INT. J. SCI. EDUC., 23 JANUARY 2004, VOL. 26, NO.

1, 85–99

RESEARCH REPORT

Making pupils’ conceptions of electricity more durable

by means of situated metacognition

Petros Georghiades, Ministry of Education and Culture of Cyprus; e-mail:

petros.georghiades@cytanet.com.cy

This paper argues for the integration of metacognitive thinking into the content of school science. It describes
the ‘metacognitive instances’ approach of implanting metacognitive activities, such as classroom discussions,
annotated drawing, keeping diary-like notes and concept-mapping, at selected points of the teaching procedure,
in an attempt to extend the durability of children’s science understanding, which is often short-lived. The
approach was tested with 60 Year 5 pupils in Cyprus studying ‘Current Electricity’, following a quasi-
experimental design with one experimental and one comparative group. Following the teaching of the unit,
participants were tested on their understanding of electricity three times in one school year. The research
showed that the practice of metacognition is feasible with primary school pupils and that children who
experienced situated metacognition retained taught concepts for a longer period of time. Although no significant
difference was recorded initially in the amount of subject matter learned by the two groups, children who
engaged in metacognitive activities appeared to have achieved more permanent re-structuring of their
understandings.

Introduction
In the context of organized systems of education, the impact of time on children’s
learning is extremely important because the progressive structuring of knowledge
and understanding is gradual. For teaching and learning to be successful we expect
learners not only to acquire new knowledge to sufficient depth, but also to retain
this knowledge for long periods of time after instruction. The implication of such an
assumption is that among the qualities that should characterize school-learned
knowledge lies durability. Arguments supporting such a view come mainly from two
directions. First, prevailing among the aims pursued by schooling and education, is
the need to prepare future citizens to use the skills and knowledge accumulated over
their school lives in real-life settings. In order for knowledge to be potentially used,
it has to be durable. Second, we want learning to be durable, in the sense of
surviving periods of disuse, in the post-training environment itself, if further
learning is to be based on existing structures (Bjork 1996). This is particularly
important in the context of science education that has come to rely heavily on
learning by conceptual change (Driver 1989, White and Gunstone 1989). The
structuring of the learning process in conceptual change that pre-supposes the
existence of a less inclusive or partly correct concept, which will eventually be
changed into a more complex concept, makes the importance of durable concepts
evident. There is currently ample research evidence demonstrating that pupils often

International Journal of Science Education ISSN 0950–0963 print/ISSN 1464–5289 online © 2004 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/0950069032000070333
86 P. GEORGHIADES

forget what they learn at school a short time after instruction (Department of
Education and Science 1985, Novak 1998), a problem that is often overlooked in
the literature on conceptual change. In their attempts to devise approaches that will
extend the lifespan of children’s learned knowledge, many researchers suggested
that metacognitive thinking is contributing to this (Hacker et al. 1998, Georghiades
2001).

Metacognition: general versus situated

Metacognition is defined as “knowledge and cognition about cognitive phenomena”
(Flavell 1979: 906). It is usually referred to in the literature as the executive
decision-making process in which individuals both perform cognitive operations
and oversee their progress (Meichenbaum et al. 1985). Learners who reflect
consciously on what they know about a given task, who describe what and why they
are currently doing, can talk about their feelings about the learning situation they
find themselves in, and can use this information to monitor and enhance their
performance, are said to be engaging in metacognitive thinking.
Over the past three decades a strong case has been made for the importance
of metacognition. It has been argued, for example: that the ability to think
reflectively is what drastically differentiates human learning from animal learning
(Karmiloff-Smith 1991); that metacognition can act as a bridge between different
areas such as decision-making and memory, and learning and cognitive develop-
ment (Nelson and Narens 1996); and that metacognitive thinking is an important
dimension of problem-solving (Gardner 1991), and, more specifically, of an
individual’s ability to identify, represent, plan and evaluate within a given problem
(Davidson et al. 1996). A growing group of researchers ascribe an explicitly
educational flavour to the importance of metacognition, suggesting that it can
have a positive impact on students’ learning and on their school performance
(Hacker et al. 1998, Georghiades 2000). This has resulted in metacognition
enjoying increased attention in a plethora of projects aiming to help children to
‘learn how to think’.
Most attempts to introduce a metacognitive dimension in children’s school
thinking can be traced in programmes that attempted to teach general thinking skills
(Baird and Mitchell 1986, Adey et al. 1989). In spite of positive outcomes from
such studies being numerously reported in the literature, an increasing number of
scholars assume a trend in the literature towards embedding instruction within
specific areas of the curriculum rather than isolating thinking skills as separate
topics (Halpern 1992, Hennessy 1993). In this way, Perkins and Salomon (1989)
moved a step further, to forecast that wider-scale efforts to join subject-matter
instruction and the teaching of thinking will be an exciting dimension of future
research and educational innovation.
In this paper, I make the distinction between general and situated metacogni-
tion. General metacognition is the kind of metacognition practised in most of the
studies aimed at improving children’s general thinking skills (for example, Baird
and Mitchell 1986, Adey et al. 1989). Such improvement was often ‘context free’
or accommodated in an artificial context unrelated to ongoing teaching, hence the
anticipation of transfer of any improved cognitive ability to other contexts.
Metacognition was explicitly taught and in most cases special time was allocated
for this purpose. Situated metacognition, on the other hand, is defined as
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 87

