A Developmental Model Of Critical Thinking

A Developmental Model of Critical Thinking
DEANNA KUHN
The critical thinking movement, it is suggested, has much to gain
from conceptualizing its subject matter in a developmental frame-
work. Most instructional programs designed to teach critical
thinking do not draw on contemporary empirical research in cog-
nitive development as a potential resource. The developmental
model ofcritical thinking outlined here derivesfrom contemporary
empirical research on directions and processes of intellectual de-
velopment in children and adolescents. It identifies threeforms of
second-order cognition (meta-knowing)--metacognitive, metas-
trategic, and epistemological--that constitute an essential part of
what develops cognitively to make critical thinking possible.
Educational Researcher, Vol. 28, No. 2, pp. 16-26, 46
E
nthusiasm for critical thinking as a goal of education
shows no signs of waning. Pluralist conceptions of
what education should be achieving--the demand
for accountability by diverse constituencies--make critical
thinking ever more attractive as a unifying objective of
every student's education. Many of the functions education
performs--making the discriminations that regulate access
to occupations, serving special needs of subgroups--rest on
human differences. Developing the competencies that en-
able people to participate fully as citizens in a democracy
remains the unifying purpose, and great promise, of public
education.
Indeed, one would be hard pressed to construct a serious
counterargument, to claim that we wouldn't like to see stu-
dents become careful, rigorous thinkers as an outcome of
the education we provide them. Yet as a serious social goal,
education for thinking is hardly trivial. By most accounts,
we remain far from achieving it. A question we might be
asking at this point, then, is whether progress is being made
in establishing a knowledge base that will support its real-
ization. Are educational researchers making reasonable
progress not only in determining how to teach critical
thinking but, even more fundamentally, in establishing ex-
actly what it entails?
In making advances regarding these tasks, especially the
latter, one might anticipate that educators would rely
heavily on a knowledge base provided by researchers who
study cognitive development. In fact, however, at least
until very recently, the burgeoning critical thinking move-
ment in education has proceeded with little apparent con-
tribution from contemporary cognitive development re-
search. If educators draw on empirical data at all, efforts to
define critical thinking skills have tended to be based on
an older psychometric literature on abilities (see Ferrari &
Sternberg, 1998, for a recent review). Nor, traditionally,
have educators looked very much to research on mecha-
nisms of cognitive development for ideas regarding how
to foster these skills.
Why has an enterprise held in such wide esteem and of
such significance to society not been guided by systematic
and extensive empirical research on the nature of thinking
and its development? A possible explanation is that rele-
vant research has been either nonexistent or untranslatable
into practice. Even if correct, this explanation, it is signifi-
cant to note, does not apply across the board to other cur-
riculum areas of concern to educators, such as reading and
mathematics. To the contrary, research in the development
of mathematical understanding and reading skills has ex-
panded enormously in recent years, with researchers pay-
ing increasing attention to the relevance of their findings for
practice.
A possible explanation for the difference is that, unlike
,reading or mathematical skills, thinking does not fall within
Sthe purview of developmental psychology because it is not
a developmental phenomenon. In other words, it does not
evidence strong age-related emergence of a sequence of
competencies of the sort that researchers have identified in
mathematics or reading. Rather, we would like thinking at
any age to be sound and rigorous and free of fallacies, and
developmental psychologists have no particular expertise
to offer in finding ways to get students to think more care-
fully and rigorously.
In this essay, I argue against such a view, claiming instead
that developmental phenomena are currently being studied
that are of direct relevance to understanding and fostering
critical thinking. Furthermore, I claim, the critical thinking
movement has much to gain from a developmental con-
ceptualization of its subject rnatter--a perspective that has
been largely absent--and it has a growing knowledge base
of cognitive development research to draw on in this re-
gard. Moreover, a developmental conceptualization has im-
plications for what is perhaps the most pressing practical
issue in current efforts to teach critical thinking, the fact that
gains most often do not generalize beyond the immediate
instructional context.
Do We (Really) Know What Critical Thinking Is?
What does it mean to be accomplished in thinking, and how
does this accomplishment manifest itself? What are the
component skills or states of partial accomplishment on
which a teacher might focus, and how can progress in these
DEANNA KUHN is a professor of psychology and education at
Teachers College, Columbia University, Box 119, New York, NY
10027. Her specialty is cognitive development and its implica-
tionsfor education.
16 EDUCATIONAL RESEARCHER

respects be gauged? The ready base of empirical knowledge
that educators can turn to to address these questions in the
case of reading or mathematics is largely absent when edu-
cators turn their attention to thinking. Nor has the recent
history, of cognitive development research made things any
easier for them. Among trends in the field have been, on one
hand, a growing attention to biological endowment and
how it constrains development to favor certain forms and
domains over others (Gelman & Williams, 1998; Spelke &
Newport, 1998) and, on the other, an increasing recognition
of the role of culture (Rogoff, 1998). Both, in their respective
ways, emphasize specificity at the expense of broad lines of
cognitive development of the sort that would be of particu-
lar relevance to educators.
In apparent compatibility with this zeitgeist, most edu-
cators today favor the view that critical thinking skills must
be taught in the context of specific subject matter (Perkins
& Salomon, 1989). Countering an historical tradition of
learning research employing minimal, arbitrary, and even
meaningless content, Brown (1997) elaborates this position,
claiming that we cannot expect children to progress in the
development of thinking unless we give them something to
think about, in other words, unless we engage them in seri-
ous learning about meaningful, rich, domain-specific sub-
ject matter of the sort contained in the traditional school
curriculum.
Although controversy, continues to exist in the education
literature, it is not my purpose here to debate claims about
how best to teach critical thinking nor even to review the
relevant evidence Instead my ur ose in this essav is to
• , p . ~ . . . . . . . . . . . . . . . . . .
argue that whatever the p0s~iTi0none takes regarding effec-
tive instructional techniques, it remains a crucial task to de-
~ne thinking skills in a way that has generality at least
across some broad range of content, is informed by, empiri-
ca| data, and is situated in a developmental framework of
whereparticular cognitive skills come from and where they
are headed. None Of the psychological or educational liter-
ature I have alluded to here, it is worth noting, is incom-
patible with such an undertaking•
I make the case for the importance of this task on two
grounds--conceptual and practicall Conceptually, it is es-
sential that we know precisely what we mean when we
refer to critical thinking or thinking skills, if the constructs
are to be useful• Brown (1997), for example, while arguing
for the need to embed the teaching of thinking skills in rich
subject matter instruction, advocates the importance of chil-
dren attaining "flexible learning and inquiry strategies of
wide applicability" (p. 399). The very use of a term such as
"inquiry strategies" implies that such strategies are not
taken to be entirely domain specific in a conceptual sense--
i.e., defined in a way, that confines them to a particular body,
of content. The extent of their empirical (as opposed to con-
ceptual) cross-domain generality--i.e., the extent to which
their application in one context is predictive of their appli-
cation in another--then becomes a matter for empirical in-
vestigation. Establishing the two kinds of generality, it
should be noted, is a sequential undertaking. Until a cogni-
tive skill is defined conceptually across a range of contexts,
one cannot investigate its empirical generality across these
contexts.
Theorists and researchers need to know exactly what we
mean by such things as "inquiry strategies" if the constructs
are to be useful to them. So, too, do practitioners. If we ex-
pect teachers to venture beyond the well-specified goals
that exist within traditional subject matter areas and em-
brace thinking skills as educational goals in a serious and
committed way, we must aid them to envision these skills
in a way, that would make them concrete realities rather
than vague abstractions. For all of the educational literature
aimed at enhancing teachers' awareness of the importance
of fostering thinking skills in their students, teachers have
been offered remarkably little in the way of concrete exam-
ples of what these skills are--what forms the}, take, how
they, will know when they see them, how they might be
measured. If such skills are really important, a teacher
might well infer, I would have a clearer picture of what they
are, and I would know how to identify them. Also, this hy-
pothetical teacher might infer, I would see more indication
that they are valued in the school community; the}, would
be regularly assessed in my students as a matter of school
polic}; as are the "basic" skills on which evaluation of
teacher performance increasingly, has come to rest. What is
a perceptive teacher to conclude when developing thinking
skills is what gets talked about in abstract terms as a wor-
thy goal and other competencies are the ones repeatedly
measured in his or her classroom?
Critical Thinking As Metacognition
In the major portion of this essay, I present one effortmmy
ow;n--to identify specific intellectual skills that might serve
as;h focus of middle- and secondary-school educators' ef-
fdrts to enhance the thinking of their students. A distinctive
characteristic of this effort is that it draws directly on con-
temporary empirical research on cognitive developnrent
from late childhood through adolescence and into adult-
hood (Moshman, 1998)• Indeed, the central thesis of this
essay is that empirical data regarding the directions and
paths in which children's and adolescents' thinking devel-
ops stand to inform educators' discussions regarding criti-
cal thinking.
A second distinctive characteristic of the present effort is
that the developing cognitive competencies I describe as
most relevant to critical thinking are m~itive--r~ather
than cognitive--competencies. In contrast to ,f,i__rst-.0_r~ier
cogn!tive skills that enable one to know about the world,
metacognitive skills are second-order meta:kllowing skills
that entail knowing about one's ow n (and others') know-
ing. O]son and Astington (1993) are the only other re-
searchers studying cognitive development to my knowl-
edge to explicitly link metacognition or meta-knowing to
critical thinking.
It should be noted, however, that a concept like metacog-
nition, even if it does not go by this specific nafne, is by no
means new to the philosophical literature on critical think-
ing. Indeed, something like metacognition figures in the defI
initions of critical thinking proposed by most educational
philosophers who have addressed the topic. It would be a
digression from the purposes of the present essay to review
this vast literature in any detail. Briefly, however, it is worth
noting that although educational philosophers have voiced
disagreement with one another regarding how critical think-
ing should be defined, their respective definitions have
tended to contain some element ef what heads Paul's (1990)
list of multiple facets that define critical thinking: "the art of
thinking about your thinking" (p. 32). Paul (1990) is critical
of the definitions of other educational philosophers--for ex-
MARCH 1999 17

