• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Creativity in Science Education: Perspectives and Challenges for Developing School Science

Posted on: Thursday, 26 April 2007, 06:00 CDT

By Kind, Per Morten; Kind, Vanessa

INTRODUCTION

Creativity, the ability to produce novel and appropriate work (Sternberg &c Lubart, 1999), is one of humanity's most important traits. The concept applies to historic novelty, generating ideas and artefacts that arise for the first time in human history, and to individual novelty; ideas and artefacts new to the person who creates them (Boden, 1990). Despite its importance, creativity is not yet fully established as a mainstream topic in psychology and/ or education research: neither does it hold a significant position in educational practice (Boden, 2001).

Educational systems world-wide are being reformed to adapt to rapid societal changes, due to global economic restructuring and technology development (Bellofiore, 1999; Bentley, 1998; Burbules & Torres, 2000; Selzer & Bentley, 1999). Students need to be prepared for:

....life in a world about we know very little, except that it will be characterized by substantial and rapid change, and is likely to be more complex and uncertain than today's world. (Hodson, 2003: 46)

In general terms, we support the view of NACCCE1 (1999) that creativity education has a part to play in helping students meet the unpredictable demands of the future. Training students' creativity may contribute significantly to their flexibility, and their ability to handle changes in their working lives. We believe that each school subject should emphasise creativity, within an agenda reflecting the characteristics of each. In school science, this means reflecting the concept of scientific creativity, leading to the tantalising and fundamental question: will training students' scientific creativity contribute to their being more able to handle the challenges and uncertainties of their future lives? The science education community is not in a position to offer answers: as this paper will illustrate, engagement with creativity in school science is currently at a much lower level than is required even to begin approaching an answer. We therefore offer this paper as a step towards constructing a platform for generating deeper engagement with creativity in science education research, and thus starting the process of making answers possible.

The paper has three main sections. In the first, we review common approaches to creativity in science education. This section serves to illustrate our point that current interpretations of 'creativity' are far removed from those needed to be meaningful in the above context. Next, we highlight psychological approaches that have received more systematic treatment. These offer the beginnings of underpinning theory necessary for taking creativity in school science beyond the approaches described in the first section. Thirdly, we summarise perspectives from the review and look for further routes towards making science education a contributor to developing students' creativity.

INTERPRETING 'SCIENTIFIC CREATIVITY'

Central to this paper is identifying the meaning and content of 'scientific creativity'. In common with creativity in general one 'right' approach or definition does not exist. Instead, we acknowledge the need to tolerate an elusive concept that is inevitably used by different workers in different ways, as subsequent discussion and description will illustrate. This does not preclude our attempt to establish a specific framework and/or meaning for 'scientific creativity' that we believe sits comfortably within the science education world. Accordingly, we set two criteria for a school-centered rationale for 'scientific creativity'. The first is that it should be based on what 'real' scientists do. By this we mean that scientific creativity in school science should be rooted in and reflect aspects of creativity seen in scientific research. Our second criterion is that any approach to scientific creativity should devise a framework appropriate to children's needs and abilities. We thus claim that any approach to scientific creativity in school science be 'authentic' in scientific research terms, and meaningful in the school context. This is essential in accepting the role of children as learners: many attempts to create 'authentic' school science have ignored the principal difference between scientists as developers of science knowledge and school children as learners of science knowledge (Hodson, 1998; Woolnough, 1998). These two criteria are in the background throughout the paper.

CURRENT APPROACHES TO CREATIVITY IN SCHOOL SCIENCE

Literature reviews on creativity, such as Baron and Harrington (1981), often have been organised along Rhode's (1961) scheme of creativity dimensions: the creative person, product, process and environment. A review on creativity literature in science education, however, supports the claim made by Runco (2004) that this categorisation does not suit literature with a 'disciplinary emphasis'. The literature mostly follows subject-related issues, often with an indirect link to creativity, rather than relating to the main trends of general creativity research. Due to this, the present review is organised around categories that show the various contexts in which science educators relate to creativity. This first section includes four categories that all link to established practices in science education:

* creative teaching

* art and science

* inquiry science

* the nature of science.

Other categories were considered but set aside: the link between creativity and the education of 'gifted and talented children' appears frequently in literature, but with weak evidence for any positive relationship between creativity and effective provision (Feist, 1999). 'Creativity and engineering' is also ignored, even though this features frequently in engineering education journals. The connection with engineering raises perspectives on differences between creativity in science and technology, drawing attention away from the specific issue of creativity in science education towards links between science and technology. A separate article would be needed to do this justice.

Creative teaching

A significant proportion of science education literature uses creativity merely as a descriptive label. Sternberg and Lubart (1999) noticed a similar tendency in their general review of creativity and made the decision to exclude such literature, because it is not truly investigating creativity. In the present review we take another approach: in education it is useful to make a distinction between teaching for creativity and creative teaching (NACCCE, 1999). The first makes creativity a learning outcome, while the second is simply a characteristic of teaching. NACCCE defines creative teaching the following way:

teachers using imaginative approaches to make learning more interesting, exciting and effective. (102)

This might be read as a general characteristic of good teaching, depending on the meaning of an 'imaginative approach'. In science education literature we clearly find 'creative' associated with good teaching in general and often used as a label contrasting with 'bad traditional teaching'. Figure 1 illustrates some of the characteristics of this use of 'labelling' in the literature, all exemplified with one reference as an example.

Figure 1: Contrasts commonly found in science education literature between creative and traditional teaching.

The figure suggests that 'creative' teaching is associated with open-ended, student-oriented, exploratory and group-based learning strategies. In science the debate is extended to include 'hands-on' activities in the laboratory or outdoors. These are settings regarded as inviting openness and freedom. In some cases, though, the distinction is not as clear cut-the contrast between 'open- ended' and recipe-like tasks, of course, also exists within the laboratory setting.