Table 1. Comparative summary of differences between general and

situated metacognition.

General metacognition Situated metacognition

Overall objective To improve general thinking To facilitate better

Anticipated improvement
Context/content
Not related to subject taught
Mode of instruction
Time of intervention
skills across contexts/
different contents
Enhance ability to transfer
thinking skills in different
contexts
General or ‘neutral’
Within subject taught
Explicitly taught
Specially allocated time
understanding of specific
subject content
Enhance performance in
taught content. (e.g. longer
durability and ability to
utilize in different contexts)
Specific
‘Blended’ with normal
classroom activities
Normal teaching time

metacognition practised in the current context of normal lessons and within the
time allocated for the teaching of curriculum subject matter, aimed at improving
learners’ performance in the specific content taught by facilitating better under-
standing. The core of the distinction between general and situated metacognition
is presented in the comparison made in table 1.
The fact that situated metacognition is not aiming directly at the improvement
of general thinking skills should not be seen as a criterion for placing it at a
hierarchically lower position to that of general metacognition. Put differently,
situated metacognition is not to be regarded as an inferior ‘variety’ of metacogni-
tion, rather it is argued as one that portrays a more ‘grounded’ approach with
differentiated focus. In essence it is an explicit attempt to achieve a balance between
subject matter knowledge and strategies for effective learning. Metacognitive
activity is used as a means of facilitating conscious learning and awareness of how
learners’ understanding progresses in order to achieve better understanding of
taught material. The effect of situated metacognition is therefore traced in the
learner’s performance in the content along which metacognition was employed,
rather than in the form of improving general thinking skills.
Employing metacognitive thinking in a way that promotes the blend of
content and reflection of the learning process is supported by research evidence
and strong arguments in the literature. Gunstone and Northfield (1994), for
instance, suggested that the development of metacognitive skills and knowledge
should not be separated from real learning tasks. Elsewhere, Gunstone (1994) has
argued for the importance of specific science content in the enhancement of
metacognition, basing his claim on work with science graduates in a 1-year pre-
service high school science teacher education programme. Boaler (1993) pre-
sented research outcomes according to which children from a learning environ-
ment characterized by the complete integration of process and content performed
better than children from an environment that concentrated mainly on content
88 P. GEORGHIADES

when encountering mathematical problems in different contexts. Referring to

work designed to test the effect of content instruction and general strategy
instruction, Garner and Alexander (1989) reported that students who are weak in
both content and strategic knowledge do not benefit from either form of
instruction. One of the explanations they gave for this finding is that ‘strategy
instruction may only work for students who have a certain amount or certain
kinds of content knowledge’ (Garner and Alexander 1989: 151), hence offering
further argumentation in favour of blending content with process.
Contrary to work previously conducted in the field, the research presented in
this paper did not attempt to ‘generally promote metacognition’ (Baird and Mitchell
1986) nor did it advocate the teaching of metacognition for the improvement of
general thinking skills. The scope was one of investigating the feasibility of practising
situated metacognition with primary school children and examining any positive
impact of such activity on the durability of their conceptions of science. This basic
difference in scope was the determining factor in not following the approach of
other researchers in the field who have incorporated metacognition in their studies.
Adey et al. (1991), for example, who worked with secondary school children,
followed an approach that involved substituting ‘normal’ science lessons with
‘thinking science’ lessons once every 2 weeks. These lessons were spread over a
2-year period and were based on specially designed material with activities that were
targeted towards encouraging students ‘thinking about their own thinking’. The
material used was clearly process oriented and there was apparently no effort to
accommodate metacognitive (or other) activities within the science content taught
during ‘normal’ science lessons. The rationale behind the ‘metacognitive instances’
approach, on the other hand, is that if it is to find its way into ordinary classrooms
on a wide scale and be realistically applicable in practice, metacognition should not
be promoted at the expense of content, but the two should be intertwined. By
incorporating metacognition alongside normal teaching activities, this paper argues
for a place for metacognition in everyday teaching and questions approaches
treating it as something ‘special’ only to be taking place during specially allocated
time.