ample, Ennis' (1987; Norris & Ennis, 1989) definition of crit-
ical thinking as "reasonable and reflective thinking [con-
cerned with what to do or believe]" (Norris & Ennis, 1989,
p. 3) or Siegel's (1988, p. 2) definition as thinking "appropri-
ately moved by reasons"---on the grounds that they rest on
concepts such as reasonableness or reflectivity that are not
themselves well defined. Although Paul remains critical of
it as well, Lipman's (1991) definition of critical thinking as
thinking that "can be assessed by appeal to criteria" (p. 116)
fares better in this respect. The evaluation of thinking by ap-
peal to criteria implicates metacognition: Thinking neces-
sarily becomes an object of cognition (just as it does in Ennis'
definition, in which one is reflecting on what one should be-
lieve). Lipman (1991) makes the interesting point, however,
that it is possible to think about one's thinking in a com-
pletely uncritical manner. Hence, he adds to his definition
the further stipulation that in order to qualify as critical
thinking, metacognition must be "self-correcting."
Educational philosophers have also concerned them-
selves with the question identified earlier--are critical
thinking skills domain specific or domain general?--and
come down variously on the domain-specific (McPeck,
1981) and domain-general (Paul, 1990) sides (although their
debates of this issue tend not to have been influenced by
empirical evidence). In contrast, an aspect of critical think-
ing that has received relatively little attention from educa-
tional philosophers is its developmental dimension. One
important exception to this generalization, however, is the
work of the educational philosopher who perhaps had the~
most to teach us about critical thinking--Dewey.
Dewey; with his deep respect for and involvement in the
empirical science of psycholog}; did not share the concep-
tion that prevails today of a dichotomy between the scien-
tific study of human development and the practical world
of children in schools. Instead, he saw schools as laborato-
ries of human development, as experiments in the possibil-
ities of human development in arranged environments. Re-
peatedly in his writings, Dewey made clear that the goal of
education could only be development (or what he called
"growth"). Education "means supplying the conditions
which foster growth" (Dewey, 1916, p. 56), not toward a
predetermined end but rather in the direction of "an in-
crease in the range and complexity of situations to which
the child is capable of applying reasoned inquiry" (Cahan,
1994, p. 158). Dewey also made it clear that he saw the ed-
ucator's task as a process of connecting with the young
child's interests and purposes, but that one could not stop
there. "The real problem of intellectual education," he said,
"is the transformation of more or less casual curiosity and
sporadic suggestion into attitudes of alert, cautious, and
thorough inquiry" (Dewey, 1933, p. 181).
Realizing Dewey's vision, I would claim, can be fur-
thered by knowledge regarding how children's intellects
develop--the directions, the processes, and the states of
partial accomplishment (Haith & Benson, 1998) that mark
this evolution. The empirical data of researchers who study
these processes have a central role to play in this endeavor.
The origins of my own concern with the development of
meta-knowing skills--the intellectual skills most closely as-
sociated with critical thinking---lie in microgenetic studies
of the strategies children or adults employ in coordinating
existing understandings with new evidence (Kuhn, 1989;
Kuhn, Garcia-Mila, Zohar, & Andersen, 1995; Kuhn,
Schauble, & Garcia-Mila, 1992; Schauble, 1996). It is through
such coordination processes that knowledge is acquired.
While much developmental research in recent years has
been devoted to detailed description of the evolution of the-
oretical knowledge within specific domains (Wellman &
Gelman, 1998), much less attention has been given to the
process of theory-evidence coordination that drives this
knowledge evolution. I have proposed that a critical change
that occurs with development is the attainment of increas-
ing control over this process (Kuhn, 1989). Because it has to
do with awareness, understanding, and management of
one's cognition (in contrast to simply its execution), this is
a meta-knowing attainment. Research I describe in this
essay supports the claim that this control increases both
with age, in cross-sectional data, and over time, in micro-
genetic data.
Traditionally in developmental psychology, thinking
about thinking has been most closely associated with
Piaget's stage of formal operations (Inhelder & Piaget,
1958). Contrary to Piaget's model, however, subsequent re-
search indicates that the second-order "operations on oper-
ations" that define his formal operational stage do not
emerge in any singular, abrupt, or decisive fashion (Mosh-
man, 1998). In this respect, Piaget's developmental claims
appear to have been too strong. In another respect, how-
ever, his developmental claims may have been too modest,
for the ability to think about one's own thought has sweep-
ing implications that extend far beyond the narrow contexts
of scientific inquiry about physical phenomena that In-
helder and Piaget (1958) examined. Thinking about one's
thought--in contrast to simply engaging in it--opens up a
whole new plane of cognitive operations that do not exist at
a simple first-order level of cognition.
Although the divisions are not rigid and aspects of all
three appear in many instances of meta-knowing, I discuss
meta-knowing here in three broad categories--metastrate-
gic, metacognitive, and epistemologicaI (Kuhn, in press). The
distinction between metastrategic and metacognitive
knowing rests on a widely employed dichotomy in cogni-
tive psychology (as well as in philosophy) between proce-
dural knowing (knowing how) and declarative knowing
(knowing that). Meta-knowing differs depending on the
kind of first-order knowing that is its object. Procedural or
strategic knowing entails the exercise of strategies to
achieve goals, thus invoking the potential for a second-
order metastrategic form of knowing that selects and moni-
tors the strategies that are applied--a manager of the reper-
tory of available strategies. Metacognitive knowing operates
on one's base of declarative knowledge, which also stands
to benefit from executive management. What do I know,
and how do I know it? Finally, epistemological knowing has
to do with an individual's broader understanding of knowl-
edge and knowing. It has both a general, philosophical as-
pect-How does anyone know?--and a personal aspect--
What do I know about my own knowing? One might be
tempted to dismiss epistemological understanding as of
marginal interest, simply a curiosity reflecting children's or
lay adults' naive efforts to grapple with philosophical con-
cepts. As I shall argue to the contrary, however, it is of crit-
ical importance, in part because of its influence on the other
two components. First, however, because it is a develop-
mental story I wish to tell, I begin at the beginning, with a
look at the developmental origins of meta-knowing.

