite commonly involves the critical examination of a more “everyday” perspective on natural phenomena

  3. (c)

    this learning process consists, at least in part, of induction into a perspective and a new discourse by a relative expert (the teacher); it is not achievable by “discovery learning.” (Mercer et al. 2004).


When Wells writes about Vygotsky and the Zone of Proximal Development it struck me that what is left unstated is that, presumably, the ZPD depends on the teacher concerned as well as on the student’s ability, mood, emotions and motivation (which, to some extent, is teacher dependent). Ausubel’s (1968) dictum that the most important single factor influencing learning is what the learner already knows, might need to be amended to “the most important factors influencing learning are what the learner already knows and the level of their motivation to learn.”

Groups and decisions

Reading Mercer’s contribution I was reminded of Group Processes by Rupert Brown (1990). In it, Brown discusses group decision making in a wide range of contexts and notes that Janis (1972) identifies five factors that led to “bad” decisions (i.e., things turned out badly for the decision makers) as a result of the decision:
  • the group making the decision was very cohesive*

  • it was insulated from information from outside the group

  • the group rarely searched systematically through alternative policy options to appraise their relative merits

  • the group was under stress to make a decision quickly

  • the group was nearly always dominated by a very directive leader

* more controlled studies have questioned the effect of this factor

Janis’ observation, that these conditions led to strong conformity pressures resulting in “groupthink” (in that the group members effectively exerted conforming pressures on possibly dissenting individuals) might help us to think more about what does go on when groups of adolescents are “working together” solving their own inquiries. A caveat that might inform our design of pedagogic strategies comes from the social psychological perspective of Eiser and van der Pligt who comment:

[evidence suggests] that the conscious thought preceding a decision may be of a relatively simple nature, given the difficulty of processing complex information. People seem to rely on simple heuristics for making probability judgements and hardly seem to think about more complex combinations of probabilities and values or utilities involved in a decision. (1988, p. 181)

So, where next?

Smardon suggests science educators could examine what research has to say about the sociology of emotions. Smardon notes that Erving Goffman’s work has continued to interest many US sociologists since his death in 1982 [Ironically for this debate, Goffman graduated with a degree in chemistry]. She argues that: “the extensions of Goffman’s work may prove useful to science education researchers” because of the focus on “the mechanics of face-to-face social interaction (Collins 2004) and the performative nature of social life (Alexander et al. 2006).” Goffman developed the already existing ideas of symbolic interaction to develop a micro-sociology. Smardon also suggests that for help with “analyzing social interaction and the negotiation of power,” the work of Fred Erickson (2004) provides some affordances: “Erickson provides accessible tools for thinking about the social ecology of mutual attention. He provides methodological means to analyze both vocal and nonvocal communication.”

Drawing on the work of Scheff, Smardon suggests that: “we need to understand what kinds of classroom environments threaten social bonds and what kinds of teaching techniques allow conflict to proceed while maintaining attunement, allowing students to feel that they are legitimate participants.” Smardon also notes that: “evidence suggests that consumption is playing an increasingly important role in social stratification among youth.” Therefore, it follows that researchers should pay careful attention to how students (Milner 2004) and parents (Pugh 2004) employ consumer culture to establish inclusive and exclusive status groups. Here the work of Cote and Levine (2002) who coined the term “Identity Capital,” might be of interest. They argue that the identity capital notion can help to interpret both the intrapersonal and interpersonal manifestations of identity in late modernity. Cote and Levine (ibid.) use “identity capital” as a way to consider people’s complex relationships with themselves and others in terms of the global economic system:

Thus, we have seen the need for a concept representing a different type of capital associated with identity formation, namely, the varied resources deployable on an individual basis that represent how people most effectively define themselves and have others define them, in various contexts (p. 142).

Smardon identifies a “meta-historical mood that contextualizes science education and that mood appears to be anxiety about the changes being wrought by a globalizing knowledge economy.” The proliferation of standards-based reform, she argues, forces educators to stop and ask: “What is the purpose of education? What is the purpose of science education?”