Description of the ‘metacognitive instances’ approach

Aims, methods and samples
The research was conducted with 60 Year 5 pupils (11 year olds) in public primary
schools in Cyprus studying the unit ‘Current Electricity’. The aim of the research
was to study the role of situated metacognition in the durability of primary pupils’
conceptions of science, in response to the common problems of pupils forgetting
what they learn a short time after instruction (Department of Education and
Science 1985). The pupils were divided into two classes of 30 children following a
quasi-experimental design with an experimental group and a comparative group.
The two groups were ‘matched’ on the basis of their average scores in a cognitive
development test and a test on Year 4 science. Metacognitive instruction was
incorporated alongside ‘ordinary’ teaching structures, following the existing science
curriculum and without the allocation of extra teaching time or the substitution of
‘normal’ science lessons with ‘special’ lessons. The teaching sequence followed was
identical for both the comparative and experimental groups, the only difference
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 89

being the implanting of metacognitive instruction at selected points of the teaching

procedure in the case of the latter group. On average, five or six metacognitive
activities were employed during each 80-minute lesson, the duration of which varied
up to a maximum of 6–7 minutes.
Pupils’ performance was repeatedly assessed by means of the same written test
1 week (Phase 1), 2 months (Phase 2) and 8 months (Phase 3) after completion of
the unit, which was covered in four 80-minute lessons, once a week. The test
addressed the five concepts taught, namely: (a) ‘electrons – current flow’, (b)
‘conductors and insulators’, (c) ‘open–closed circuits/switches’, (d) ‘burnt-out light
bulbs’ and (e) ‘short circuit’. Each concept was tested three times by means of
exercises with different contextual characteristics: (a) exercises requiring context-
free recall of knowledge, (b) exercises placed in contexts that resembled the
experiments performed in the classroom, and (c) exercises placed in ‘unfamiliar’
contexts that referred to real-life problems. The test did not include any exercises
that were in any way similar to the activities practised only by the experimental
group, such as concept mapping or annotated drawing.

Metacognitive activities
Among the growing literature on theoretical as well as practical aspects of
metacognition in relation to the field of education has been the appearance of ‘lists’
of metacognitive activities for teaching purposes (for example, Novak 1985, Baird
and Mitchell 1986). Four different metacognitive activities were used in this
research, namely: (a) classroom discussion, (b) keeping a diary, (c) annotated
drawing, and (d) concept mapping (Georghiades 2002). The primary scope of the
present research was to investigate whether the practice of metacognition could have
any positive impact at all on children’s science learning. The question was initiated
by the debate in the literature as to whether metacognition as a higher-order
thinking skill is or is not available to primary school children, and by research
showing that boys of 11 years of age, or younger, did not benefit from intervention
programmes that made use of metacognitive activities (Adey et al. 1991). Choosing
and incorporating a set of activities advocated as being the ‘best’ or ‘most
appropriate’ ones would be taking for granted that metacognitive instruction does
have a positive impact with young children and that the scope was to enhance that
impact. The fact that this would be overlooking a question of high significance
contributed to the decision to employ a number of different activities without any a
priori claims regarding their success. When just one metacognitive activity is
systematically used, not only does it become unpopular with the students, but it is
also treated in a mechanistic way without really causing any reflection (Baird and
Mitchell 1986). My intention, therefore, was to have a variety of activities, so that
the children maintained high interest in the process and were exposed to different
stimuli. In summary, the idea behind the selection of these activities was to have
simple, easy to carry out, activities that could blend with the rest of the teaching
sequence, without taking considerable time for training children in carrying them
out. By trying to blend metacognition with the rest of the teaching activities and by
not presenting these as ‘new’ activities, the scope was also to avoid appearance of the
well known ‘Hawthorne effect’; that is, the condition whereby any improvement in
performance can be attributed to the novelty of new treatment or conditions,
irrespective of their nature.
90 P. GEORGHIADES

The timing of metacognitive activities in the teaching procedure was largely

pre-specified in planning the lessons, although on occasions it took place in a
more spontaneous way. The use of ‘metacognitive instances’ did not follow a
strictly pre-determined course or sequence; rather, it was adjusted to the specific
circumstances of each lesson. Although all four activities were used in all lessons,
this was not done in a ‘routine-like’ way. However, some features of the
approach were inevitably repeated in subsequent lessons. It was more likely, for
instance, to use classroom discussion early in a lesson and to reserve the use of
concept mapping for subsequent stages of the lesson. The only pre-specified task
was the writing of diaries that was assigned as homework, or as a last activity
for those pupils who finished their work early. At the beginning of the next
lesson children were encouraged to share what they had written with their
classmates. The following sections present samples of children’s reactions to
metacognitive stimuli during the implementation of the metacognitive instances
approach.