Developmental Origins of Meta-Knowing
Metacognitive Knowing
Somewhere in the age range of 3 to 5 years--the exact
age being a matter of debate--children acquire the insight
that assertions are expressions of someone's belief (Olson &
Astington, 1993). It is an attainment, I will claim, that is a
critical marker in the development of meta-knowing and
lays an essential foundation for critical thinking. It is also
the cuhnination of its own set of developmental precursors.
By the time they achieve this insight, children will have de-
veloped the ability to represent mental states; they distin-
guish, for example, thinking about a dog from actually per-
ceiving one (Estes, Welhnan, & Woolley, 1989). They also
have begun to use mental-state concepts such as desire and
intention as a means of explaining their own and others' be-
havior (Wellman & Gelman, 1998). In these very simple re-
spects, even very young children are able to think about
thinking as a human activity that they and others engage in.
And yet, prior to the attainment of the insight I highlight
here, young children remain curiously limited in their con-
ceptualization of the products of this thinking--the beliefs
people hold and the assertions they make that express these
beliefs. A good deal of evidence now exists that children
below the age of about 4 regard the universe of assertions
that people make, and the beliefs that these assertions rep-
resent, as isomorphic to an external reality. An account of an
event differs from the event itself only in that one exists on
a representational plane whereas the other is perceived di-
rectly. In other words, the world is a simple one in which
things happen and we can tell about them. There are no in-
accurate renderings of events.
A major source of evidence for this characterization of
young children's understanding is their poor performance
in simple tasks involving the concept of false belief. In the
now-classic experimental paradigm from which a burgeon-
ing wave of research on "theory of mind" derives, 3-year-
olds believe that a newcomer will share his or her own ac-
curate knowledge that a candy container in truth holds
pencils (Perner, 1991; Wimmer & Perner, 1983). It is impos-
sible that the other person could hold a belief that the child
knows to be false. And children may even deny that they
earlier had professed this false belief before being shown
the contents.
Underlying these claims is the naive epistemological the-
ory that beliefs are nothing more or less than mental copies
of reality. The implications of this "realist" epistemological
stance for critical thinking are profound, rendering it, in ef-
fect, moot. If we accept a very broad and general definition
of critical thinking as the evaluation of assertions (Olson &
Astington, 1993), the realist stance lends little sense to such
an activity. If assertions merely duplicate and reflect reality,
they do not need to be evaluated. They simply exist, on a
plane parallel to physical reality, as representations of the
way things are.
Recognition of assertions as belief states, in contrast, im-
plies the understanding that they may conflict with a phys-
ical reality. As a result, their truth status becomes suscepti-
ble to evaluation. Perner (1991) characterizes this
acquisition as the ability to metarepresent--i.e., to mentally
model the human representational function. We do not
know why this achievement occurs at exactly this time, but
apparently the child accrues enough experience with
human knowing to link the products of knowing---beliefs
and assertions--to the generative process that gives rise to
them. Their source is a human representational system that
is recognized as having intent and volition to represent
what it wishes to represent. Accordingly, an assertion's clos-
est link becomes to this source--the subject who is doing
the representing, rather than the object being represented.
The kinds of evaluation that recognition of assertions as
belief states makes possible are rudimentary consisting of
no more than comparing an assertion to reality and pro-
nouncing it true or false. Yet this achievement has two im-
portant implications. One is the implication of at least some
distinction between an assertion and external evidence beac-
ing on it; assertion and evidence must exist as distinct enti-
ties if one is to be evaluated in light of the other. Second, the
status of evidence and assertion as distinct entities makes it
possible for relations to be constructed between them. Para-
doxically, achievement in the second respect (establishment
of correspondences between assertion and evidence) out-
strips achievement in the first (maintenance of the distinc-
tion between the two). Young children, for example, readily
draw on covariation evidence to make causal inferences
(Shultz & Mendelson, 1975). (Noncovariation as evidence
for noncausality turns out to be more difficult.) Moreover, in
very simple contexts, they can identify such correspon-
delhces even when the assertion is contrary to their own be-
li4'fs (Ruffman, Perner, Olson, & Doherty, 1993).
More problematic, it turns out, is maintenance of firm dif-
ferentiation between evidence bearing on an assertion and
the assertion itself. The reason may be that making the as-
sociation between an assertion and the evidence that sup-
ports it remains in the realm of first-order, strategic cogni-
tion--e.g., observing a character eat green food supports
the assertion that she likes green food (Ru ffman et al., 1993).
Differentiating assertion and evidence as belonging to dif-
ferent epistemological categories, in contrast, is an achieve-
ment that entails meta-knowing. At stake are the questions,
"What do I know?" and "How do I know it?"
Several studies have documented the difficulty pre-
schoolers have in achieving and maintaining awareness of
the sources of their knowledge. The evidence suggests that
some time is required beyond the acquisition of false--belief
understanding--the understanding that assertions are gen-
erated by human minds and therefore differentiated from
an external reality--for the child to recognize as significant
the sources of their own and others' beliefs. Gopnik and
Graf (1988), for example, found preschool children unable
to indicate whether they had just learned the contents of a
drawer from seeing them or being told about them. Simi-
larly, Taylor, Esbensen, and Bennett (1994) reported
preschoolers as showing little ability to distinguish when
they had acquired knowledge--whether it had just been
taught to them or was something they had "always known"
(as most of them claimed regarding a newly learned fact).
In Gopnik and Graf's (1988) study, the alternative sources
of knowledge that a child was asked to maintain awareness
of were arbitrary (whether the contents of a drawer were
learned from seeing or being told about them) and might
have been dismissed by the child as of no significance once
the knowledge had been acquired. In a current study (Kuhn
& Pearsall, 1998b), in contrast, we ask preschoolers to make
a distinction between alternative sources of knowledge, a
MARCH 1999 19