Do we really believe the Education Gospel (Grubb and Lazerson 2004), that science and technology will provide jobs and salvage our flagging economy and solve all of our social problems? In an era where national science policy is a disguised economic policy designed to close the “innovation gap” and create a nationally competitive edge (and these policies exist in many countries not just the U.S.), what does it mean to democratize science? Do we have an obligation to produce entrepreneurial scientists for our nation?

This point resonates with me as it is time that we, as a science education community began to challenge a major shibboleth. As Jonathan Osborne and I have argued in a recent report on “Science Education in Europe”:

Nevertheless, much of the concern [about science education] is focussed around the issue of supply and fails to recognise that science operates in a global context. Here the evidence would suggest that, at a global level, there no shortage of doctoral scientists. For instance, evidence from the US context shows that there is an oversupply of students with biomedical PhDs and, as a consequence, the success rate on applications to the National Institute of Health, the government agency responsible for funding research, has declined from 26% to 19% from 2000 to 2005. Likewise unemployment in science and engineering professions in the US follows the overall rate and is not markedly lower. If there is no substantive demand for scientists overall then increasing supply without increasing demand is, at best, unwise and wasteful and, at worst, morally questionable. (2008, p. 14)

Hewson’s, Vosniadou’s and Tiberghien’s responses to Roth, Lee and Hwang

The responses to the Roth, Lee and Hwang article are qualitatively different from those to the Treagust and Duit paper. This difference can partly be accounted for because Roth et al. set out to take on a task that they perceive as being more challenging than the one that Treagust and Duit set themselves. Starting from the premise that: “Science educators have however yet to ask themselves what it would mean to consider the possession of conceptions as well as conceptual change from the perspective of cultural studies,” Roth et al. aim to set out:

the first principles of such a cultural approach to scientific conceptions with a view to generating less presupposing and more parsimonious explanations of this core issue within science education than if conceptions are supposed to be structures inhabiting the human mind.

Note that for Roth et al., the term “conception” means “the ensemble of ways of talking about some entity” (although their point about the number of planets in the solar system is rather spurious). They go on to define “concepts” as “cognitive entities, pieces of furniture of the conscious mind that are unlike signs (Pines 1985). They are human inventions that: “once labeled become communicable through the use of language” (p. 108)” and then, again, describe “conceptions” as “the way in which individual humans conceive of concepts.”

Their message is, instead of “having to deduce internal models and cognitive frameworks” which require levels of abstraction, “teachers only deal with ‘ways of talking’ as people get about the serious business of living in the world.” And so, “What teachers have to assist students in, therefore, is developing ways of talking that are contextually appropriate.”

Between them, Peter Hewson’s and Stella Vosniadou appear to refute, dismantle and undermine Roth et al.’s position. While Vosniadou brushes aside their points with elegance mixed with barely disguised contempt, in boxing terms, Hewson offers the punch that Ali never gave Foreman.

Both Hewson and Vosniadou observe the cardinal rule of feedback—start with the positive. Hewson agrees with Roth et al. about a number of points: the coordination of speech and gesture; the significant role gestures play in the here-and-now; the need to be aware of the limitations of data reduction; the ability to speak about novel topics; and, the need to recognize that participants seek to make sense of concepts even if they go against accepted scientific ideas. However, that’s a bit like evolutionists and creationists agreeing that bears shit in the woods, and, at this point Hewson’s gloves disappear.

In Hewson’s view, Roth et al. make the following mistakes, they: “largely ignore the external context”; hold back important information; take a “restrictive view”; use a “a hard, dichotomous logic”; fly “in the face of sound logic”; take “far too extreme a view to be taken seriously”; and, produce “grotesque caricatures that have no place in a scholarly article.” In short, Roth et al.’s position is “too simplistic and too extreme” and “to put it charitably” the authors are “ill-informed.” Hewson argues that what we need is “a reasoned exploration of views, starting, perhaps, with an identification of points of agreement.” Which is what we find in the measured responses to the Treagust and Duit paper.