Children’s engagement in metacognitive activities

Classroom discussion
Classroom discussion was the metacognition-promoting means employed most
frequently during the course of the research. This related to the ease with which
discussion can be triggered, its independence from any materials, and the
catholic way in which children can potentially participate. ‘What is the purpose
of this activity?’, ‘What are you asked to do?’, ‘Have you learned something from
this activity?’, ‘What was the information you had that helped you reach this
conclusion?’, or ‘Explain to your friend the way you solved that problem’, are
some of the prompts used for triggering reflective thinking. As expected, some
children engaged more fully in the discussions than others who were less willing
to reveal their thoughts and ideas, or to attempt any kind of analysis that would
potentially add to the complexity of school science. The great number of pupils
in the class made the involvement of all children in the discussions a difficult
task to accomplish. On the other hand, the fact that some pupils did not seem
to participate was laden with ambiguity as to whether they benefited from these
discussions, since metacognition is a highly personal and internal mental activity.
The following are a few characteristic quotes from classroom discussions:
I knew that it was dangerous to switch on the lights with my hands wet, but I didn’t know
why. I didn’t know that water has got electrons.
I used to think that the lamp glows instantly because current is very fast. Now I know that
electrons move just like the seats of a ski lift.
First I said that the good lamps would glow but when I did the experiment I saw that they
didn’t. It’s because the burnt-out lamp breaks the circuit.
I will never forget that pencil tip (graphite) lets electricity pass through it. It was so
strange!
I no longer think that a burnt-out lamp can glow a little because in the experiment it didn’t
glow at all.
I liked the experiment with the wire-wool burning. It helps me remember how short-circuit
can set fire.
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 91

The pupils were encouraged to compare prior and current beliefs and to consider
different potential explanations to the phenomena observed, something that
explains the strong comparative elements in their remarks. They also made
judgements on the examples or analogies used by the teacher and the experiments
performed in class, specifying what was in their opinion successful or not. Many of
them could justify their views or choices by referring to recently constructed
knowledge.
In general, classroom discussions managed to engage the majority of the pupils
in metacognitive reflection of varied degrees, which, considering the young age of
the children and their unfamiliarity with metacognitive activity, is judged to have
been successful. Engagement in metacognitive reflection is thought to have acted in
a twofold way. It helped pupils who had grasped intended meanings correctly to
reflect on their understanding and become more conscious of their learning. It
offered those pupils who held partial or inaccurate understandings of the
phenomena studied the chance to re-consider their views towards the scientific
explanation. Cases of children admitting ‘Now I see!’, following these discussions,
contributed to the shaping of such an impression.

Diaries
Diaries were used as a means of engaging shy and introverted children, who were
not willing to share their thoughts or queries with their classmates, in metacognitive
thinking, and one way of promoting children’s private reflection. In keeping diary-
like notes as a metacognitive activity, sheets were used rather than an exercise book
(pupils were asked to write on the back of the worksheets used in class). This way
the task was given an informal character and there was a better chance of pupils
remembering to bring these sheets with them for the next lesson rather than a
booklet to which they were not used in their science lessons. The use of diaries,
however, was not a problem-free choice. Not all pupils remembered to do the task
when this was assigned as homework, and in most of the cases their writing was
restricted to one or two sentences. Moreover, most of their remarks were primarily
descriptive of work done in class, rather than projecting a reflective nature, and a
tendency to be rather generic in their thoughts was evident. The contribution of the
use of diaries partly lay in the highlighting of points that were not clearly understood
by children at the end of the lesson, giving me the chance to re-visit briefly these
areas in the following lesson, before addressing new concepts.
In those cases where some reflection was evident, children’s comments were,
once again, of a comparative nature, usually comparing their ideas prior and
following teaching. In the two following diary extracts, the children demonstrate
some reflection on learned material, contrasting new knowledge to their under-
standing prior to teaching.
I knew that ‘short circuit’ can be caused by water on wires. Now I know that it can also be
caused when two wires touch together because electrons always choose the easiest route.
(Gregory, age 11)
When we started learning about open and closed circuit I thought it was very confusing
because at home we say ‘open the light’ and ‘close the light’, but here we learned about ‘open
circuit’ and ‘closed circuit’. I think that the example with the bridge that rises to let boats
through will help me remember that when the switch is open, electrons stop like cars do.
(Angela, age 11)
92 P. GEORGHIADES