distinction that is of critical epistemological significance.
'Ihey are asked to differentiate theory, as a source supporting
!he plausibility of a claim, and evidence, as a source that can
>..~pport the truth of the claim. Do young children appreciate
whether they are claiming something to be true because it
~..~kes sense as a way for things to be or on the basis of em-
pirical evidence of its correctness? Young children, we hy-
p~.~thesized, would not be sensitive to this distinction in iden-
titving the sources of their claim that an event had occurred.
We showed preschoolers a sequence of pictures in which,
for example, two runners compete in a race. Certain cues
suggest a theory as to why one will win---e.g., one has fancy
nmning shoes. The final picture in the sequence provides
c~ ,,tence of the outcome--e.g., one of the runners holds a
~rophy and exhibits a wide grin. When children are subse-
quently asked to indicate the outcome and to justify this
kn~,wledge, younger children show a fragile distinction be-
lween the two kinds of justification--'How do you know?"
ate,{ "Why is it so?"--i.e., the evidence for their claim (the
,,~:tcome cue in this case) versus their explanation for it
(the in itia] theory-generating cue). The two appear to jointly
support a single representation of a state of affairs that
"makes sense," with the respective cues treated as inter-
changeable contributors to this knowledge state. In the race
example, young children often answered the "How do you
kn,:~w [he won]?" question not with evidence (e.g., "He's
h~dding the trophy") but with a theory of why this state of .v
aftairs makes sense (e.g., "Because he has fast sneakers"). InJe
another instance, in which a boy is shown first climbing a
tree and then down on the ground holding his knee, the
"How do you know [that he fell]?" question was often an-
:~wered with "Because he wasn't holding on carefully."
Fhese findings do not suggest that 4-year-olds can never
answer "How do you know?" questions, Indeed, they do so
~_c,,nmonly when justifications for their claims are readily
available (e.g., "How do you know it's a zebra? Because it
has stripes."). Rather, the findings suggest that children
who have not yet achieved the epistemological under-
standing in question do not distinguish justifications of dif-
fering epistemological status (such as theory and evidence)
wh~n multiple cues suggest different justifications.
{hese confusions between theory, and evidence diminish
sharply among 6-year-olds, who are more likely to distin-
guish the evidence for their claim from a theow that ex-
plains it. The question, "How do you know?," becomes a
nleaningful one, distinct from the question of whether this
knowledge claim is a plausible one. It is important to em-
FG~:~ize, however, that 6-year-olds who have achieved this
epi>temological understanding still differentiate theory and
evidence in only a very limited sense and set of contexts.
The child is able to distinguish evidence for the claim that
an e'vent occurred from a causal theory that makes occur-
r~'nce of the event plausible. The evidence in question is
simply evidence that the event occurred (e.g., the runner
v,,,,~ the race), which is the claim under consideration. A
ttv,,,,~ryas to why or how the event occurred is a separate en-
tit'. Such a theory can be used to support the plausibility of
the claim, but it remains only a theory and, hence, one that
properly requires its own supporting evidence. Evidence
c,tpable of supporting a causal theory is of more complex
folms. These more complex forms of evidence (such as co-
~,~riation evidence) are those that older children and adults
confuse with their own causal theories, as I note later.
It is the development just described during the preschool
years that reflects a growing metacognitive capacity to re-
flect on one's own knowing and, hence, to distinguish the-
ory and evidence as different forms of knowing, rather than
demonstrations of children's ability to reason about corre-
spondences between assertions and patterns of evidence,
which has been the focus of most studies of early scientific
reasoning skill (see Kuhn et al., 1995, for review). I highlight
this development here as a foundational achievement in
metacognitive and epistemological understanding, as well
as the developmental origin of subsequent achievements in
scientific thinking (Kuhn & Pearsall, 1998b).
Still to develop, however, is the epistemological insight
that different minds can arrive at genuinely different and le-
gitimate understandings of the same evidence. At age 8,
children do not make appropriate judgments regarding the
mental representations of observers exposed to ambiguous
stimuli (Carpendale & Chandler, 1996; Pillow & Henrichon,
1996) and tend to assume that the other will interpret the
stimulus in the same way they do. They do not yet realize
that two people can hold genuinely different beliefs, except
in the case where one party's belief is misinformed and in-
correct. Children at this age lack the interpretive or "con-
structive" theory of mind (Carpendale & Chandler, 1996;
Wellman, 1990) that would lead them to understand con-
flicting representations of the same event as legitimate
products of individuals' unique meaning-making efforts.
This achievement is an important stepping stone in the de-
velopment of epistemological understanding, which we re-
turn to later. First, we examine the developmental origins of
metastrategic knowing.
MetastrategicKnowing
Metastrategic knowing invokes an entirely different, older
developmental literature that has remained unconnected to
the theory-of-mind literature despite their common concern
with meta-knowing. Research on metastrategic awareness
of knowing as process is nonetheless consistent with what
the theory-of-mind literature would predict. If mental rep-
resentations are understood as reflecting the external world
more than the mental activity of the representor, this men-
tal activity does not assume great significance in its own
right and is unlikely to be the object of the child's attention.
Research by Flavell and colleagues confirms this predic-
tion. Although their responses make it clear that they un-
derstand thinking as internal mental activity, preschool
children show limited awareness of its occurrence or con-
tent in themselves or others. They may not acknowledge
that a person engaged in making a decision is thinking and,
despite explicit cues, are poor at judging what the person
has been thinking about. They also show limited awareness
of their own thought. When asked, for example, to think
about the room in their house where their toothbrush is
kept and asked immediately afterward what they have
been thinking about, 5-year-olds failed to mention either
bathrooms or toothbrushes (Flavell, Green, & Flavell, 1995).
Flavell's recent work grows out of the earliest empirical
research in developmental psychology to examine meta-
knowing--a literature on meta-memory (Brown, 1975,
1978; Flavell & Wellman, 1977; Kreutzer, Leonard, & Flavell,
1975). It focuses on children's knowledge and use of strate-
gies, such as rehearsal or categorization, to aid their perfor-
mance in traditional memory tasks such as recalling a list of
20 EDLICATIONAL RESEARCHER

words. In contrast to the early metacognitive achievements
described earlier, no marked achievements appear during
the preschool years. Unless they are prompted to do so,
children tend not to apply strategies spontaneously in con-
texts where they would be useful. Such findings led to the
conclusion that children have poor metastrategic awareness
and understanding of their memory processes (Brown,
Bransford, Ferrara, & Campione, 1983; Flavell & Wellman,
1977; Schneider, 1985). Studies that have directly assessed
children's knowledge about memory support this conclu-
sion (Kreutzer et al., 1975). Not before middle childhood do
children understand, for example, that a memory strategy
such as categorization aids recall (Moynahan, 1978).
Other research points to a more wide-ranging meta-
strategic deficit not confined to memory skills. Children
show little evidence of engaging in metastrategic monitor-
ing of their comprehension of what they read to judge
whether it has been successful (Flavell, Speer, Green, &
August, 1981; Markman, 1979). Such monitoring is a first
step toward the use of metastrategic understanding as a
means to regulate and improve cognitive performance. Al-
though early studies did not indicate strong relations be-
tween metastrategic knowledge and strategic performance,
more recent work suggests a bidirectional relationship in
which metastrategic knowledge both directs and, in turn, is
enhanced by strategy use (Kuhn & Pearsall, 1998a; Schnei-
der & Bjorklund, 1998; Sophian, 1997).
What makes the weaknesses in early metastrategic and
metacognitive skills that have been described here espe-
cially significant is the fact that, with development, we see
improvement but by no means mastery, even by adulthood.
Developmentalists characteristically probe children's early
years in search of the origins of later-appearing competen-
cies. In this case, we appear to be observing the early origins
of a later lack of competence, one that has profound impli-
cations for critical thinking. These basic forms of second-
order cognition--knowing what one knows and how one
knows it and effectively managing and deploying one's
cognitive resources--are the foundation of the critical
thinking skills that we hope to impart to students during
the remainder of their school years. Next, we examine the
progress children make in these respects as they proceed
into and through adolescence.
Life-Span Development of Meta-Knowing
Research involving older children, adolescents, and adults
suggests that meta-knowing competencies--in contrast to
most of the competencies that developmental psychologists
study--remain incompletely developed. My own microge-
netic studies of these age groups have involved two kinds of
cognitive skills--those of analysis (of cause-and-effect rela-
tionships operating in multivariable systems; Kuhn &
Phelps, 1982; Kuhn et al., 1992, 1995) and of argument (in a
framework of alternative assertions, each associated with
supporting and discrepant evidence; Kuhn, Shaw, & Felton,
1997; Kuhn, Weinstock, & Flaton, 1994). The studies involv-
ing analysis skills provide a rich window on the acquisition
of new knowledge as a process of theory-evidence coordina-
tion, invoking the potential for cognitive control and, hence,
critical thinking. For example, participants are asked to ex-
amine a database on children's ratings of TV programs to de-
termine which features of the programs influence their pop-
ularity (Kuhn et al., 1995). The method allows us to follow
over time both their evolving knowledge base and changes
in the strategies by which they acquire that knowledge.
Because the method is a microgenetic one in which change
over time is observed, the data address fundamental issues
regarding the nature of the change process. Findings of other
investigators who have used a microgenetic method to ex-
amine change converge with our own in showing that indi-
viduals approach a task with a repertory of strategies that
they apply variably over time, even when the task environ-
ment remains constant (Kuhn, 1995; Siegler, 1996). Such find-
ings lead to a revised conception of developmental change.
In contrast to the traditional conception of a single transition
in which a new strategy replaces an old one, with the focus
on the challenge of mastering the new strategy, the newer
portrayal of change features continuing shifts in the fre-
quencies of usage of multiple strategies, with the diminished
usage and eventual relinquishment of less adequate strate-
gies a developmental challenge at least equal to that of mas-
tering new, more adequate ones. This newer model of de-
velopmental change has major implications regarding the
role of metastrategic factors. Rather than the traditional
focus on metastrategic understanding of a particular strat-
egy (such as categorization in meta-memory research) as a
factor influencing performance, the major metastra tegic task
becomes strategy selection from the repertory of strategies an
individual has available. Strategy selection is a metastrate-
gi4',--not a strategic--function. Hence, the burden of expla-
nation shifts to the metastrategic level.
At the same time as it calls on metastrategic skill in the
selection and monitoring of strategies, coordinating theo-
ries and evidence requires metacognitive skill in justifying
knowledge claims. It is here that we see in adolescents and
adults metacognitive weaknesses that parallel those ob-
served among preschoolers in the studies described earlier.
Like preschoolers, many older individuals blur the dis-
tinction between theory-based and evidence-based sources
of their beliefs. Rather than seeing their theories as belief
states subject to disconfirmation and representing theory
and evidence as distinct entities to be reconciled with one
another, they merge the two into a single representation
of "the way things are" with little apparent awareness of
the sources of their belief. Evidence serves merely to illus-
trate what one knows to be true, with evidence-based and
theory-based justifications functioning as interchangeable
supports for a claim. Theories may eventually change in re-
sponse to discrepant evidence, but often with the individ-
ual manifesting little awareness or control of the process.
Like young children in the theory-of-mind research de-
scribed earlier, older participants in our studies are likely
to deny that they ever held a belief different from the one
they are now professing.
In the strategic domain, investigative strategies are often
driven by belief, with features believed irrelevant never ex-
amined. Although the strategies of adults we found on av-
erage superior to those of adolescents (Kuhn et al., 1995), a
common pattern at both ages is the use of an adequate strat-
egy to interpret theory-compatible evidence regarding one
feature and an inadequate strategy to interpret theory-
discrepant evidence regarding another feature, even though
the evidence with respect to the two features is identical.
Weak metacognitive awareness of the basis for one's beliefs
and metastrategic inconsistency in the application of infer-
ence strategies thus reinforce one another. A third form of
MARCH 1999 21

