Professional Documents
Culture Documents
Textbook International Handbook of Thinking and Reasoning 1St Edition Linden J Ball Ebook All Chapter PDF
Textbook International Handbook of Thinking and Reasoning 1St Edition Linden J Ball Ebook All Chapter PDF
https://textbookfull.com/product/the-routledge-international-
handbook-of-clinical-hypnosis-1st-edition-julie-h-linden/
https://textbookfull.com/product/handbook-of-item-response-
theory-volume-three-applications-1st-edition-wim-j-van-der-
linden-editor/
https://textbookfull.com/product/critical-thinking-an-
introduction-to-reasoning-well-watson/
https://textbookfull.com/product/reasoning-unbound-thinking-
about-morality-delusion-and-democracy-1st-edition-jean-francois-
bonnefon-auth/
Handbook of Item Response Theory, Three Volume Set -
First Edition Van Der Linden
https://textbookfull.com/product/handbook-of-item-response-
theory-three-volume-set-first-edition-van-der-linden/
https://textbookfull.com/product/hypothetical-thinking-classic-
edition-dual-processes-in-reasoning-and-judgement-jonathan-st-b-
t-evans/
https://textbookfull.com/product/on-reasoning-and-argument-
essays-in-informal-logic-and-on-critical-thinking-1st-edition-
david-hitchcock-auth/
https://textbookfull.com/product/the-enlightenment-of-bees-1st-
edition-rachel-linden/
https://textbookfull.com/product/the-power-of-critical-thinking-
effective-reasoning-about-ordinary-and-extraordinary-claims-6th-
ed-6th-edition-lewis-vaughn/
“If there is any doubt that the field of thinking and reasoning is central to current psychological
science, this Handbook should dispel those doubts. I can’t imagine a young scholar in cognitive
science who wouldn’t find something of interest in this set of up-to-date empirical and theo-
retical chapters.”
—Keith E. Stanovich, University of Toronto, author of The Rationality Quotient
“This book offers authoritative overviews of every major topic in reasoning research, written
by leading scholars from around the world. Its 35 chapters capture the field’s history as well as
its exciting cutting-edge developments, making this Handbook a valuable resource for years
to come.”
—Norbert Schwarz, University of Southern California
The Routledge International Handbook of Thinking and Reasoning is an authoritative reference work
providing a well-balanced overview of current scholarship spanning the full breadth of the
rapidly developing and expanding field of thinking and reasoning. It contains 35 chapters by
leading international researchers, covering foundational issues as well as state-of-the-art devel-
opments, both in relation to empirical evidence and theoretical analyses.
Topics covered range across all sub-areas of thinking and reasoning, including deduction,
induction, abduction, judgment, decision making, argumentation, problem solving, expertise,
creativity and rationality. The contributors engage with cutting-edge debates and pressing con-
ceptual issues such as the status of dual-process theories of thinking, the role of unconscious,
intuitive, emotional and metacognitive processes in thinking and the importance of probabilistic
conceptualisations of thinking and reasoning. In addition, authors examine the importance of
neuroscientific findings in informing theoretical developments, as well as the situated nature of
thinking and reasoning across a range of real-world contexts such as mathematics, medicine and
science.
The Handbook provides a clear sense of the way in which contemporary ideas are chal-
lenging traditional viewpoints in what is now referred to as the “new paradigm psychology
of reasoning”. This paradigm-shifting research is paving the way toward a far richer and more
encompassing understanding of the nature of thinking and reasoning, where important new
questions drive a forward-looking research agenda. It is essential reading for both established
researchers in the field of thinking and reasoning as well as advanced students wishing to learn
more about both the historical foundations and the very latest developments in this rapidly
growing area.
Linden J. Ball is Professor of Cognitive Psychology and Dean of Psychology at the University
of Central Lancashire. He is Editor-in-Chief of Journal of Cognitive Psychology, Associate Editor
of Thinking & Reasoning and Editor of Routledge’s Current Issues in Thinking & Reasoning book
series.
Linden to Jonathan St. B.T. Evans, who helped to set me off down the fascinating
path of thinking and reasoning research and who was always an exemplar of what it
means to be an exceptional cognitive scientist.
Valerie to the memory of Allan Paivio, who was a model of experimental rigour,
theoretical precision, and integrity in and out of the laboratory. I constantly aspire to
meet the stellar standard he set.