Notably, the second extract shows what appears as substantial change in Angela’s
understanding, whereas the case in the first extract is one of addition of new
information without rejecting that already held by Gregory. Of course, these
statements do not themselves furnish sufficient evidence of learners truly holding
changed understanding, yet they are an indication in this direction. It is also
noted that in the second extract Angela referred to the use of analogies during
teaching (Georghiades 1999), acknowledging the contribution of these analogies
in changing her understanding, therefore exhibiting reflection on the means of her
learning.
Keeping diary-like notes has, in general, proved rather problematic, the
majority of the pupils restricting their writing to brief descriptive statements. In
those cases where more extensive notes were made, there was evident re-visiting of
knowledge, expressed by means of comparisons between pupils’ prior and post-
teaching understanding, and reflection on the methods or examples used by the
teacher. The latter cases suggest that keeping diary-like notes can be a successful
metacognitive activity with some children, once they acquire some experience in
carrying it out.

Annotated drawing
Annotated drawing proved to be the most enjoyable activity for the children who,
in general, appeared very willing to be assigned this task. The children were asked
to draw a simple arrangement in relation to a given concept, and produce brief
annotations to explain their drawings. In order to achieve involvement of more
pupils in the activity, including children who might not be willing to draw
something (because they felt that the quality of their drawing was poor) and
children who were very slow in producing a drawing, I also used ready-drawn
figures that children had to annotate. The level of reflection involved in their
annotations varied considerably, from being merely descriptive of the object they
drew, to explaining and justifying the why of things. In one case, for instance, one
pupil’s drawing was plainly descriptive of a plug without showing any reflection on
behalf of the learner as to why different parts of the plug are made from different
materials. A different child, on the other hand, who was describing the same
object, clearly showed in her drawing that she engaged in re-visiting her
knowledge at a level higher than the descriptive (figure 1). She managed to
describe the materials from which the plug is made, she explained why these
materials are used in particular and, in doing so, she used the right scientific
terminology. She even moved a step further to explain what can possibly happen
if the plastic cover of the wire is removed, exhibiting hypothetical thinking in
relation to her understanding.
Figure 2 presents a rather unusual case. When asked to draw a short circuit and
briefly explain what happens, one pupil constructed his own multiple-choice
question. This is thought to be an important case of metacognitive thinking in
action. The pupil not only produced a drawing that represented correctly the
arrangement for a short circuit to happen, but he further presented the scientifically
correct and a wrong response to choose from. This suggests that this pupil was
conscious of different potential explanations of short circuit and was able to choose
the correct response by revisiting his knowledge, as suggested by the arrows he used
to represent electron flow.
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 93

Figure 1. Sample of annotated drawing.

Figure 2. Sample of annotated drawing.

94 P. GEORGHIADES

Figure 3. Sample of concept mapping.

Overall, children produced rich material by means of annotated drawings. Even

in their most basic form of being merely descriptive, annotated drawings were useful
in enabling pupils to represent the knowledge they were holding. In their more
advanced forms, children’s drawings entailed signs of reflection on their under-
standings, proving them to be a successful metacognitive tool.

Concept mapping
Concept mapping was used initially as a whole-class activity and later as a pair or
individual activity, aimed at summarizing taught concepts and highlighting
relationships between them. The children had never used concept mapping before,
nor had they received any substantial training during the research on the
construction of concept maps, other than a few examples on unrelated topics.
During the first lesson the children were given the concepts they were expected to
use on cards, and were asked to place them on paper, draw links and label them. In
the remaining lessons the children were only given a ‘key concept’ and were then
asked to draw their concept maps based on everything they knew in relation to the
given concept. The complexity of the concept maps produced by the pupils varied
considerably. On one hand were cases of pupils restricting their concept maps to
basic connections or relationships between learned concepts, and on the other hand
were cases of pupils presenting more complex and detailed associations.
The concept maps produced at the end of the first lesson that introduced
children to the concepts of open and closed circuit were very simple yet managed to
represent correctly basic relationships and connections between taught concepts of
electricity. The concept map in figure 3 was produced during the second lesson
following learning about conductors and insulators. This concept map accom-
modates both knowledge acquired on that specific day (conductors and insulators)
and knowledge obtained in the previous lesson (electrons/open–closed circuit). This
is important because in producing this concept map the pupil had to re-visit
previously held knowledge and relate this to new information in a way that the two
co-exist in a mutually supportive manner. This kind of mental activity is taken as a
sign of thinking that entails metacognitive reflection.
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 95

Figure 4. Sample of concept mapping.