meta-knowing also has a role to play--the epistemological,
to which we now turn.
EpistemologicalMeta-Knowing
Research on the development of epistemological under-
standing, originating with the pioneering work of Perry
(1970), has remained curiously isolated from other meta-
knowing research, especially the theory-of-mind work to
which it is most directly related. A possible explanation is
that theory-of-mind research has been largely confined to
children only up to the age of about 6, whereas work on the
development of epistemological thinking has focused on
adolescents and adults. The conceptual connection between
the two bodies of work is nonetheless evident.
With the understanding of assertions as belief states, as-
sertions are recognized as emanating from--and therefore
connected to--the human activity of knowing. Nonethe-
less, the initial absolutist epistemological stance--the norm
in childhood and into adolescence and even adulthood
(Chandler, Boyes, & Ball, 1990; Hofer & Pintrich, 1997; King
& Kitchener, 1994; Kuhn, 1991; Perry, 1970)--does not ac-
cord a pivotal role to the knower as a constructor of knowl-
edge. Rather, the locus of knowledge remains in the exter-
nal world, where it awaits discovery by human knowledge
seekers. A child by 4 or 5 appreciates knowing as connected
to and generated by a knowing agent to a sufficient extent
to understand that beliefs may deviate from a single, true
reality. Yet much slower to be achieved--if, indeed, it is
achieved at all--is a truly constructivist theory of mind that
recognizes the primacy of humans as knowledge construc-
tors capable of generating a multiplicity of valid represen-
tations of reality.
The transition from a realist pre-epistemological un-
awareness of belief states to the initial epistemological
stance of absolutism is nonetheless a profound one. It is a
transition from simply knowing that something is true to
evaluating whether it might be. To carry out such evalua-
tion, absolutists rely on the concept of a certain truth, one
that is known or potentially knowable through either direct
apprehension or the authority of experts. Belief states can
be judged as correct or incorrect in relation to this truth.
Most salient in the present context is the fact that the ab-
solutist stance allows the acquisition of elementary critical
thinking skills that serve as a foundation for more advanced
forms of critical thinking that may develop later. Assertions
are not taken at face value as simple descriptions of an ex-
ternal reality. Rather, they must be evaluated against some
external standard by means of cognitive operations that are
thereby exercised and strengthened.
Although an absolutist epistemological understanding is
adequate to allow development of elementary critical
thinking skills, it does not follow that it provides the
strongest possible support for their development and use.
In fact, an absolutist epistemology is more likely to function
as a significant constraint on the development of critical
thinking. Within the absolutist epistemological framework,
claims under dispute can be resolved by seeking and ob-
taining information found to be lacking, either first hand
from direct observation or second hand from an appropri-
ate authority figure. Finding it may be difficult, but once the
necessary information is available, questions of the truth or
falsity of assertions should be answerable more or less di-
recfly. To the extent, then, that individuals confine them-
selves to an absolutist epistemology, the demand for critical
thinking skills--and, hence, the impetus to exercise and fur-
ther develop them--is slight.
People can spend entire lifetimes within the protective
wraps of either a pre-absolutist stance in which assertions
are equated with reality or, more commonly, the absolutist
stance in which assertions can conflict but disagreements
are resolvable by appeal to direct observation or authority.
In the modern world, however, it is hard to avoid exposure
to conflicting assertions not readily reconcilable by obser-
vation or appeal to authority. As a result, most people
progress beyond absolutism, venturing onto the slippery
slope that will carry them to a multiplist epistemological
stance, which becomes prevalent at adolescence. A critical
event leading to the first step down the slope toward mul-
tiplism is likely to be exposure to the fact that experts dis-
agree about important issues. If even experts cannot be
counted on to provide certain answers, one resolution is to
relinquish the idea of certainty itself, and this is exactly the
path the multiplist takes. As the next inductive leap along
this path, if experts with all of their knowledge and author-
ity disagree with one another, why should their views be ac-
cepted as any more valid than anyone else's? A better as-
sumption is that anyone's opinion has the same status and
deserves the same treatment as anyone else's. Beliefs or
opinions are the possessions of their owners, freely chosen
.!~according to the owner's tastes and wishes and, accord-
•
" ingly, not subject to criticism. In the words of one of the ado-
lescents in our research on argument (Kuhn, 1991, p. 182),
"You can't prove an opinion to be wrong because an opin-
ion is something somebody holds for themselves." Hence,
in a conceptual sleight of hand that represents the final step
down the slippery slope, because all people have a right to
their opinions, all opinions are equally right.
In contrast to the absolutist stance, which is difficult to
maintain in pure form, people often remain multiplists for
life. Only a minority progress to an evaluative epistemology,
in which all opinions are not equal and knowing is under-
stood as a process that entails judgment, evaluation, and ar-
gument. Evaluative epistemologists have reconciled the
idea that people have a right to their views with the under-
standing that some views can nonetheless be more right
than others. They see the weighing of alternative claims in a
process of reasoned debate as the path to informed opinion,
and they understand that arguments can be evaluated and
compared based on their merit (Kuhn, 1991). Those who re-
main at the multiplist level of epistemological understand-
ing, in contrast, conceive of no basis for judging the strength
of an argument, except possibly its power to persuade. As a
result, their critical thinking skills are taxed to an even lesser
extent than those of the absolutist, who stands ready to eval-
uate assertions against a criterion of truth.
The core dimension underlying and driving the progres-
sion in epistemological understanding is the coordination
of the subjective and objective components of knowing. The
absolutist sees knowledge in largely objective terms, as lo-
cated in the external world and knowable with certainty.
The multiplist becomes aware of the subjective component
of knowing, but to such an extent that it overpowers and
obliterates any objective standard that would provide a
basis for comparison or evaluation of opinions. Only the
evaluativist is successful in integrating and coordinating
the two, by acknowledging uncertainty without forsaking