The claim may be made that stereotypes of traditional and creative teaching are of little value, as good science teaching can be found by adopting a balanced approach between the two extremes rather than within one of them. Much 'traditional teaching' could provide excellent contexts for creative activity. Figure 1 perhaps above all indicates a tendency among educators towards idealism, and a loss of perspective when describing teaching. The term creativity may be seen as contributing towards this problem, as it becomes a self-justifying label which is so 'politically correct' that it avoids critical scrutiny. Alternatively, the extreme stereotypes listed in Figure 1 can be seen as reflecting real contradictions among educators, or more importantly, real problems found in many science classrooms, and creativity thus becomes a positive aim that may guide science educators towards a solution. Breaking away from the negative aspects of 'traditional' teaching requires teachers to be idealistic and have belief in alternative values. The concept of 'creative' teaching may serve as something to aspire to in order to achieve this. The significance of labelling this idealism as 'creative' is indicated by the TIMSS survey (Beaton et al., 1996), which reports that a majority of teachers in most countries regard creativity as 'very important' for success in school science.

However,extant literature offers few examples of nuanced analysis of what it is that actually makes science teaching creative, and without this the labelling loses its value. Some hints for further engagement to clarify this may be found in the NACCCE report:

[creative teachers] must recognise when encouragement is needed and confidence threatened. They must balance structured learning with opportunities for self-direction; and the management of groups with attention to individuals. They must judge the kinds of questions appropriate to different purposes and the kinds of solutions it is appropriate to expect. (110)

Thus, 'creative' teaching is not restricted to a certain context, say, the laboratory, classroom, working in groups or as individuals, but means the way a teacher manages and organises learning. Escaping from the 'creative means good' argument is nonetheless difficult, fuelled additionally by the sense that if we want creativity to be more than a 'label' we need to focus on the 'ends', that is how best to teach for creativity, rather than just 'the means'.

Art and science

Science and art are commonly regarded as contrasting areas: one represent rationality and logical reasoning while the other is seen as primarily aesthetic. Unsurprisingly, creativity for this reason is associated more with art than science. Both, however, include creativity. This is often a key feature when the two areas are integrated in education (Dobbs, 1995). Science education literature offers two contrasting approaches in work with creativity in the context of art. One follows the 'common sense' view regarding art as representing creativity much more than science and, therefore, as a 'tool' for making 'the rational' science education more creative. The other stresses similarities between art and science, emphasising science as an aesthetic and creative activity, both in knowledge development and in the nature of the knowledge itself. Broadening the pedagogy of school science and enriching science learning environments are common features of most projects in this area.

Watts (2001) illustrates 'the art as a tool' approach by arguing that poetry is a way to gain an aesthetic, poetic experience in school science. His argument is based on three premises. First, that 'poetry and science can work together to enable learners to grow in familiarity with the concepts, facts, principles and processes with which they are working' (200). Poetry is presented as a special way of observing the world, contributing to deeper understanding of nature. second, poetry gives a chance to 'play with words and toy with language' (201). This encourages students to play creatively and imaginatively with science concepts and ideas, especially when ideas are 'half-formed'. Third, using poems may 'fuel the process of learning' (201) by strengthening the emotional dimension. degaard (2003) uses similar arguments to justify the inclusion of drama in school science, showing examples for how this may be used to increase students' learning of science concepts and their understanding of the nature of science and science in society. Projects integrating a wide range of different types of art and science along these lines are relatively commonplace, although most are presented as teaching ideas in teacher journals (for example Herold, 1992; Khourey-Bowers & Baxter, 1995; Lock, 1991; Mesure, 2005; Stokes, 2001; Ward, 1987 and Yasso, 1991). Similar arguments to those discussed above about the positive impacts on learning that such 'creative' teaching methods bring are found in justification. However, these projects take creativity as a stronger focus in itself, as the chosen artforms claim to generate a learning environment of creativity and imagination (degaard, 2003). Significantly, though, all the projects take art, not science, to represent creativity.

An approach emphasising the similarities of creativity in art and science is found in Massoudi (2003), who discusses 'creativity and spirituality' in scientific writing. To him creativity is 'a manifestation of inspiration' (115), emphasising that people being creative are wholehearted participants in a state of wonder and awe:

Being in awe means to see the most seemingly normal and obvious events as amazements. (117)

Artists and scientists share this sense. In their creative moments both enter a spiritual state, achieving a special form of seeing:

This form of seeing is not with our ordinary eyes, but with the Eye of the Heart. (118)

Massoudi links these aspects of creativity to writing, claiming that scientific writing is similar to creative writing. He uses this argument to suggest that school science should include more creative writing. Girod, Rau & Schepige (2003) also present scientific creativity as relying upon the same aesthetic tools of thinking as arts: science involves not just the process of stepping back and analysing the world with 'cold logic and rigorous methods', but also '...stepping forward in an attempt to get inside of objects, events and ideas' (577). They argue that science involves understanding nature with ones' heart and mind, emotions and cognitions, imagination and reason. Art, therefore, acts as a source of skills and insights that science needs to progress. These authors argue that the same thinking should apply to teaching and learning science and art at school: Teachers should strive for similar but developmentally appropriate experiences with beauty and aesthetic appreciation of science ideas' (577), applying Dewey's (1934 /1980) aesthetic theory to identify elements of aesthetic understanding.

Studies such as these raise the question as to the degree to which art really is 'more' creative than science, and in what ways art and science share a common creativity. The first is probably true as far as numbers go: anyone who makes a living from art must be creative, but scientists may work in the full spectrum, from doing routine technical jobs to frontline creative thinking (Feist, 1999). In fact, as Kuhn (1970) argues, most science is 'normal' and just rarely does some individual produce truly 'revolutionary science'. This, however, is not an argument saying that art of itself is more creative than science. As will be discussed later in the paper, science clearly has its own means of creative expression.

The second suggestion, that art and science share the same creativity, raises more complicated issues. There is strong evidence for similarities between the creative process in art and science. Bohm (1998), for example, makes such claims. He argues for this based on the common origins of art and science. Early humanity's forms of art and science comprised a unified set of perceptions and responses to the surrounding world, that can be traced through to today. For example, we may feel a sense of 'wholeness and beauty' when in the presence of a magnificent painting and, separately, appreciating a major scientific theory. The truth of scientific theories arises not only because they correspond to observable facts, but also because they contribute a sense of wholeness in our understanding of the world.