During the last lesson of the unit, concept maps of greater detail and complexity
were produced. This can be attributed to the fact that pupils learned a number of
concepts or phenomena associated with electricity and had, in the meantime,
acquired some experience in constructing concept maps. The concept map in figure
4 clearly presents more concepts and more relationships, showing ability on behalf
of the learner to depict a system of inter-related pieces of information (e.g. electron
flow, role of the battery, open–closed circuit, conductors and insulators, switches
and burnt-out light bulbs).
In general, most of the concept maps produced by the children of the
experimental group entailed some signs of metacognition. This judgement is based
on the fact that children gradually managed to simultaneously represent numerous
concepts and identify relationships and connections between them, something that
could not possibly be achieved if the pupils had not exhibited reflection on their
constructed knowledge. Had the pupils received further training in concept
mapping, this would have possibly resulted in the production of even richer concept
maps.

Results and discussion

The results presented in table 2 show that scores achieved by pupils of the
experimental class were consistently higher than those obtained by their counter-
parts of the comparative class, in the three tests administered during the course of
96 P. GEORGHIADES

Table 2. Test results in the three phases of the research.

Comparative group Experimental group

Degrees
Standard Standard of
Phasea % Mean deviation % Mean deviation t freedom p

1 68.7 41.2 11.3 71.6 43.0 9.8 0.655 58 0.515

2 64.5 38.7 12.3 70.2 42.1 12.4 1.043 56 0.302
3 62.9 37.8 13.0 72.9 43.8 9.6 2.018 57 0.048*

a
Phase 1, test conducted 1 week after the completion of the unit; Phase 2, test conducted 2 months later; Phase
3, test conducted 8 months later.
* Significant at the 0.05 level.

the research. In the case of the comparative class there is a descending tendency in
overall performance over the three testing instances, whereas for the experimental
class the picture is one of a more consistent (even increasing) performance.
Differences recorded between the two groups were initially small (Phase 1, p =
0.515), only to increase over the three phases of the research in favour of the
experimental group (Phase 3, p = 0.048). This suggests that although children from
the experimental group did not construct significantly greater amounts of subject
knowledge, compared with the performance of the comparative group, they more
successfully accommodated the ideas of current electricity into their long-term
memories and experienced more permanent re-structuring of their understandings.
This is an important finding already reported in the metacognitive literature (Blank
2000) and one that strengthens the case in favour of introducing metacognitive
activities in school science. Due to the small number of participants, however, the
outcomes of the present research will inevitably have to be treated as ‘signs’ rather
than as ‘evidence’ of situated metacognition having an effect. More confident claims
can only derive from longitudinal research with larger samples.
What can be claimed with confidence is that situated metacognition is feasible
with young children, albeit difficult and demanding on both teacher and pupils.
Children’s responses, questions, and ideas expressed during classroom discussions
and the material produced by the pupils of the experimental group during the
implementation of the ‘metacognitive instances’ approach demonstrated that most
children of this age are capable of reflecting upon their learning, to a varying extent
and within the limitations of their mental and cognitive development. The majority
of the children successfully re-visited their knowledge, made comparisons between
their understandings of electricity prior to and following teaching, traced
connections and relationships between taught concepts, and made judgements on
activities and examples used in class. Achieving this kind of learning behaviour was
neither easy nor problem free, not least because of the difficulty in engaging all
children in metacognitive thinking. My attempts were often hindered by children’s
varying levels of cognitive maturation, their limited concentration span, and their
reluctance to acknowledge misconceptions or weaknesses in their learning.
Consequently, the implementation of the metacognitive activities required sub-
stantial support, continuous feedback and encouragement, both on the class level
and the individual learners’ level.
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 97

Due to the personal mental mechanisms involved in the learning process and,
even more so, in engaging in metacognitive activity, identifying in detail the way
in which the practice of situated metacognition may have affected the durability of
pupils’ understanding is difficult. One plausible assumption is that metacognition
acted as the ‘marking event’ on children’s learning. By engaging in metacognitive
activities, pupils’ conceptions of electricity were ‘marked’ and distinguished in
their memories from other pieces of knowledge that were not highlighted in a
similar reflective way. For example, encouraging metacognitive reflection on
relationships characterizing learned material, by means of concept mapping,
helped children to relate their newly constructed understanding to knowledge
they already held, consciously identify links between concepts, and to obtain a
more global perception of the unit taught. Arguably, this kind of learning
behaviour resulted in more meaningful, solid and, consequently, more durable
conceptions of science. The fact that children from the experimental group
demonstrated generally more durable ‘correct’ science, without reverting to
previously held understandings, could also be suggesting that metacognition acts
in the direction of preventing the re-appearance of alternative explanatory
frameworks (Driver et al. 1994). Children’s engagement in more conscious
learning may have contributed towards the construction of new concepts that
proved intelligible, plausible and fruitful (Strike and Posner 1985) to the learners
for a longer period of time, hence eliminating the need for regression to non-
scientific constructs. Notably, any attempt to interpret the effects of metacogni-
tion on human learning can only be partly successful, as the mechanisms of
metacognition have not, to date, been fully decoded.