Let’s examine how Hewson approaches each of Roth et al.’s three major claims.

Claim 1: The Primacy of the here-and-now

Hewson argues that Roth et al. largely ignore the external context providing “no information about who I and M are.” Roth et al., in Hewson’s eyes, are hoist with their own petard because the “dichotomous view of context that emerges is (part of) the price for insisting on the primacy of the here-and-now.” It is obvious to Hewson that: “knowledge of the external context of an interview is essential to any claim that it is at the very least comparable to other interviews in the science education research literature.” I am not convinced by this argument, the verbal interaction, whether it’s an interview or a teaching sequence, seems to have many of the characteristics of exchanges in the conceptual change literature. The point, though, surely is that many/most sociocultural researchers would want to know more about the context. Roth et al. appear to be reinventing “sociocultural” without much culture beyond the here and now. Hewson recognizes this point when he argues that a “less restrictive view of the relationship between the interview itself and its precursors and successors might accord primacy to the interview but could also give importance to what came before and after.” Indeed, but my question would be what does “give importance” look like?

Claim 2: Language provides all necessary resources

Hewson examines Roth et al.’s second claim that language provides all necessary resources. But Hewson argues that M and I have agency and that: “Language doesn’t preordain the exact form and substance of this paragraph; rather it provides the opportunities for considerable variety.”

Claim 3: Interview participants do not hold fully-formed mental models prior to the interview

Hewson’s reading of the transcript shows that: “we can conclude that M brings her language competency to the interview, a competency that exists outside of the context, the confines of the interview.” Hewson is happy to accept Roth et al.’s claim that their analysis demonstrates conclusively that M. did not have a fully-formed model that she introduced into the interview, but he makes the obvious point that: “to generalize on the basis of this one case to deny the existence of fully-formed models under any circumstances flies in the face of sound logic.”

Stella Vosniadou, editor of the new Handbook of Research on Conceptual Change appears in no mood to suffer fools gladly. While agreeing with Roth et al. on many points, she begins by pointing out that they have set up a “straw man” in that: “over the years practically all of the above mentioned tenets of the classical conceptual change approach have been subjected to serious criticisms.” Put another way, Vosniadou is saying “Well, yes, OK, but you’re a bit late in arguing against the strong conceptual change research model.” Vosniadou, politely but firmly, points out that in the “reframed approach, the emphasis is not on misconceptions as unitary, faulty conceptions, but on knowledge acquisition as a complex and intricate process that proceeds through various kinds of modifications.” Researchers now make a distinction “between naïve explanations of natural phenomena, based on everyday experience in the context of lay culture and scientific explanations.”

Vosniadou clearly sees the value of focusing on learners’ mental models:

In this reframed approach, many misconceptions are seen to result often from students’ attempts to synthesize the new, scientific information, with existing beliefs based on naïve physics. This is why we call them “synthetic models.” Synthetic models are not stable, alternative theories, but dynamic, situated, and constantly changing representations that adapt to contextual variables and/or to the learners’ developing knowledge.

However, Vosniadou’s disagreement with the position advanced by Roth et al. “centers mostly around the attempt to ‘eradicate’ mental representations.” Roth et al.’s explanation:

is without explanatory power because it results in cutting the ties between speech and perception, and leaves open the question of how we can talk about phenomena in the physical world without appealing to representations of our experiential and cultural knowledge in the world.

Using the example of Mary and her interlocutor, Vosniadou points out that we all create mental models and derive from them possible explanations of phenomena. Mental models are cognitive resources that: “allows us to use perceptual information creatively for the purpose of explanation.” Vosniadou sees Roth et al.’s argument as being that: “because communication is produced in real time and because we can speak about topics we have never talked or thought about before. We do not need mental models or something that must predate our talk.” In Vosniadou’s opinion, Roth et al. “seem to think that mental models are static and must predate our talk,” but she argues, this is not the case:

Mental models can be situationally produced and coordinated in real time to deal with the demands of the situation. They can predate our talk or follow it. We can create a mental model and run it in our mind to derive new information and new explanations which are based on perceptual knowledge and which are not readily conceptually available.