Evidence is, however, also apparent for the opposite view; that creativity in art and science are different. For this, we look simply at the importance of domain-specific knowledge on creativity (Csikszentmihalyi 1996, Gardner 1993, Weisberg 1993). It is generally acknowledged that people who do noteworthy creative work in any given domain are almost invariably very knowledgeable about the domain (Nickerson, 1999). Another argument is, of course, the role of rationality. Creativity in science aims to develop consistent theories and any step therefore includes both context of discovery and context of justification (Popper, 1963). Even if there might be interesting common principles in art and science, these may therefore prove less important when illustrating authentic creativity in the two areas.

In our opinion, successful creativity projects integrating art and science include some elements that may add vitally to science education. All include an element of presenting creative products; for example performing theatre plays or role plays (Mesure 2005; 0degaard, 2003), making pictures or writing poems (Lock 1991; Watts, 2001). In this respect, art is very different from science. Art education from the beginning focuses on expressing personal feelings and opinions through creative 'products' and showing these to others. Science education, as shown earlier in the paper in the image usually offered of 'traditional teaching', has had a practice of treating students as 'passive learners'. This stark contrast may be the main reason why science educators turn to art when seeking to re-vitalise their subject. Science teaching, we think, would benefit from adapting the artistic process of self-expression, including 'risking failure, taking leaps of faith and trusting in a more creative approach when the outcome is not at all certain' (Mesure, 2005: 13). This might be a way of making teaching meet children's needs, while truly emphasising creativity. The challenge, of course, is to ensure that the teaching still has science as a main focus rather than artistic expression. To achieve this, awareness of the characteristics of scientific creativity and clear goals for the learning outcomes are needed.

Inquiry science

In investigative, or inquiry science students work on open- ended, investigative tasks, the underlying idea being that by so doing school science mimics real scientists' creativity. This idea occurs frequently in science education literature (for example, Gangoli, 1995; Washton, 1966). The words 'discovery', 'inquiry' and 'creativity' have been used as synonyms to describe this practice (Lucas, 1977). Our task here is to explore the extent and ways in which investigations in science really offer an arena for developing students' scientific creativity.\A contrast exists between the ideals on which inquiry based science is founded and the reality of its practice. Inquiry science is attractive to science educators for idealistic and realistic reasons. The approach dominated 1960s and '70s science curricula, about which DeBoer (1991: 206) writes:

If a single word had to be chosen to describe the goals of science educators during the 30-year period that began in the late 1950s, it would have to be inquiry.

Inquiry projects run by the UK's Nuffield Foundation (for example, Nuffield Foundation, 1971) and the American Association for the Advancement of Science in the US (American Association for the Advancement of Science, 1967) established strong, influential trends and ideas, leading many educators to become passionately convinced about the correctness and benefits of inquiry-based science teaching. Although the dominance has faded in more recent decades, inquiry still remains a strong slogan in the US (Anderson, 2002; Crawford, 1999; NRC, 1996; Zachos et al., 2000; Zion et al., 2004) and elsewhere (Abd-El-Khalick, 2004).

In reality, however, inquiry science struggles to achieve the hoped-for effectiveness in practice. The passion for the approach has led to too much credence being given to weak evidence of positive effects on students' learning. Some papers, for example, make conclusions such as: 'In general, research shows that inquiry teaching produces positive results' (Anderson, 2002: 2). Strong evidence contradicting its effectiveness has frequently been ignored (Newton, 1968). For example, Welch et al. (1981) analysed the role of inquiry in US science education between 1960 and 1980. They document a gap between the 'desired state' and the 'actual status', suggesting that while the aims for inquiry science may be desirable, what is actually achieved is much less than that hoped for. In the UK, Donnelly et al. (1996) evaluated the 1980s-1990s National Curriculum (DfEE, 1988) initiative for 11-16 year old students to undertake individual investigations (known as 1ScI' or 'Science 1') which, they concluded, is fundamentally ill-conceived. Many science educators have either not noticed or conveniently ignored the messages given in these papers.

The main flaw with inquiry science is that the promised freedom and openness that is perceived to exist in real science is rarely achieved in reality-for a variety of pragmatic (and no doubt well- intentioned) reasons, teachers inevitably frame students' investigations ('inquiries'), either by providing a fool-proof 'recipe', restricting apparatus or providing heavy guidance towards a specific route for achieving a solution. Embedded within this is the assumption that 'real science' has the qualities of freedom and openness that inquiry science so desires to match. In many cases, though, scientists work comprises repetitive tasks carried out within tight frameworks and limits. We also see here the elements of our initial discussion on scientific creativity-within inquiry science, the assumption that children should 'behave like scientists' is seen frequently, although the school context and children's naivety as learners makes this impossible to achieve in practice.

The convincing argument to support inquiry science as a way to train students' creativity, of course, would be empirical intervention studies testing its effect. Such studies, however, are hard to find. In the huge amount of literature linking creativity and inquiry science the only study found is Bills (1971), in which 306 fourteen year old (eighth grade) students participated in a quasiexperimental study: the experimental group had been trained on 'divergent thinking' through open-ended inquiry tasks. The result, however, was negative-no particular effect of the training was found. Bills suggests two different explanations; either training does not develop creativity or that creativity developed in the science tasks does not transfer to the testing tasks. Either explanation is a serious message that inquiry science offers no guarantee for developing students' scientific creativity.

A more satisfying approach to working with inquiry science to stimulate creativity may be found in award programs that invite school children to run science projects over a longer period, such as the Australian CREST2 project. These programs offer extended practical opportunities demanding student commitment and ownership. So far as we are able to determine in the absence of a critical evaluation, these appear to offer inquiry science in ways that meet 'creativity' criteria to a greater extend than traditional inquiry teaching.

The nature of science

Teaching the nature of science (NoS, or 'ideas-about-science' and 'how science works') is a long-standing aim for school science. Understanding NoS has been seen as a key contribution to achieving scientific literacy, the drive for which has been significant in recent years. Hence, NoS has achieved a degree of prominence (AAAS, 1989; DeBoer, 2000; Driver et ai, 1996; McComas & Oison, 1998; Millar & Osborne, 1998; Schwartz, Lederman & Crawford, 2004). Students' understandings of NoS are also linked to scientific problem solving abilities (Matthews, 1994) and their conceptual understanding of scientific knowledge (Leach 1999). In common with many concepts in science education, achieving a formal definition for NoS is difficult, as so many workers adopt the phrase with differing meanings, reflecting the wide range of explorations of this area. A basic comparison of NoS literature (Abd-ElKhalick, Bell & Lederman,1998; Anderson, 2000; Driver et al, 1996; Lawson, 2004; McComas & Olson, 1998; Wong, 2002; Yore, Hand & Prain, 2002) reveals these issues: methods of science (also called 'scientific approach to inquiry' and 'scientific inquiry'); the nature of scientific knowledge (that is, the issue of epistemology of science); the social practice of science (also called 'science as a social enterprise') and historical aspects of science. These last two areas are sometimes treated as internal to science, but also as an interaction between science, technology and society (STS).