Conclusions
This research has shown that the practice of metacognition is feasible with Year 5
pupils, hence joining forces with scholars who previously advocated this potential
(for example, Adey et al. 2002, Gunstone 1994, Rudd 1992), and those who
supported the early involvement of young children in reflective or philosophical
thinking (Costello 2000, Fisher 1995, Lipman 1985). This conclusion is partic-
ularly important when viewed in relation to the bulk of existing literature, which
selectively focused on secondary school or adult populations when it came to
studies on metacognition. Primary ages were either ignored or excluded as unable
to engage in metacognitive thinking, or to benefit from such activity.
Situated metacognition has given first signs of a positive impact on the
durability of children’s science learning. Although metacognitive thinking does not
necessarily lead to greater amounts of knowledge, it seems it can have a positive
impact on the strength or depth of children’s constructed conceptions. The
repetitive character of the outcomes over the three phases of this research, and the
significant difference in group performance in Phase 3, in particular, indicate that
this impact is durable over a long period of time. Research with greater numbers of
participants is needed before any definite conclusions are reached on the effect of
situated metacognition on children’s science learning. In considering directions for
future research, it is essential that exposure to metacognitive activities is prolonged,
by employing such activities across different units in science. It is possible that
metacognitive experience ‘builds up’ with time, in the sense that learners gradually
use numerous metacognitive techniques in different settings, and the positive
98 P. GEORGHIADES

impact of this activity aggregates. Consequently, the case could be made for a dual
positive outcome from the use of situated metacognition, both short term and long
term. Short term because the positive impact of metacognitive activity may reflect
on children’s understanding of taught material, hence maintaining better perform-
ance in the subject-area covered. Long-term impact, on the other hand, could stem
from accumulated experience of metacognitive activity gained by the pupils as a
result of their systematic ‘enculturation’ (Perkins 1993) in metacognition throughout
their school year(s). More longitudinal research that will explore these potentials is
therefore called for.
In shaping this new perspective to the practice of metacognition, it is by no
means suggested that situated metacognition should substitute any attempts to
teach metacognition explicitly. What is proposed is that situated metacognition can
offer an approach to the practice of metacognition with a potentially positive long-
term impact on pupils’ understanding, that can be applicable in everyday teaching
without allocating special time or the need for specially devised resources. Situated
metacognition does not escape the difficulties or criticism associated with more
traditional forms of metacognition, yet it can be a realistic alternative in teachers’
attempts to achieve a balance between subject matter knowledge and strategies for
effective learning.

References
ADEY, P., ROBERTSON A. and VENVILLE G. (2002). Effects of a cognitive acceleration programme
on Year 1 pupils. British Journal of Educational Psychology, 72(1), 1–25.
ADEY, P., SHAYER, M. and YATES, C. (1989). Cognitive acceleration: the effects of two years of
intervention in science classes. In P. Adey, J. Bliss, J. Head, and M. Shayer (Eds.), Adolescent
Development and School Science (London: Falmer Press), 240–247.
ADEY, P., SHAYER, M. and YATES, C. (1991). Better learning. A report from the Cognitive
Acceleration through Science Education (CASE) project (London: Kings College, University
of London).
BAIRD, J. R. and MITCHELL, I. J. (Eds.) (1986). Improving the quality of teaching and learning: an
Australian case study – The PEEL Project (Melbourne: Monash University).
BJORK, R. A. (1996). Memory and metamemory considerations in the training of human beings.
In J. Metcalfe and A. P. Shimamura (Eds.), Metacognition (Cambridge, MA: MIT Press),
185–205.
BLANK, L. M. (2000). A metacognitive learning cycle: a better warranty for student under-
standing? Science Education, 84(4), 486–506.
BOALER, J. (1993). Encouraging the transfer of ‘school’ mathematics to the ‘real world’ through the
integration of process and content, context and culture. Educational Studies in Mathematics,
25, 341–373.
COSTELLO, P. J. M. (2000). Thinking skills and early childhood education (London: David Fulton).
DAVIDSON, J. E., DEUSER, R. and STERNBERG, R. J. (1996). The role of metacognition in problem
solving. In J. Metcalfe and A. P. Shimamura (Eds.), Metacognition (Cambridge, MA: MIT
Press), 207–226.
DEPARTMENT OF EDUCATION AND SCIENCE (1985). Science at Age 15 – science report for teachers no.
5 (London: HMSO).
DRIVER, R. (1989). Changing conceptions. In P. Adey, J. Bliss, J. Head, and M. Shayer (Eds.),
Adolescent Development and School Science (London: Falmer Press), 79–99.
DRIVER, R., SQUIRES, A., RUSHWORTH, P. and WOOD-ROBINSON, V. (1994). Making sense of
secondary science (London: Routledge).
FISHER, R. (1995). Teaching children to think (Cheltenham: Stanley Thornes).
FLAVELL, J. H. (1979). Metacognition and cognitive monitoring: a new area of cognitive
developmental inquiry. American Psychologist, 34, 906–911.
GARDNER, H. (1991). The unschooled mind (New York: Harper and Row).
CONCEPTIONS OF ELECTRICITY BY METACOGNITION 99