CONTENTS
Contributorsxii
Prefacexix
3 Intuitive thinking 37
Tilmann Betsch and Pablina Roth
6 Fallacies of argumentation 88
Peter J. Collins and Ulrike Hahn
ix
Contents
x
Contents
29 Several logics for the many things that people do in reasoning 523
Keith Stenning and Alexandra Varga
31 The sense of coherence: how intuition guides reasoning and thinking 559
Sascha Topolinski
Index639
xi
CONTRIBUTORS
Rakefet Ackerman
Faculty of Industrial Engineering and Management
Technion – Israel Institute of Technology
Haifa, Israel
Linden J. Ball
School of Psychology
University of Central Lancashire
Preston, UK
Tilmann Betsch
Department of Psychology
University of Erfurt
Erfurt, Germany
Isabelle Blanchette
Département de Psychologie
Université du Québec à Trois-Rivières
Trois-Rivières, Québec, Canada
Jean-François Bonnefon
Center for Research in Management
Université Toulouse 1 Capitole
Toulouse, France
Alexander P. Burgoyne
Department of Psychology
Michigan State University
East Lansing, Michigan, USA
xii
Contributors
Ruth M. J. Byrne
School of Psychology and Institute of Neuroscience
Trinity College Dublin
University of Dublin
Dublin, Ireland
Guillermo Campitelli
School of Arts and Humanities
Edith Cowan University
Joondalup, Western Australia, Australia
Serge Caparos
Psychology, Literature, Language, and History
Université de Nîmes
Nîmes, France
Nick Chater
Behavioural Science Group
Warwick Business School
University of Warwick
Coventry, UK
Peter J. Collins
Department of Psychological Sciences
Birkbeck, University of London
London, UK
Pat Croskerry
Department of Emergency Medicine
Dalhousie University
Halifax Infirmary
Halifax, Nova Scotia, Canada
Nicole Cruz
Department of Psychological Sciences
Birkbeck, University of London
London, UK
Wim De Neys
Laboratory for the Psychology of Child Development and Education Sorbonne
Paris Descartes University
Paris, France
Shira Elqayam
School of Applied Social Sciences
De Montfort University
Leicester, UK
xiii
Contributors
Aidan Feeney
School of Psychology
Queen’s University Belfast
Belfast, UK
Dedre Gentner
Department of Psychology
Northwestern University
Evanston, Illinois, USA
Bart Geurts
Department of Philosophy
Radboud University
Nijmegen, The Netherlands
Kenneth J. Gilhooly
Psychology Department
University of Hertfordshire
Hatfield, UK
Vinod Goel
Department of Psychology
York University
Toronto, Ontario, Canada
Geoffrey P. Goodwin
Department of Psychology
University of Pennsylvania
Philadelphia, Pennsylvania, USA
Michael E. Gorman
Department of Science, Technology and Society
University of Virginia
Charlottesville,Virginia, USA
Sebastian Hafenbrädl
IESE Business School
University of Navarra
Barcelona, Spain [Current Address]
and
School of Management
Yale University
New Haven, Connecticut, USA
xiv
Contributors
and
Faculty of Business and Economics
University of Lausanne
Lausanne, Switzerland
Ulrike Hahn
Department of Psychological Sciences
Birkbeck, University of London
London, UK
David Z. Hambrick
Department of Psychology
Michigan State University
East Lansing, Michigan, USA
Rebecca K. Helm
School of Law
University of Exeter
Exeter, UK [Current Affiliation]
and
Department of Human Development
Cornell University
Ithaca, New York, USA
Denis J. Hilton
Laboratoire Cognition, Langage, Langues, Ergonomie
Université de Toulouse Le Mirail
Toulouse, France
Ulrich Hoffrage
Faculty of Business and Economics
University of Lausanne
Lausanne, Switzerland
Philip N. Johnson-Laird
Psychology Department
Princeton University
Princeton, New Jersey, USA
and
Department of Psychology
New York University
New York, New York, USA
Sangeet S. Khemlani
Navy Center for Applied Research in Artificial Intelligence
Naval Research Laboratory
Washington, District of Columbia, USA
xv
Contributors
Barbara Koslowski
Department of Human Development
Cornell University
Ithaca, New York, USA
Brooke N. Macnamara
Department of Psychological Sciences
Case Western Reserve University
Cleveland, Ohio, USA
Francisco Maravilla
Department of Psychology
Northwestern University
Evanston, Illinois, USA
Julian N. Marewski
Faculty of Business and Economics
University of Lausanne
Lausanne, Switzerland