evaluation. (See Table 1 for summary.) This conception is
compatible with that proposed by Hofer and Pintrich (1997)
in a recent summary and review of the literature on the de-
velopment of epistemological thinking.
Although often overlooked, the evolution of episte-
mological understanding is a fundamental part of meta-
knowing and of cognitive development more broadly. In-
deed, as I argue next, epistemological understanding may
have a pivotal role to play when we turn to the concerns of
educators concerned with critical thinking.
Critical Thinking As a Developmental Phenomenon
Developing competence in meta-knowing, I have sug-
gested, warrants attention as a major component of cogni-
tive development. A unifying dimension of this develop-
ment is that of thought becoming increasingly aware of itself
and under the individual's control. To be competent and
motivated to "know how you know" puts one in charge of
one's own knowing, of deciding what to believe and why
and of updating and revising those beliefs as one deems
warranted. To achieve this control of their own thinking is
arguably the most important way in which people both in-
dividually and collectively take control of their lives.
How does critical thinking connect to this achievement?
Each of the three kinds of meta-knowing that have been ex-
amined here--the metacognitive, metastrategic, and episte-
mological--is central to critical thinking. The development
of metacognitive understanding is essential to critical think-
ing because critical thinking by definition involves reflect-
ing on what is known and how that knowledge is justified.
Individuals with well-developed metacognitive skills are in
control of their own beliefs in the sense of exercising con-
scious control over their evolution in the face of external in-
fluences. They know what they think and can justify why.
Their skills in the conscious coordination of theory and ev-
idence also put them in a position to evaluate the assertions
of others.
Metastrategic skill is also essential to critical thinking.
Those who have developed strong metastrategic skills
apply consistent standards of evaluation across time and
situations. They do not succumb to a view of a favored as-
sertion as more probable than its alternatives because of its
favored status and, therefore, subject to different standards
of evolution. They also resist the temptation of "local inter-
pretation" (Klahr, Fay, & Dunbar, 1993) of an isolated piece
of evidence as supportive because it is considered out of the
context of a broader pattern of which it is a part.
The development of epistemological understanding may
be the most fundamental underpinning of critical thinking.
If knowledge is entirely objective, certain, and simply accu-
mulates, unconnected to the human minds that do this
knowing--as the absolutist conceives---or if knowledge is
entirely subjective, subject only to the tastes and wishes of
the knower--as the multiplist conceives--critical thinking
and judgment are superfluous. People must see the point of
thinking if they are to engage in it. Put simply by one of the
multiplists in our studies, "I feel it's not worth it to argue
because everyone has their opinion." In such cases, educa-
tors can undertake to teach intellectual skills, but the rea-
sons to apply them will be missing.
,~ituating these attainments within a developmental
framework makes it possible to investigate ways in which
earlier attainments prepare the way for later ones. In con-
trast to most of the achievements studied by developmen-
tal psychologists, the course of development sketched out
here is one that most individuals never complete. It is here,
then, that we find the intersection of the concerns of re-
searchers seeking to understand cognitive development
Table 1
Levelsof EpistemologicalUnderstanding
Level Assertions Reality Knowledge Critical thinking
Realist Assertions are copies that Reality is directly knowable. Knowledge comes from an Critical thinking is
represent an external external source and is unnecessary.
reality, certain.
Assertions are facts that are Knowledge comes from an
correct or incorrect in their external source and is
representation of reality certain.
(possibility of false belief).
Absolutist Reality is directly knowable. Critical thinking is a
vehicle for compar-
ing assertions to
reality and determin-
ing their truth or
falsehood.
Multiplist Assertions are opinions Reality is not directly Knowledge is generated Critical thinking is
freely chosen by and knowable, by human minds and is irrelevant.
accountable only to their uncertain.
owners.
Assertions are judgments that
can be evaluated and com-
pared according to criteria
of argument and evidence.
Knowledge is generated
by human minds and is
uncertain.
Evaluative Reality is not directly
knowable.
Critical thinking is
valued as a vehicle
that promotes sound
assertions and
enhances under-
standing.
MARCH 1999 23

and the concerns of educators seeking to maximize the re-
alization of intellectual potential. Educators who wish to
foster critical thinking, I have suggested, stand to gain from
conceptualizing students' potential for critical thinking in a
developmental framework of what has preceded and what
is likely to follow. Let us turn, then, to efforts to teach criti-
cal thinking as they exist today and ask how the develop-
mental picture that has been sketched here might enrich
this undertaking.
Teaching CriticalThinking As Cognitive
(and Metacognitive) Development
For a number of years, the major debate among those con-
cerned with teaching critical thinking has been the one
noted earlier between teaching critical thinking skills as
general entities or in the context of subject matter instruc-
tion (Perkins & Salomon, 1989). Today, that debate has been
eclipsed by another that goes directly to the heart of what
we have been concerned with here--identifying exactly
what critical thinking is. On one side of the debate, tradi-
tionalists see it as a set of mental competencies that reside
inside individual heads. On the other, advocates of a newer,
situated-cognition perspective regard intellectual skills as
social practices exercised and shared within a community
(Cole, 1996; Resnick & Nelson-LeGall, 1997; Rogoff, 1998).
Although I have not undertaken to review the evidence
here, the first debate, as noted earlier, has largely come.7
down on the side of instruction embedded in rich subject.,"
matter. However, the successful wedding of thinking-skill
development to subject matter instruction--identified by
Perkins and Salomon (1989) as the major challenge facing
educators--depends on the explicit definitions of thinking
skills argued for here, making them readily identifiable
within varied subject matter. Even if it is the best or only
way to teach them, embedding them in rich content makes
them harder--not easier--to identify. Furthermore, in ad-
dition to needing clear definitions of the skills they are seek-
ing to instill, I have argued, educators benefit from seeing
these skills in the developmental context of where they
have come from and where they are headed, as guideposts
for instructional efforts.
The second, more recent debate between cognitive-
competency and social-practice conceptions might also be
enriched by the developmental framework advocated here.
Transfer has been a major focus of this debate. Although
proponents dispute the charge (Greeno, 1997), critics of the
situated cognition perspective (Anderson, Reder, & Simon,
1996) claim that it has little to tell us about occasions and re-
spects in which behavior does generalize across contexts, as
it often does. People draw on past experience as a powerful
organizer of new encounters. They use analogy as a cogni-
tive tool. Indeed, to apply this critique to the topic of this
essay, critical thinking only makes sense as a construct if we
believe humans have some ability to bring established
modes or methods of thought to bear in approaching new
situations.
One attempt to explain transfer has been to add a "dis-
positional" component to the traditional skill conception of
critical thinking (Perkins, Jay, & Tishman, 1993): Critical
thinking entails the disposition, as well as the skill, to think
well. A problem with the disposition construct, however, is
that it leaves much of the variance unaccounted for. What
determines whether someone has a disposition to behave in
a particular way? If disposition is interpreted in the sense of
habit, as it often is, someone might get used to thinking well
in the same way they are in the habit of exercising high stan-
dards of personal hygiene. Humans, however, are not sim-
ply creatures of habit; their beliefs and values regarding
what is important help to shape their behavior. In the end,
people think carefully and reflectively not out of habit, be-
cause such thinking is not an effortless habit to maintain,
but because they are convinced of the value of doing so.
This conception invokes meta-knowing functions as a
key factor in disposition (and, hence, performance). As a
major and influential part of what develops cognitively
during the childhood and adolescent years, they can help to
bridge the contrasting conceptions of intellectual skills as
individual abilities versus situated practices. Tied to the
practices that identify groups and cultures are beliefs and
values that represent how these practices are understood by
members of the community. These are an essential---even if
less visible--aspect of what is practiced and valued in the
community and thereby appropriated by its younger mem-
bers (Resnick & Nelson-LeGall, 1997). Hence, the power of
social practice as a shaper of human behavior can be recog-
nized without ignoring the knowing and ineta-knowing
powers that enable people to attribute meaning to what
they do and, in the course of so doing, to shape their indi-
vidual and collective behavior.
Where and how, then, might educators intervene to sup-
port the developmental process that has been described
here, a process that the majority of students may never com-
plete? Regular practice of the skills we would like to see de-
velop is essential, we know, but practice does not make per-
fect in the absence of understanding. The best approach,
then, may be to work from both ends at once--from a
bottom-up anchoring in regular practice of what is being
preached so that skills are exercised, strengthened, and con-
solidated as well as from a top-down fostering of under-
standing and intellectual values that play a major role in
whether these skills will be used. The developmental goal
is to put people in metacognitive and metastrategic control
of their own knowing.
The essential contribution that modern cognitive devel-
opment research can make to this challenge is a growing
knowledge base regarding what is developing and, hence,
what needs to change. Skills are only one part of a more
complex structure that develops, with second-order meta-
knowing components both arising from and feeding back to
support the use of first-order cognitive skills. In building a
knowledge base that describes this structure in concrete
enough detail to be useful to educators, researchers can
help define educational objectives rather than only advise
on how to implement objectives determined by others, their
more traditional role as educational consultants. A number
of cognitive development researchers have begun to con-
cern themselves in a serious way with building a bridge
from their research findings to educational practice. This
essay, I hope, will prove a useful starting point to educators
seeking a bridge in the opposite direction, one enabling
them to draw on empirical data on how children's intel-
lects develop, as a means of enriching their visions of good
practice.
24 EDUCA~IIONAL RESEARCHER