Vosniadou’s point is that the:

presence and use of mental models does not have to be assumed. It is exhibited in students’ drawings and their constructions of models during the interview […] Mental models are not useless. They provide the necessary link between language and perception.

Hammering home the point, she argues that: “Linguistic communication cannot be the sole basis for explanation in science. If that were the case there would be no reason to conduct experiments.”

Vosniadou comes closer than any of the authors of papers in the special issues to identifying differences between conceptual change approaches and sociocultural approaches when she argues that: “sociocultural approaches to learning emphasize the importance of cultural artifacts and their role as cognitive facilitators.” “But,” she argues, “they do not explain how human culture created these artifacts in the first place.” In Vosniadou’s eyes, “model-based reasoning is the key to understanding how humans construct the rich cultural environments that mediate our social and intellectual life.” Looking forward, Vosniadou notes that:

Many researchers are now arguing for a distributed cognitive system that can generate internal representations of the environment when necessary, but can also use salient resources in the environment, such as cultural artifacts, in a non-reductive way.

In such a system, “mental models play an important role. Individuals can form mental models not only of their everyday, physical experiences but also of the cultural artifacts they use.” So, she concludes by refuting one of Roth et al.’s points by arguing that:

It would not help science instruction to prevent educators from teaching children how to use mental modelling for conceptual change in science. Mental models can play an important role in conceptual change because they can form the basis on which new information can enter the cognitive system in ways that can modify what we already know. They can be used by children and by scientists to conduct thought experiments and simulations that can help them see the differences between alternative explanations of phenomena and to test the implications of principles or theories.

In that last sentence, Vosniadou comes closer to bridging conceptual change research and sociocultural perspectives than most, if not all, of the other contributors.

Andrée Tiberghien presents an outwardly more detached perspective on Roth et al.’s paper. Her final words are encouraging “I believe that science education research needs theoretical debates in relation to methodological components of the theories and in this perspective this issue contributes to the advancement of research.” Her critique of Roth et al.’s paper, begins matter-of-factly: “Commenting on a paper that deals with rather new theoretical issues in science education is a challenge. A main reason for this challenge is that the different theoretical frameworks are not very developed.” So, Tiberghien’s view is that we’re not dealing with anything too sophisticated in Roth et al.’s paper and that colors her view of what they have written.

In what, to a British academic, seems unusual for a continental European scholar, Tiberghien takes a “rather pragmatic epistemological position.” Although, shortly afterwards pragmatic becomes “pragmatic” in that:

in the sense that several research results lead me to consider that students’ understandings during a teaching sequence (of several weeks or months) can be illuminated by several components of a classroom situation as diverse as social situation, kinaesthetic perception, type of knowledge, type of lexical or syntactic forms of language, etc. It is not a question of first or second order of a component, each of them can at a time determine the students’ constructions.

So, Tiberghien is, in effect, advocating methodological pluralism based on her experiences of researching teaching and learning for nigh on three decades.

Interestingly, in a paper that deals with the importance of language, Tiberghien notes some linguistic issues, such as when she writes that Sensevy (2007) notes that: “the term ‘to teach’ in a certain way requires the term ‘to learn’; the term to learn requires the term to teach.’” Researchers who speak English as a first language miss out on some crucial understandings of knowledge, learning and teaching because they can’t access continental European languages or literature.

Tiberghien concludes her opening salvo by arguing that:

In conclusion, one could wonder to what extent the object of study chosen by the cultural approach is more adapted to study teaching/learning situations than students’ evolution over a longer period of time than the duration of the situation. […] Personally, I think that the combination of socio constructivist view of learning and of the joint action theory of teaching and learning focused on knowledge allows me to study the relations between classroom practice and students’ evolution based on an individual questioning.