GARNER, R. and ALEXANDER, P. A. (1989). Metacognition: answered and unanswered questions.

Educational Psychologist, 24(2), 143–158.
GEORGHIADES, P (1999). Take your class skiing! Primary Science Review, 58, 15–16.
GEORGHIADES, P. (2000). Beyond conceptual change learning in science education: focusing on
transfer, durability and metacognition. Educational Research, 42(2), 119–139.
GEORGHIADES, P. (2001). Dimensions of meta-‘Conceptual Change Learning’ in science education: the
role of metacognition in the durability and contextual use of pupils’ conceptions. Unpublished PhD
thesis, Roehampton, University of Surrey.
GEORGHIADES, P. (2002). Making children’s scientific ideas more durable. Primary Science Review,
74, 24–27.
GUNSTONE, R. F. (1994). The importance of specific science content in the enhancement of
metacognition. In P. Fensham, R. F. Gunstone, and R. White (Eds.), The content of science
(London: Falmer Press), 131–146.
GUNSTONE, R. F. and NORTHFIELD, J. (1994). Metacognition and learning to teach. International
Journal of Science Education, 16(5), 523–537.
HACKER, D. J., DUNLOSKY, J. and GRAESSER, A. C. (Eds.) (1998). Metacognition in educational
theory and practice (Mahwah, NJ: Lawrence Erlbaum).
HALPERN, D. F. (1992). Enhancing thinking skills in the sciences and mathematics (Hillsdale, NJ:
Lawrence Erlbaum).
HENNESSY, S. (1993). Situated cognition and cognitive apprenticeship: implications for classroom
learning. Studies in Science Education, 22, 1–41.
KARMILOFF-SMITH, A. (1991). Beyond modularity: innate constraints and developmental change.
In S. Carey and R. Gelman (Eds.), The epigenesis of mind (Hillsdale, NJ: Lawrence
Erlbaum).
LIPMAN, M. (1985). Thinking skills fostered by philosophy for children. In J. Segal, S. Chipman,
and R. Glaser (Eds.), Thinking and learning skills 1 (Hillsdale, NJ: Lawrence Erlbaum),
83–108.
MEICHENBAUM, D., BURLAND, S., GRUSON, L. and CAMERON, R. (1985). Metacognitive
assessment. In S. R. Yussen (Ed.), The growth of reflection in children (New York: Academic
Press), 3–35.
NELSON, T. O. and L. NARENS (1996). Why investigate metacognition? In J. Metcalfe and A. P.
Shimamura (Eds.), Metacognition (Cambridge, MA: MIT Press), 1–25.
NOVAK, J. D. (1985). Metalearning and metaknowledge strategies to help students learn how to
learn. In L. H. T. West and A. L. Pines (Eds.), Cognitive structure and conceptual change
(London: Academic Press), 189–209.
NOVAK, J. D. (1998). Learning, creating and using knowledge (Mahwah, NJ: Lawrence Erlbaum).
PERKINS, D. N. (1993). Creating a culture of thinking. Educational Leadership, 51(3), 98–99.
PERKINS, D. N. and SALOMON, G. (1989). Are cognitive skills context-bound? Educational
Researcher, 18(1), 16–25.
RUDD, T. J. (1992). Exploring self assessment with primary school students. Unpublished M.Ed.
Studies project, Faculty of Education, Monash University.
STRIKE, K. A. and POSNER, G. J. (1985). A conceptual change view of learning and understanding.
In L. H. T. West and A. L. Pines (Eds.), Cognitive structure and conceptual change (London:
Academic Press), 211–231.
WHITE, R. T. and GUNSTONE, R. F. (1989). Metalearning and conceptual change. International
Journal of Science Education, Special Issue, 11, 577–586.