References
Anderson, J., Reder, L., & Simon, H. (1996). Situated learning and ed-
ucation. Educational Researcher,25(4), 5-22.
Brown, A. (1975). The development of memory: Knowing, knowing
about knowing, and knowing how to know. In H. Reese (Ed.), Ad-
vances in child development and behavior (VoI. 10, pp. 103-152). New
York: Academic Press.
Brown, A. (1978). Knowing when, where, and how to remember: A
problem of metacognition. In R. Glaser (Ed.), Advances #z instruc-
tional psychology (Vol. 1, pp. 77-165). I-tillsdale, NJ: Erlbaum.
Brown, A. (1997). Transforming schools into communities of thinking
and learning about serious matters. American Psychologist, 52,
399-413.
Brown, A., Bransford, J., Ferrara, R., & Campione, J. (1983). Learning,
remembering, and understanding. In P. Mussen (Series Ed.), J.
Flavell, & E. Markman (Vol. Eds,), Handbook of child psychology:
Vol. 3. Cognitiw! development (4th ed., pp. 77-166). New York: Wiley.
Cahan, E. (1994). John Dewey and human development. In R. Parke, P.
Ornstein, J. Rieser, & C. Zahn-Waxler (Eds.), A century of develop-
mental psychology (pp. 1452167). Washington, DC: American Psycho-
logical Association.
Carpendale, J., & Chandler, M. (1996). On the distinction between false
belief understanding and subscribing to an interpretive theory of
mind. Child Development, 67, 1686-1706.
Chandler, M., Boyes, M., & Ball, L. (1990). Relativism and stations of
epistemic doubt. JournalofExperimental Child Psychology,50, 370-395.
Cole, M. (1996). Cultural psycholoyy: A once andfuture discipline. Cam-
bridge, MA: Harvard University Press.
Dewey, J. (1916). Democracy and education: An introduction to the
philosophy of education. In J. Boydston (Ed.), The middle worksofJohn
Dewey (Vol. 9). Carbondale: Southern Illinois University Press.
Dewey, J. (1933). The process and product of reflective activity: Psy-
chological process and logical forms. In J. Boydston (Ed.), The later
works of Johll Dewey (Vol. 8, pp. 171-186). Carbondale: Southern
Illinois Universify Press.
Ennis, R. (1987). A taxonomy of critical thinking dispositions and abil-
ities. In J. Baron & R. Sternberg (Eds.), Teaching thinking skills: Theory
and practice (pp. 9-26). New York: Freeman.
Estes, D., Wellman, H., & Woolley, J. (1989). Children's understanding
of mental phenomena. In H. Reese (Ed.), Advances in child develop-
ment and behavior(pp. 41-87). New York: Academic Press.
Ferrari, M., & Sternberg, R. (1998). The development of mental abilities
and styles. In W. Damon (Series Ed.), D. Kuhn, & R. Siegler (Vol.
Eds.), Handbookof childpsychology: Vol 2. Cognition, languaxe, and per-
ception. (5th ed., pp. 899-946). New York: Wiley.
Flavei1, J., Green, F., & Flavell, E. (1995). Young children's knowledge
about thinking. Societyfor Researchin Child Development Mom~raphs,
60(Serial No. 243).
Flavell, J., Speer, J., Green, F., & August, D. (1981). The development of
comprehension monitoring and knowledge about communication.
Socfetyfi~rResearchin ChildDevelopmentMonographs,46(Serial No. 192).
Flavell, J., & Wellman, H. (1977). Metamemory. In R. Kail & J. Hagen
(Eds.), Perspectives on the development of memory and cognitio~ (pp.
3-33). Hillsdale, NJ: Erlbaum.
Gelman, R., & Williams, E. (1998). Constraints on cognitive develop-
ment and learning. In W. Damon (Series Ed.), D. Kuhn, & R. Siegler
(Vol. Eds.), Handbook qf child psychology: Vol. 2. Cognition, hmgltage,
and perception (5th ed., pp. 575-630). New York: Wiley.
Gopnik, A., & Graf, P. (1988). Knowing how you know: Young child-
ren's ability to identify and remember the sources of their beliefs.
Child Development, 59, 1366-1371.
Greeno, J. (1997). On claims that answer the wrong questions. Educa-
tilmal Researcher,26(1), 5-17.
Haith, M., & Benson, J. (1998). Infant cognition. In W. Damon (Series
Ed.), D. Kuhn, & R. Siegler (VoL Eds.), Handbook of child psychology:
Vol. 2. Co~,}nition,language, and perception (Sth ed., pp. 199-254). New
York: Wiley.
Hofer, B., & Pintrich, P. (1997). The development of epistemological
theories: Beliefs about knowledge and knowing and their relation to
learning. Review qf Educational Research,67(1), 88-140.
Inhelder, B., & Piaget, J. (1958). The ,~rowthof logical thinking from child-
hood to adolescence. New York: Basic Books.
King, P., & Kitchener, K. (1994). Developing reflectivejudgment: Under-
standino~ and promoting intellectual growth and critical thinking in ado-
lescents and adults. San Francisco: Jossey-Bass.
Klahr, D., Fay, A., & Dunbar, K. (1993). Heuristics for scientific experi-
mentation: A developmental study. Cogldtive Psycholo,~y,25, 111-146.
Kreutzer, M., Leonard, C., & Flavell, J. (1975). An interview study of
children's knowledge about memory. Societyfol Researchin Child De-
velopment Monographs, 40(Serial No. 159).
Kuhn, D. (1989). Children and adults as intuitive scientists. Psycholug
ical Review, 96, 674-689.
Kuhn, D. (1991). The skills qfmxument. New York: Cambridge Univer-
sity Press.
Kuhn, D. (1995). Microgenetic study of change: What has it told us?
PsychologicalScience,6, 133-139.
Kuhn, D. (in press). Metacognitive development. In C. Tamis-
Le-Monda (Ed.), Child psychology:A handbookq~contemporary issues.
New York: Garland.
Kuhn, D., Garcia-Mila, M., Zohar, A., & Andersen, C. (1995). Strategies
of knowledge acquisition. Societyfor Research in Child Developme~lt
Monographs, 60(Serial No. 245).
Kuhn, D., & Pearsall, S. (1998a). Relations between metastrategic
knowledge and strategic performance. Cognitive Development, 13.
227--247.
Kuhn, D., & Pearsall, S. (1998b). Metaco~fftive dimensio~isofknowledxc
acquisition: Awareness of theory and evide~zceas sources of one's knowl-
edge. Unpublished manuscript, Teachers College, Columbia Univer-
sity, New York.
Kuhn, D., & Phelps, E. (1982). The development of problem-solving
strategies. In H. Reese (Ed.), Advances in child deveh~pment and behav-
ior (Vol. 17, pp. 1-44). New York: Academic Press.
Kuhn, D., Schauble, L., & Garcia-Mila, M. (1992). Cross-domain devel-
opment of scientific reasoning. Cognition and Instruction, 9, 285-332.
Kuhn, D., Shaw, V., & Felton, M. (1997). Effects of dyadic interaction
on argumentive reasoning. CoRnitionmid Instruction, 15, 287-315.
Kuhn, D., Weinstock, M., & Flaton, R. (1994). Historical reasoning as
tlJ~eory-evidence coordination. In M. Carretero & 1-Voss (Eds.), Co,~-
~d'tive and instructional processes in history and the social scieJ~ce,,
(pp. 377-401). Hillsdale, NJ: Erlbaum.
Lipman, M. (1991). Thi!~kingin educatioH. New York: Cambridge Uni-
versity Press.
Markman, E. (1979). Realizing that you don't understand: Elementary
school children's awareness of inconsistencies. Child Developmc~tt,
50, 643-655.
McPeck, J. (1981). Criticalthinkiuq,and education. New York: St. Martin's
Press.
Moshman, D. (1998). Cognitive development beyond childhood: Con-
straints on cognitive development and learning. In W. Damon (Se-
ries Ed.), D. Kuhn, & R. Siegler (Vol. Eds.), Hamlbookof child psycho/
ogy: Vol.2. Coxnition, langua,ge,alut perception (5th ed., pp. 947-978).
New York: Wiley.
Moynahan, E. (1978). Assessment and selection of paired associate
strategies: A developmental study. Journalof Experimental Chiht Psy-
chology, 26, 257--266.
Norris, S., & Ennis, R. (1989). Evaluating criticalthinking. Pacific Grove,
CA: Critical Thinking Press and Software.
Olson, D., & Astington, J. (1993). Thinking about thinking: Learning
how to take statements and hold beliefs. Educational Psycholo,~ist.
28(1), 7-23.
Paul, R. (1990). Critical thinking. Rohnert Park, CA: Center for Critical
Thinking and Moral Critique, Sonoma State University.
Perkins, D., Jay, E., & Tishman, S. (1993). Beyond abilities: A disposi-
tional theory of thinking. Merrill-Pahner Quarterly, 39(1), 1-21.
(Special Issue on "The Development of Rationality and Critical
Thinking" )
Perkins, D., & Salomon, G. (1989). Are cognitive skills context-bound?
Educational Researcher,18(1), 16-25.
Perner, J. (1991). UuderstandinR the representational mimL Cambridge,
MA: MIT Press.
Perry, W. (1970). Formsof intellectualand ethicaldevelopment in the colhxe
years. New"York: Holt, Rinehart & Winston.
Pillow, B., & Henrichon, A. (1996). There's more to the picture than
meets the eye: Young children's difficulty understanding biased in-
terpretation. Child Development, 67, 803-819.
Resnick, L., & Nelson-LeGall, S. (1997). Socializing intelligence. In L.
Smith, J. Dockrell, & P. Tomlinson (Eds.), Piaget, Vygotsky, and beyomt
(pp. 145-158). London: Routledge.
Rogoff, B. (1998). Cognition as a collaborative process. In W. Dam¢}n
(Series Ed.), D. Kuhn, & R. Siegler (Vol. Eds.), Handbookqfchild psy-
chology: Vol. 2. Cognition, lauguage, and perception (5th ed., pp.
679-744). New York: Wiley.
(coutiuued oI~p. 4(~)
MARCH 1999 25