Creativity is an aspect of the nature of science that normally features in school science research literature. McComas and Olson (1998), for example, studied aspects of NoS in eight state and national standard documents from Anglo-American countries, finding creativity as a common thread. All documents mentioned science knowledge as 'tentative', implicitly suggesting that it comprised created ideas rather than being a 'true' description of the world. Six documents mentioned the scientist as a 'creative' person. These two aspects generate a core message that runs through much NoS literature; that scientific ideas are creative products and that scientists are creative people. This is reflected in the outcomes of Osborne et al's (2003) Delphi study involving twenty-five scientists working in different domains, including teaching, 'pure' science and medicine:

Students should appreciate that science is an activity that involves creativity and imagination as much as many other human activities, and that some scientific ideas are enormous intellectual achievements. Scientists, as much as any other profession, are passionate and involved humans whose work relies on inspiration and imagination. (702)

In addition, Schwartz, Lederman and Crawford (2004) indicate that in science creativity is always matched by rationality, with experiments playing a crucial role:

Science knowledge is created from the human imaginations and logical reasoning. This creation is based on observation and inferences of the natural world. (613)

The Delphi study (op cit.) also rated 'experimental methods and critical thinking' highly. These studies suggest that science as 'creative ideas which have been subject to critical testing' is central to what is agreed as 'important for school'. Interestingly, Osborne et a.l (2003) found that teachers were the sub-group of scientists who most strongly favoured the creativity aspect.

Interestingly, little NoS literature examines creativity beyond these core messages. Given that NoS receives poor treatment in most classroom practice (Gallagher, 1991; Schwarz & Lederman, 2002) this might be seen as problematic, contributing to what McComas (1996, 1998) calls the 'myth of science'. His analysis of science textbooks found a commonly held view that 'science is procedural more than creative' (60). Other studies show that students think of data collection in association with creativity in science, rather than working with science ideas (Schwartz, Lederman & Crawford, 2004). The myth about science, however may also work the other way, as science is sometimes regarded as too creative, ignoring rational components. The potential exists for the development of more detailed theoretical frames to establish a more balanced and informed view of NoS in school science. A full discussion of which is outside the scope of this paper, but we offer the following comments as a starting point.

A way forward for better and more effective teaching about creativity in NoS relies on theoretical perspectives accessible to teachers. We may, of course, claim that a theory for creativity is already established in the much referred-to picture offered by Reichenbach's (1938) two modes of science: the mode of discovery and the mode of justification. Popper (1959 and 1963) implemented this in his science philosophy, but, like most science philosophers, he is not explicit about how 'discovery' actually happens. Another, more recent frame, might be that of Simonton (2004), who proposes a theoretical framework to explain scientific creativity based on the components chance, logic, genius and Zeitgeist. For Simonton, logic plays a crucial role in generating ideas, not just testing them. This is demonstrated through the development of many computer programs, so called'discovery programs', which replicate great scientists' achievements by applying analyses to empirical data (Kulkarni & Simon, 1988). Simonton further emphasises the interplay between geniuses and the environment through the concept of Zeitgeist, the spirit of the time. When many people work on the same problem(s) new ideas will be 'in the air' waiting for anyone to pick them up. Numerous examples of the phenomenon of 'multiples' (Merton, 1961) exist in science, when scientists have simultaneously come up with the same original idea: the conflicts between Newton and Leibniz and Darwin and Wallace are well-known. The contribution of Zeitgeist leads Simonton to claim that the image of the 'lone genius' is a myth, although he admits that in many cases separating the contribution from the individual and the science environment is difficult. Perhaps the most astonishing element of his theory, and the factor to which he gives most weight, is chance. This is not to imply that scientific creativity is accidental, but complex and affected by so many unknown causes that making anything more than probabilistic assertions is impossible.

Simonton's theory, along with others, is complicated. When teaching about the nature of science and scientific creativity in school science we therefore have to judge between presenting too simple a 'story' that may create 'myths' and misunderstandings, or a more authentic version that may be difficult for children to understand. There is no obvious compromise, but our suggested statements for a reliable picture are:

* scientific theories are creative products (ideas) made by scientists

* many scientists work on the same problems and new ideas (theories, laws) emerge by common effort

* most science theories develop over a long period in small steps

* some scientists are highly creative and make substantial contributions in their fields, but they always build on other people's ideas

* all scientists must use their imagination when contributing to the development of science.

* scientific theories are created in many different ways. The processes are sometimes highly creative and/or highly logic, rational and/or accidental.

* in science creativity and rationality always work together. Scientific creativity never works without rationality and strict empirical testing.

Whatever is agreed, an important issue is how to develop these understandings in school science among teachers and students. Gallagher (1991) has shown this task is more complicated than just helping teachers develop their understanding. Teachers may have different perspectives on the nature of science, treating their classroom practice of science as an established body of knowledge and techniques requiring minimal justification. Improving teaching therefore requires teachers to reconsider their roles, use of discourse, conception of learning goals and the nature of classroom activities (Bartholomew, Osborne & Ratcliffe, 2003).

A further issue is developing appropriate activities and material. Irwin (2000), for example, used historical case studies with 11-13 year olds to develop understanding of the power of imagination and creativity. His study suggests that concrete examples are more relevant to students than general statements about 'science being creative1. Similar ideas are found in McComas (1998) and Solomon (1991) who make students 'play' with scientific ideas and analyse their evidence in historical and modern contexts.

DEVELOPING SCIENTIFIC CREATIVITY: PSYCHOLOGICAL PERSPECTIVES

The previous section of the paper revealed a range of approaches to scientific creativity, most with pragmatic rather than theoretical underpinnings and rationales. In the next section we explore attempts that are more deeply rooted in creativity research. This takes us into psychological perspectives that offer more detail about the creative processes occurring in scientists and children. We explore these using a historical thread, looking first at psychometric approaches then more recent developments in cognitive research and imagination. This section deviates from being purely a review of science education literature, enabling deeper reflection on the possibilities offered by these areas of research.