Tiberghien sees value in the cultural approach because it “emphasizes the conversation as an important phenomenon in science education” and goes on to note “the richness of introducing a new theoretical framework to investigate questions already studied” because “they show how the implicit aspects can be made explicit and interpreted and open new questions in particular about the social competencies and their role in creating sense.” Noting further that, depending on the questions they want to answer, researchers have to “select elements of the situation and they have to make implicit hypotheses at least for some types of selection even if the theory orients the types of elements taken into account.” Tiberghien then contrasts researchers such as Hans Niedderer who engage in long term studies of teaching and learning with Roth et al.’s focus on the conversation, commenting that the latter researchers “do not employ an explicit methodology dealing with evolution on a much longer time, even though it is probably implicit.” It’s at times like this when you wish you’d got all the authors in a room together so that you could thrash out all these unknowns.

Tiberghien’s key criticism of Roth et al.’s paper is that when they claim that: “the in-between forms of talk, which neither can be due to a misconception nor to a conception” are “beyond the capability of a conceptual change approach” they contradict evidence of intermediary conceptions. “Intermediary conceptions,” Tiberghien argues are: “precisely an in-between, partly misconception and partly conception in reference to physics school knowledge.” During her career, Tiberghien has moved towards examining “small elements” of students’ utterances—at the level of the sentence in order to contrast learners’ and teachers’ views of the world. Tiberghien points out the somewhat irrational nature of conceptual development as learners struggle to make yet another sense of the world, which involves adding new experiences, ideas from teachers and others and their own extant views of natural phenomena. No wonder effective learning can be hard.

Some years ago, Rod Watson, Teresa Prieto and I studied children’s alternative frameworks within the realm of combustion. We found many children who operated within one or other of two coherent views of the phenomena but many others who held what appeared to be inconsistent views. We concluded that:

In changing from one framework of chemical transformation to another, students develop the precision with which they use language, replace aspects of the old framework with aspects of the new, incorporate new concepts, and sometimes retain aspects of both frameworks simultaneously. In view of the complexity of the change between the frameworks, the process of change may take place over a substantial period of time. Inconsistency in the students’ explanations may, therefore, be an indicator of conceptual change. (Watson et al. 1997, p. 440)

These processes can happen quickly or slowly and do not have to involve interactions with teachers or other learners. The issue is that seeming inconsistency to an outsider might be a good thing and an indicator of developing thinking. This aspect of learning has implications for teachers who need to be aware of, and track, changes in thinking rather than simply monitor and assess conceptual understanding at the end of a topic. The irony is that it is the behaviour of researchers, studying long term development of understanding in students in relatively non-judgemental ways that might provide models for teachers keen to teach science more effectively. Although work has gone into promoting teachers as researchers, the demands of summative assessment would seem to be philosophically at odds with the need for promoting meaningful learning.

Tiberghien’s approach to data collection, which can involve collecting hours of video-taped interactions “has methodological consequences.” The researcher must “break down the productions into small elements […] and at the same time [keep] the information about the context in which the student or the teacher utters this element.” She argues that: “data should be quasi systematically analysed over time and not by selecting particular moments.” It is somewhat ironic, then, that Tiberghien presents us with a particular moment to illustrate her arguments.

In the example, students L and N are throwing and catching a medicine-ball. Tiberghien argues that the transcript of what happens illustrates that the students’ “own perception mediates their interpretations and their answers; not only the perception of catching the ball themselves but also viewing the other doing it play a role; perceiving and talking play a major role in this case.” Tiberghien’s extract and analysis are more convincing than that provided by Roth et al. and by Treagust and Duit. However, the words on the paper don’t tell us enough to make complete sense of the exchange. In line 7, L’s statement, “upwards like that,” could be said with conviction or with hesitation. Similarly in line 9, L says “but you absorb” might actually be “but you absorb …?” There will always be problems in presenting convincing arguments about learning when our main medium of communication is words on paper.