(continuedfrom p. 25)
Ruffman, T., Perner, J., Olson, D., & Doherty, M. (1993). Reflecting on
scientific thinking: Children's understanding of the hypothesis-
evidence relation. Child Development, 64, 1617-1636.
Schauble, L. (1996). The development of scientific reasoning in knowl-
edge-rich contexts. Developmental Psychology, 32, 102-119.
Schneider, W. (1985). Developmental trends in the metamemory-
memory behavior relationship: An integrative review. In D. Forrest-
Pressley, G. MacKinnon,&T.Waller(Eds.), Cognition,metacognition,and
human performance(Vol. 1, pp. 57-109). Orlando, FL:Academic Press.
Schneider, W., & Bjorklund, D. (1998). Memory. [n W. Damon (Series
Ed.), D. Kuhn, & R. Siegler (Vol. Eds.), Handbook ~child psychology:
Vol. 2. CGgnition,language, and perception (5th ed., pp. 467-522). New
York: Wiley.
Shultz, T., & Mendelson, R. (1975). The use of covariation as a princi-
ple of causal analysis. Child Development, 46, 394-399.
Siegel, H. (1988). Educating reason:Rationality, critical thinking, and edu-
cation. New York: Routledge.
Siegler, R. (1996). Emergine~minds: The process of change in children's
thinking. New York: Oxford University Press.
Sophian, C. (1997).Beyond competence: The significanceof performance
for conceptual development. Cognitive Development, 12, 281-303.
Spelke, E., & Newport, E. (1998). Nativism, empiricism, and the devel-
opment of knowledge. In W. Damon (Series Ed.) & R. Lerner (Vol.
Ed.), Handbook qf child psychology: Vol. 1. Theoreticalmodels ~f human
development (5th ed., pp. 275-340). New York: Wiley.
Taylor, M., Esbensen, B., & Bennett, R. (1994).Children's understand-
ing of knowledge acquisition: The tendency for children to report
they have always known what they have just learned. Child Devel-
opnlent, 65, 1581-1604.
Wellman, H. (1990). The child's theory ~ mind. Cambridge, MA: MIT
Press.
Wellman, H., & Gelman, S. (1998). Knowledge acquisition in founda-
tional domains. In W. Damon (Series Ed.), D. Kuhn, & R. Siegler
(Vol. Eds.), Handbook of child psychology: Vol. 2. Cognition, language,
and perception (5th ed., pp. 523-574). New York: Wiley.
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation
and constraining function of wrong beliefs in young children's un-
derstanding of deception. Cognition, 13, 103-128.
Manuscript received October 19, 1995
Revisions received September 9, 1996,
February 19, 1998, and May 5, 1998
Accepted May 8, 1998
Publications-Related Events at
the 1999 Annual Meeting
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
The Publications Committee is pleased to announce that this year'sAnnual
Meeting includes several events, described below, which should be of interest
to participants who would like to discuss publishing.
Open Meeting of the Publications Committee: The Future of the AERA Publica-
tions Program, Wednesday, April 21, 12:25-1:55
The Publications Committee will discuss how AERA publications are striving to
respond to the interests and commitments of the educational research
community.
Journal Talk I and Journal Talk II: Contributing to Educational Journals
Thursda~ April 22, 12:25-1:05; and Tuesday, April 20, 10:35-11:15
These roundtable sessions will give those interested in contributing to journals a
chance to speak with editors of AERA journals and other education journals.
Welcoming New Voices Into the AERA Publications Program: An Open House
and Reception, Thursday, April 22, 6:15-7:45
¢

A Developmental Model Of Critical Thinking

More Related Content

Similar to A Developmental Model Of Critical Thinking (20)

More from Andrew Molina (20)

Recently uploaded (20)

A Developmental Model Of Critical Thinking