The psychometric approach to measuring scientific creativity

Guilford (1950), during the peak era for psychometric research, introduced 'divergent thinking' as a key concept to describe the creative process, and suggested this could be measured with paper- and pencil tests. In so doing, he laid the foundation for a major change in creativity research (Sternberg & O'Hara, 1999). The change was two-fold. First, his test enabled creativity to be understood as an intellectual ability, and second, test instruments became available so that researchers could study this ability in a variety of situations and contexts. Today, psychometrics has lost dominance and cognitive processes are studied in new and more varied ways, but Guilford's concept of divergent thinking and his creativity tests remain. We explore the influence this has had on creativity research in science education; first by reviewing direct uses of Guilford's theory and, in the subsequent section, by reflecting more widely on the cognitive approach to scientific creativity.

Guilford (1967) describes divergent thinking as thinking in various directions in order to arrive at alternative solutions to a problem. He contrasted it to convergent thinking; thinking logically to arrive at one correct solution. Both types are cognitive operators in Guilford's Structure of the Intellect (SI) model, presented as a three-dimensional cube-like 'space' in which operators form one dimension, with contents and products being the other two. Combining the numbers of operators (5), contents (4) and products (6) allows the model to identify a total of 120 (= 5 x 4 x 6) intellectual abilities, each making a cell in the three- dimensional space. Among the abilities relating to divergent thinking are fluency, the ability to produce many solutions/ideas to a problem; flexibility, the ability to generate different types of solutions/ideas; and originality, the ability to generate rare and uncommon solution/ideas. Guilford devised tests to measure creativity and used statistical factor analysis to investigate his model. An example of a 'classic' task testing divergent thinking is the 'brick problem': 'List all sensible ways you may make use of a brick.' (Guilford 1967). Many creativity tests have been developed from Guilford's theories (see Sternberg & Lubart, 1999; Sternberg & O'Hara, 1999) with Torrance's Tests of Creative Thinking (Torrance, 1990) being perhaps the most well-known.

Guilford's theory has been adopted for science education by several workers (e.g. Diakidoy & Constantiou, 2000-2001; Endean & George, 1982; Felder, 1988; Schlichter, 1983). Diakidoy and Constantinou (2000-2001) explored the context-dependency of creativity among university physics students who were asked to generate as many responses as possible to three physics tasks each designed to be 'open-ended', that is, having more than one solution. The tasks were scored according to Guilford's divergent thinking abilities: 'fluency' was a count of the number of solutions and 'flexibility' was the number of different types of solutions. 'Originality' was calculated on a sliding scale: responses given by fewer than 5% of the students scored 3, fewer than 15% scored 2 and fewer than 50% scored 1. In addition, the authors measured students' domain-specific knowledge and gathered information about their grades. They found that creativity varied with the tasks; it was context-dependent, but did not correlate strongly with subject matter knowledge.

Hu and Adey's (2002) work has generated great interest due to their attempt to address the specific issue of scientific creativity. In developing a test for scientific creativity for secondary science, they copied Guilford by constructing the Scientific Structure Creativity Model (SSCM) as a definition for creativity. Similar to Guilford's SI-model, the SSCM model comprises three dimensions in a 'space' of factors. These are: process, comprising imagination and thinking; trait, comprising fluency, flexibility and originality; and product, having the components technical product, scientific knowledge, science phenomena and science problem. In total the model offers 24 cells or 'factors'. Hu and Adey initially designed two items for each cell. These were reduced later to seven in total, each measuring a core element of scientific creativity, namely; unusual use, problem finding, product improvement, creative imagination, problem solving, science experiment and product design. The items are clearly recognisable as the Guilford/Torrance-type, but comprise problems set in scientific contexts. For example, the 'unusual use' item is the 'brick problem', cited above, set in a science context: 'Please write down as many as possible scientific uses as you can for a piece of glass. For example, make a test tube.' (ibid.: 394). Scoring procedures follow the Guilford/Torrance tradition in a similar way to that demonstrated by Diakidoy and Constantiou (2000-2001). The test showed high internal consistency and inter-scorer reliability when tested on a sample of 160 UK students.

Two later studies used Hu and Adey's scientific creativity test, helping to elaborate the meaning of their approach to creativity. Lin, Hu, Adey and Shen (2003), tested the influence of the Cognitive Acceleration through Science Education (case) programme (Adey, Shayer & Yates, 1989) on students' creativity with a sample of 11- 15 year old Chinese students. The study revealed that those who had been taught case material increased their creativity test score on 5 out of 7 items, indicating a strong link between creativity and other intellectual abilities. The case material is designed to increase students' sci\entific reasoning through specially designed learning tasks, focusing mainly on analytical skills rather than creativity. We may therefore conclude either that the Hu and Adey test measures analytical thinking, or that analytical and creative cognitive abilities correlate. Alternatively, as Lin et al. (2003) suggest, case teaching may enhance students' metacognition (their reflection on their own thinking), thus aiding students' creative thinking. The second study, Weiping, Adey, Shen and Lin (2004), applied Hu and Adey's test to compare Chinese and British teenagers' creativity. Data patterns obtained in this study confirm similarities between the science creativity test and earlier Guilford/Torrance tests. Torrance (1962) showed that creativity does not develop evenly through adolescence, but in stages, with stagnation occurring at certain points. Weiping et al.(2004) found similar results using the science creativity test in the UK and China, although the levelling off occurred at age 14 rather than among the 13 year olds found by Torrance. Further results from these studies will be discussed later. For the moment, we note the consistent and coherent results obtained by the Hu-Adey test when used with a single national sample and when comparing samples between two countries.

