A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Educational Psychology
This content is subject to copyright. Terms and conditions apply.
Journal of Educational Psychology
1988,
Vol. 80. No. 4. 419-423Copyright 1988 by the American Psychological Association, Inc.
Ofm-066 3/88/$OO. 7 5
Hidden Costs of Reflectiveness:
Aspects of Successful Scientific Reasoning
David Duemler and Richard E. Mayer
University of California, Santa Barbara
We tested a prediction of Baron's (1985) rationality theory: that reflectiveness is directly related
to creative problem-solving performance. In two experiments, subjects solved a series of four
induction problems involving conventional rules, followed by an induction problem involving
an unconventional rule. Subjects who had been extremely reflective or extremely impulsive on
successfully solving conventional problems performed more poorly on solving the unconventional
problem than did subjects who had been moderately reflective or moderately impulsive on
conventional problems. This U-shaped relationship between degree of reflectiveness/impulsive-
ness on conventional problems and problem-solving performance on an unconventional problem
conflicts with the predictions of rationality theory.
Scientific reasoning involves generating and testing hy-
potheses based on collected data. In the course of scientific
thinking, one person may
go
public with a theory more rapidly
than another person; that is, some people may be willing to
publically state their theories on the basis of a small amount
of collected data, whereas other people prefer to wait until
more data are collected before accepting a theory. Premature
acceptance of a theory may be considered one aspect of
impulsive thinking, whereas waiting for data may be consid-
ered one aspect of reflective thinking.
Baron's (1985) theory of rational thinking is based on the
idea that a reflective thinking style is more conducive to
successful performance on creative problem-solving tasks
than is an impulsive thinking style. Baron (1985) lists
"insuf-
ficient search" as one of the defining features of
the
reflective-
ness/impulsiveness distinction:
I suggested that there might be a general bias to stop too soon
when collecting evidence. We can call this bias impulsiveness.
(P.
157)
Baron (1985) goes on to argue that students should be taught
to think in a way that prevents them from forming hypotheses
(or jumping to conclusions) on the basis of insufficient evi-
dence:
My purpose here is to show ... the teaching of good thinking
.. . should direct itself at the removal and prevention of the
major
biases
[including] insufficient search (and
its
total absence)
.... (p. 244)
One straightforward prediction of Baron's rationality theory
is that reflectiveness (or absence of impulsiveness) is directly
related to the probability of success on creative problem-
solving
tasks,
that
is,
the more reflective a student
is,
the more
successful the student will be in creatively solving problems.
Although we are sympathetic to Baron's, as well as to
Dewey:s (1933), call for the teaching of proper habits of
Correspondence concerning this article should be addressed to
David Duemler or Richard E. Mayer, Department of Psychology,
University of California, Santa Barbara, California 93106.
reflective thinking, the present study explores some limita-
tions on the positive effects of reflective thinking. In particu-
lar, we focus on one aspect of the reflectiveness/impulsiveness
distinction, what Baron (1985, p. 157) calls the "bias to stop
too soon when collecting evidence" during hypothesis testing.
This bias is indicated by the tendency to go public with a
hypothesis after collecting minimal evidence. Like Baron, we
also avoid the view of reflectivity/impulsivity as a personality
trait (Baron, Badgio,
&
Gaskins, 1986; Kagan, Rosman, Day,
Albert, & Phillips, 1964; Messer, 1976), and we focus on just
one aspect of
this
dimension.1
To investigate the predictions of rationality theory, we used
a modified version of the classic einstellung procedure (Lu-
chins,
1942). Students solved a series of
rule
induction prob-
lems that were all based on the same type of salient rule (i.e.,
conventional problems), and then we gave the students an
identical-looking problem that was based on a qualitatively
different and less salient kind of rule (i.e., unconventional
problem). Some students (whom we defined as reflective)
solved the conventional problems without ever overtly stating
an incorrect hypothesis along the way; other students (whom
we defined as impulsive) stated incorrect hypotheses on all or
most of the conventional problems before eventually solving
them; and other students (whom we defined as moderate)
stated incorrect hypotheses on some but not most of the
conventional problems before eventually solving them. Our
interpretation of rationality theory predicts a linear relation-
ship between degree of reflectiveness on conventional prob-
lems (i.e., the tendency to withhold stating a hypothesis until
it is completely tested) and probability of success on solving
the unconventional problem.
1 It should be noted that our operational definition of reflective-
ness/impulsiveness is more limited and specific than, and possibly
different from, some of
the
more traditional, general definitions. For
example, in contrast to some definitions of reflectiveness/impulsive-
ness,
we do not use measures of latency in our definition. Therefore,
the reader may prefer to substitute different terminology such as low
versus
high risk taking for
reflectiveness versus
impulsiveness.
419
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