Finally, Tiberghien’s position on the value of models and modelling seem more at ease with Vosniadou’s stance discussed above. For Tiberghien, “modelling the material world is at the core of physics” and, presumably, the rest of science. However, somewhat confusingly, Tiberghien writes:

This choice of modelling in physics leads us to distinguish two worlds: the world of objects and events which refers to the inanimate material world and that of theories and models which refers to the theoretical aspects and models of studied material situations. The models are intermediary between the material world and theories.

Elaborating on this position, Tiberghien states the “rather strong hypothesis that, in everyday life, there is also a modelling activity of the material world.” One could argue that Tiberghien herself is trying to articulate a model of one aspect of the material world—conceptual change in learners’ minds.


I began by stating that I came to these eight papers with “a conviction that science teacher preparation should include developing an understanding of conceptual change, constructivist science teaching, dialogic teaching and inquiry-based science inside and outside the classroom.” Nothing I have read has changed my opinion substantially. What I feel that I have gained is a broader view of the relationship between different perspectives on the discussions.

I’m not convinced that we saw much “exploratory talk” from the contributors. It was a bit like watching the first round of a panel discussion and then finding that you’d run out of time and the hotel staff were waiting to clear the room for the “Awards Luncheon.” All of the contributors deserve awards for their scholarship and openness but I’d be more interested in hearing what they had to say in response to the first round of opening comments.

Mercer notes that:

It is our natural habit to express our ideas in dialogue, to test our views against those of others, and to attempt to persuade other people to share the conceptual understandings that we believe are the best. It is of course also normal that we resist changing our minds, if the views we hold are bound up with aspects of our social identities. But, nevertheless, most of us proceed as if we believe that one of the most important ways of changing someone’s mind is to talk with them. It seems, though, that cognitively-orientated conceptual change researchers stand outside this near-consensus – which of course is why sociocultural researchers like myself are keen to engage them in dialogue.

I don’t see much dialogue in this set of papers. In the same way that reading the interview transcripts of the two initial papers is frustrating, then so it is in reading the scholarly commentaries. I see polite agreement with the basics, the uncontroversial and, in many cases, the broad thrust, but I don’t see much evidence of learning, of development or of changing directions. What I do see is more like that old joke about someone who is stopped in the street and asked, “How do I get to New York?” After a short pause he replies, “Well I wouldn’t start from here.”

My involvement with science education research began in earnest when I undertook a part-time Masters in Science Education at King’s from 1985–7. In 1985, John Head, who was a lecturer at the college, had published a book, The Personal Response to Science (Head 1985). Almost a quarter of a century ago, Head was arguing for science teachers to make the subject more meaningful. He offered four tactics: reformulation; application; making use of prior knowledge; and, involving the learner in the teaching strategy. One has to ask, how far have we come when one reads Head’s description of “reformulation”:

If a student has to explain a concept “in his own words” then the repetition, parrot-fashion, of rote learnt material will prove inadequate. In order to express the concept in everyday language with familiar metaphors the learner has to relate the newly met concept to the pre-existing cognitive structure, and in so doing meaningful learning occurs […] The obvious managerial problem for the teacher is to find ways of giving all the pupils in a class opportunities to talk and write in this fashion. It would be glib to minimise this problem, but clearly the form in which written work is sought from the members of the class and the nature of discussions within small groups, as during practical work, can determine the opportunities given to individual students for such expression. (Head 1985, p. 86)

I began by wondering to what extent I could use the eight papers as a resource for myself, for colleagues and for our preservice teachers. Between them they provide snapshots of contemporary thinking as well as summaries of previous work. But I worry that the pace of progress in science education research is painfully slow and that the insights that we have now that we didn’t have in the 1980s aren’t really much to write home about. We owe the previous generation of researchers a great debt for the work that was done in developing understandings of what learning science involves. But we owe more to future generations who will build on our ideas and evidence. Would that they be impressed.

Copyright information

© Springer Science+Business Media B.V. 2008