Guilford's and Torrance's work on creativity has received criticism. Mansfield and Busse (1981) find fault with the strong overlap with intelligence and dislike paper-and-pencil tests measuring 'creativity on request'. Sternberg and O'Hara (1999) claim the psychometric approach to creativity has lost its appeal, because tests seem only weakly to relate to other ratings of creativity and measure trivial aspects of the phenomenon. These critiques may apply to Hu and Adey's approach, but their test does seem to correspond to science educators' understanding of scientific creativity. They subjected their items to validity testing by an 'expert-panel' including thirty-five teachers and science educators from the UK and China. Two of the seven items had less than 50% support for measuring scientific creativity, but overall, support was strong. The 'product design' item 'Please design an apple picking machine. Draw a picture, point out the name and function of each part' received strongest support and was the task most students found interesting.

The cognitive approach to scientific creativity

Cognitive approaches seek to understand the mental processes underlying creative reasoning rather than just identifying particular intellectual skills (Sternberg & Lubart, 1999). These are gradually replacing psychometrics in creativity research. Research methods utilise human subjects and computer simulations (Boden, 1999; Simon, 1995). The general emphasis of the research is that creative thinking is normal, and not a distinct domain of activity (Ward, Smith & Finke, 1999), meaning that the mind continually uses creative processes in everyday thought. An example is language use, where we apply a limited set of rules in flexible ways to create meaning. Creative reasoning, in all forms, may therefore be traced back to a fundamental set of common generative processes. These include recalling structures from memory; the formation of associations among structures; mental synthesis of new structures; mental transformation of existing structures into new forms; and analogical transfer of information from one domain to another. These processes also account for thinking in extremely creative people, who exhibit enhanced intensity of application of the processes; greater richness or flexibility of stored cognitive structures to which the processes are applied, and higher memory capacity. Several models and theories attempt to explain how such 'ordinary' cognitive processes can work together in creative reasoning (see Sternberg & Lubart, 1999; Ryhammar & Brolin, 1999.)

The cognitive approach suggests that teaching may be adapted to develop creative reasoning patterns, but focusing on such general skills may at first sight suggest we are taking a wrong path in terms of teaching scientific creativity. The situation is akin to the 'science skills' and 'science process' trends of the 1960s- 1980s, when a series of reasoning skills, like 'classifying', 'hypothesising' and 'interpreting', were presented as representing scientific thinking, being useful in people's ordinary life and teachable through science courses (Gagne, 1965: 65-8). This approach received strong criticism on the grounds that such skills, even if used by scientists, are not science-specific and develop regardless of science teaching (Millar & Driver, 1987). The characteristics of scientific reasoning, and therefore what teaching should focus on, were instead thought to be the type of problem, knowledge and contexts in which these skills are used.

In spite of this critique, reasoning patterns still exist that most science educators agree should be taught; for example, learning to 'control variables' and 'co-ordinate theory and evidence' (Driver, Newton & Osborne, 2000; Gott & Duggan, 1996, among others). These are generally agreed upon as important aspects of 'science', but also recognised as general cognitive abilities with importance (Kuhn, Amstel & O'Loughlin, 1988; Piaget, 1952). Learning science obviously includes cognitive development beyond factual and procedural knowledge. The problem might therefore be what reasoning skills should be taught and how, rather than if we should develop them or not.

An answer may be found in the long list of reasoning patterns identified by studies in the field of psychology of science (Feist & Gorman, 1998). For example, scientists typically form abstract representations (Chi, Feltovich & Glaser, 1981); work on problems in a forward, abstract manner rather than backward and concretely (as novices often do) (Larkin, 1983); use 'confirm early-disconfirm late' heuristics (Mynatt, Doherty & Tweney, 1977) and are able to think simultaneously in two problem spaces (that is, reflect on possible alternatives) (Klahr, Dunbar & Fay, 1990). Worth noticing though, is that the most general attribute of scientists' thinking seems to be the complexity (Feist & Gorman, 1998). Expert scientists are characterised by complex cognitive networks of knowledge (Gruber, 1989) and by being complex thinkers:

The simple thinker makes relatively few qualifications and sees things in black and white terms. In contrast, the complex thinker not only makes distinctions and qualifications, but integrates into a synthetic whole the opposing points of view (Feist & Gorman, 1998: 29).

Compulsory science students in school are 'simple thinkers', so may tend to use any of the above processes differently from scientists. The initial conclusion to the issue of finding creative reasoning skills associated with science is that we should respect, not simplify differences between real science and school science, and between scientists and student. This raises the fascinating possibility that students could become better creative thinkers by teaching them to think the same way as scientists, but in saying this we underestimate strongly science's complexity. Hence, when identifying any skill as useful or important we need arguments and evidence from school science and real science. The two rational skills mentioned above, controlling variables and coordinating theory and evidence, for example, have achieved status in science education not just because they are scientific processes, but because of their intrinsic value for science education. On this basis, the paper will explore imagination and thinking through analogies as a particular area of cognition associated with creativity. These skills are central to scientists, who aim to look beyond the surface level of objects and phenomena (Dunbar, 1995, 1999; Holyoak & Thagard, 1995). They may also be important to school sciencea topic worth some curriculum developers' discussion time.

Imagination

Imagination is often described as 'seeing with the mind's eye'. As a concept, the term has general and narrow meanings. The general meaning refers to making up or thinking about fictional situations or worlds, such as fairy tales and fiction novels. These emerge from fantasy, but are based on real world knowledge and experiences. This makes imagination a general tool humans use to reflect on situations that are beyond their experience. Rugg (1963) and Holton (1978, 1998), for example, study scientists' imaginations when developing new ideas and theories, while Harries (2000) and Vygotsky (2004) study children's use of imagination in role-play and pretend play.

The narrow meaning of imagination operates in the same context but more specifically means the mind's ability to create and explore mental images of situations that are not physically present (Block, 1981; Kosslyn, 1994, 1999). In this context, psychologists often refer to more specific concepts such as imaging, imagery and visual- spatial thinking. Other terms such as 'auditory imagery' (Reisberg, 1992) and 'kinaesthetic (or motor) imagery' (Jeannerod, 1994) have been coined to demonstrate that 'image' is not necessarily a 'picture' but all types of 'quasi-sensory experiences' (Richardson, 1969). For the purposes of this paper, only picture-like images are considered, although we will move between general and narrow meanings of imagination, reflecting on creativity in the light of the role imagination plays, and may play, in science education.

We begin this sub-section with a basic description of mental images presented in cognitive psychology. Many cognitive psychologists argue against a strong emphasis on a pictorial association of mental entities, claiming the focus should be on propositional representations underlying the experience of mental images. Others claim pictorial representations are unique mental entities that are treated differently from other types of infor\mation. This debate, known as the 'imagery debate', has run since the 1950s (see Kosslyn, 1999 for an account). While the existence of mental images is not disputed, lack of agreement persists over their nature and the roles they play in information processing. Kosslyn defends the pictorial view, comparing the purpose of imagination to ordinary vision (Kosslyn, 1980, 1999) used to recognise objects and for 'spatial processing', that is, tracking moving objects and navigating when moving. Similarly, we may recall images from memory when needing to recognise an object, mentally moving the object or ourselves in an imagined situation. Mental imagery has many similarities with visual perception: recent brain- scan research shows that these activities engage overlapping parts of the brain (Kosslyn, 1999). The parallel between imagination and vision can only go so far. When a situation is imagined, this includes conceptual understanding and not just visual (or other sense-type) memories. Images can't be regarded as 'true' pictures of the world any more than other individual understandings, but are ideas in picture-like form constructed for specific cognitive purposes. Mervis and Rosch (1981) claim people form 'prototype' images; for example, we imagine 'bird' as an interpreted prototype rather than a particular bird we have seen. Some images are naturally more realistic than others. Rugg (1963), for this reason, differentiates between 'reproductive imagination' and 'creative imagination', the first being simple perceptual memories while the latter is formed using memories derived from external objects. Kosslyn (1999) identifies four different imagery abilities describing more fully how humans process mental images: image generation, image inspection, image maintenance (retaining images of previously considered items) and image transformation (mentally moving the items around). Research suggests that people differ substantially on these abilities and that they involve combinations of mechanisms located in various parts of the brain (Kosslyn, 1980).

Harris (2000) presents a very different route to understanding imagination. His and others' research on young children's pretend play and role play shows that imagination holds a central place in cognitive development and learning. Earlier research on children dismissed the importance of imagination. Piaget (1962), for example, built on a Freudian (Freud, 1961) understanding interpreting children's imagination used in play as egocentric thinking. In contrast, Harris shows that imagination is a child's way of actively exploring and learning to operate in the outer world. He reasons that children uphold the same logic and causal principles in their imagined worlds as in the real world. For example, when role- playing a tea party children transfer real world attributes to the toys-for example, The tea is hot so needs to be drunk carefully'. Children are normally aware of differences between what is real and imagined, so can reflect on two 'worlds' in a meta-perspective. This creates distances between the actual experiences and permits logical reflection on premises, causes and effects. When role-playing, children are observed going beyond copying their observation of adults. Instead, observations from everyday routines provide material for their imagination. These are deconstructed and analysed into component features, played with and put together in new combinations. In adult life, this deconstruction and reassembly ability helps humans solve problems, predict dangerous situations or simply avoid social embarrassment. Harris also underlines the importance of imagination in children's play as a way of developing their empathetic and social skills.

Obviously, imagery and imagination are important skills for scientists. When developing new theories they use the ability to imagine and visualise physical phenomena and 'play' with possible outcomes. Examples include simple analogies, as when Einstein, while working out the general theory of relativity, imagined what it would be like to ride on a ray of light and Faraday visualised electro- magnetic field lines. Psychologists (for example, Rothenberg, 1979a, b; Simonton, 1988), historians and philosophers (Holton, 1996, 1998) have studied numerous such examples. The outcomes, however, do not give straightforward pictures of how scientists use mental images in creative thinking (LeBoutillier & Marks, 2003). Manipulation of mental images may be used, as described by Rothenberg's (1996) 'Janusian imagery' in which contrasting images are compared and explored side by side. 'Thought experiments' are also carried out- these are discussed later. Elsewhere, scientific creativity might result from an unconscious, vivid flow of images (Suler, 1980), as in Kekule's 'discovery' of the benzene-ring prompted by a daydream of a snake biting its own tail (Holyoak & Thagard, 1995). Scientists are frequently reported as having ideas and images while in a dream- like state (Rothenberg, 1996; Shepard, 1978). The common feature seems to be an ability to operate in an imaginative visual-spatial mode of thinking. Shea, Lubinski and Benbow (2001) indicate that this is characteristic of scientists in general, not just scientific 'geniuses'. Their longitudinal study tested talented 12-14 year olds spatial visualisation (the ability to visualise concrete objects and manipulate these visualisations, 605) and verbal abilities. The study aimed to evaluate how these tests predict educational and vocational outcomes later in life. Follow-up questionnaires distributed to a sample of 331 individuals at ages 18, 23 and 33 led to this very clear conclusion:

Intellectually talented adolescents with stronger (visual) spatial ability relative to verbal ability were more likely to be found in engineering and computer science-mathematics fields. (611)

Gardner (1993) supports this conclusion claiming that spatial ability is a key factor in determining 'how far one will progress in science' (192).

Two points emerge from the discussion thus far: first, imagination and the particular skill of imagery are important learning tools; second, these tools are important for those entering science. An emerging question is: what roles do these abilities play in science learning in school? Does imagery help younger students learn basic science principles? Perhaps surprisingly, little research addresses these questions directly. Mathewson (1999, 2005) claims imagery is crucial. He argues that: 1If visual-spatial cognition is fundamental in science, it should be important for the successful teaching in science' (39).

Mathewson uses Holton's (1996) work as a theoretical rationale to classify scientists' imagination as 'visual' (similar to imagery described above), 'metaphoric' (using analogies) and 'thematic'. This last type refers to 'unifying ideas' that scientists apply across many phenomena, such as symmetry, conservation, stability and system. Mathewson combines these types of imagination into master images that he believes should be taught to give better understanding of science. 'Circuits' is an example of a master image, meaning closed paths, loops and networks, which can be applied to topics like capillaries, circulation systems and electronics (Mathewson, 1999: 40). He suggests students are taught 'visual-spatial self-awareness' and 'meta-cognitive visual skills' using specially developed activities and encouraged to draw their perceptions of natural objects and events. He also encourages practical work and use of physical materials. While interesting, Mathewson's research does not include empirical evidence that this approach enhances science learning.

Indirect evidence supporting Mathewson's claim, if not his solution, can be seen in other areas of science education research. Work on 'thought experiments' (Ireson, 2005; Reiner & Burko, 2003; Lattery, 2001; Gilbert & Reiner, 2000; Reiner & Gilbert, 2000) is an example. This concept originates in studies of how scientists use their imaginations to run 'experiments' including problems, hypotheses, 'results', discussions and conclusions (for example, Sorensen, 1992; Horowitz, Tamara & Massey, 1991). Gilbert and Reiner (2000) show this phenomenon is common practice in science education, even if the 'experiments' are often turned into 'simulations' because students are given insufficient time to reflect on problems themselves. Students are often expected to reason within an imagined world of invisible particles and phenomena. Reiner and Gilbert (2000) explain conceptual learning from this reasoning as an effect of combining tacit knowledge and logical processes: students use their visual and bodily experiences as a basis for imagining invisible phenomena, helping them generate new states of knowing. Evidence for this comes from their study of university physics students who were video taped while running discussions about thought experiments.

Research on analogies (for example, Bryce & MacMillian, 2005; Heywood, 2002; Pittman, 1999), metaphors (Tobin & Tippins, 1996) and models (Clement, 2000; Gilbert, 2004; Greca & Moreira, 2000; Harrison & Treagust, 2000) provide further information about the role of imagery in science learning. A majority of these studies explored analogies, metaphors and models provided by teachers rather than investigating these techniques in promoting students' own mental imagery skills. Two studies, however, used student-generated analogies in science contexts. These support a positive learning effect arising from developing imagery skills. Spier-Dance et al. (2005) found positive learning effects when undergraduate students were asked to develop individual analogies for halogen oxidation and discuss these in group and class sessions. Interestingly, lower- achieving students seemed to benefit most. We cannot determine whether the ability to generate analogies (and thereforeimagery) or the subsequent structured discussion contributed most to this change. In the second imagery-related study, Pittman (1999) found positive connections between 14-year-old American students' abilities to produce their own analogies in drawing tests and their scores on a written test. Students' analogies were varied, drawing heavily on their everyday experiences and interests. The study also revealed that the ability to produce analogies related strongly to students' confidence and self esteem. Taken together, the two studies indicate students' mental imagery abilities are important for science learning, but that their 'images' or analogies must be subjected to critical analysis in order to contribute positively to learning (Franco et al. 1999; Greca & Moreira, 2000; Harrison & Treagust, 1996).

A different approach to answering the question whether imagery and imagination help students develop science knowledge is found in Hadzigeorgiou (2005). He argues for the value of narrative thinking (Bruner, 1986) and romantic understanding (Egan, 1990). These concepts suggest that scientists do not only think along paradigmatic (logico-mathematical) lines but rely on complementary alternatives. Scientists often make use of everyday analogies and develop personal, affective and aesthetic understanding (Dawkins, 1998; Root-Bernstein, 1997). Hadizigeorgiou's point is that science education neglects these human elements in science and by so doing ignores a useful approach for science learning. He wants science teaching to capture students' imagination, for example by presenting ideas that conflict with everyday common-sense, such as, 'a spaceship may travel at thousands of miles per second in the absence of an external force' or encouraging students to create their own analogies for science phenomena and ideas. The argument is not primarily that these elements are a first step towards correct conceptual change, but that they contribute towards more personalised involvement in science (our italics). Girod, Rau and Schepige (2003), Ieong (2004) and Linden (2005) present similar arguments. In their defence, Girod, Rau and Schepige (2003) offer results from a study of 9 and 10 year old students encouraged to produce imagined stories on science topics. Interviews show that these stories apply as personal explorations of subject matter knowledge and ways of linking this to everyday experiences.

Our analysis of imagination demonstrates high relevance to both our real and school science worlds. Just like scientists, students, it seems, benefit from being able to create images of 'hidden' science phenomena like atoms, field lines and eco-systems, and mentally manipulate these. The benefits may include the personalised relationship and romantic understanding students develop alongside any actual conceptual learning. Encouraging students to work with images also demonstrates important ideas about science. However, the most challenging perspective is that people able to operate in imagined worlds seem to be more creative and enjoy benefits in learning in many areas of life (Claxton, 1999; Finke, 1989; Johnson- Laired, 1983; Kosslyn, 1994; Nersessian, 1999). Although evidence that imagination in science education transfers to other areas is not convincing and needs far more exploration, this perspective is important to the educational challenges of our time. Imagination offers the promise of making scientific creativity more concrete and helping to identify a potential starting point for further research.

SUMMARISING PERSPECTIVES AND SETTING CHALLENGES

In this third section we aim to summarise the perspectives on creativity in science education and set challenges showing how science education may develop to enable the field to contribute to developing students' creativity.

The first two sections yield, as we hinted at the outset, three frames within which 'scientific creativity' is interpreted in different ways. Each has its own set of educational aims and objectives. The term 'scientific creativity' is defined loosely or more tightly in each frame depending on the context. We call the first of these the creative teaching frame, drawing mainly on our sections headed creative teaching, art and science and, to some extent, inquiry science. A range of possibilities is featured; from generally student-oriented, to working with open-ended tasks and doing scientific investigations. The aims of these may generally be described as enhancing science learning through stimulating students' interest and excitement. Much current fashionable practice in introducing 'creativity' into science classrooms is represented here, mainly concerned with engaging science teachers in the development of their classroom skills and practice. Practitioners and researchers working in this frame may not necessarily agree that they are inculcating 'scientific creativity'hence here we see the term defined loosely. This leads to the main problem with the 'creative teaching' approach: the obvious lack of guiding theoretical perspectives, using 'creativity' or 'creative' to label or support any practice that may be justifiably called 'good' teaching. Clearly, given our earlier statements, much, if not all, of the work done in this interpretation falls short of our own beliefs about scientific creativity. This is not to say these efforts are not valuable or lacking in positive outcomes as examples of 'good' teaching and/or cross-curricular links. This frame provides pertinent and clear illustration that the current position on creativity in science education is not 'scientific creativity' as it could and, we suggest, should be.

A second frame, teaching about scientific creativity, is drawn from the work we report on inquiry science and the nature of science. This focuses more distinctly on one aim-to help students understand how science researchers work creatively to develop new theories. This is taken seriously by educators involved with teaching about the nature of science, but statements about how science is creative are kept at a very simple level. Hence, we find the vie

Source: Studies in Science Education

More News in this Category