Henry Markovits
Département de Psychologie
Université du Québec à Montréal
Montréal, Québec, Canada
Michael J. McCormick
Department of Psychology
Auburn University
Auburn, Alabama, USA
Hugo Mercier
Centre National de la Recherche Scientifique
Institut des Sciences Cognitives Marc Jeannerod—UMR5304
Bron, France
Kevin C. Moore
Department of Mathematics & Science Education
University of Georgia
Athens, Georgia, USA
Ben R. Newell
School of Psychology
University of New South Wales
Sydney, Australia
Mike Oaksford
Department of Psychological Sciences
Birkbeck, University of London
London, UK
xvi
Contributors
David E. Over
Department of Psychology
Durham University
Durham, UK
Tim Rakow
Department of Psychology
Institute of Psychiatry, Psychology and Neuroscience
King’s College London
London, UK
Valerie F. Reyna
Human Neuroscience Institute
Cornell University
Ithaca, New York, USA
Pablina Roth
Institut für Sport und Sportwissenschaft
University of Heidelberg
Heidelberg, Germany
Mark A. Runco
American Institute of Behavioral Research and Technology
Leucadia, California, USA
and
University of Georgia
Athens, Georgia, USA
Christin Schulze
Center for Adaptive Rationality Max Planck Institute for Human Development
Berlin, Germany
Peter Sedlmeier
Institut für Psychologie
Technische Universität Chemnitz
Chemnitz, Germany
William J. Skylark
Department of Psychology
University of Cambridge
Cambridge, UK
Keith Stenning
School of Informatics
University of Edinburgh
Edinburgh, UK
xvii
Contributors
Valerie A. Thompson
Department of Psychology
University of Saskatchewan
Saskatoon, Saskatchewan, Canada
Maggie E. Toplak
LaMarsh Centre for Child and Youth Research
Department of Psychology
York University
Toronto, Ontario, Canada
Sascha Topolinski
Department of Psychology
Social and Economic Cognition
University of Cologne
Cologne, Germany
Bastien Trémolière
Département de Psychologie
Université du Québec à Trois-Rivières
Trois-Rivières, Québec, Canada
Alexandra Varga
Titu Maiorescu University
Bucharest, Romania
Randall Waechter
School of Medicine
School of Veterinary Medicine
Windward Islands Research and Education Foundation
St. George’s University
Grenada, West Indies
Keith Weber
Graduate School of Education
Rutgers University
New Brunswick, New Jersey, USA
Robert W. Weisberg
Department of Psychology
Temple University
Philadelphia, Pennsylvania, USA
Hiroshi Yama
School of Literature and Human Sciences
Osaka City University
Osaka, Japan
xviii
PREFACE
What is the goal of a handbook? Clearly one goal must be to cover a wide overview of a disci-
pline. It should be thorough and authoritative and provide both researchers and students with a
well-balanced synopsis of contemporary scholarship across the full breadth of an area of enquiry.
To this end, our Handbook surveys the current state of the art in the rapidly developing and
expanding field of thinking and reasoning research, in relation to both empirical evidence and
theoretical analyses.
Having said that, we also believe that it is insufficient for a handbook merely to provide an
overview of the current state of affairs and the history leading up to it. We also wanted our
Handbook to capture the incredible momentum that is apparent in the field of thinking and
reasoning research at the moment. As such, we sought to provide the reader with a strong sense
of where the field is going by addressing the new questions that researchers are attempting to
answer – and with which tools: what do researchers perceive to be profitable avenues for future
research? What are the latest conceptual challenges? Thus, our second goal was to cover emerg-
ing trends and cutting-edge debates and to position these in relation to established findings and
theories so as to give a clear impression of the way in which new ideas are challenging tradi-
tional viewpoints and charting new territory.
Within this context, a discerning reader might note that many of our chapter titles are
similar to those that appeared in the first Handbook of Thinking and Reasoning that was edited
by Keith Holyoak and Robert Morrison back in 2005 and published by Cambridge Univer-
sity Press. Indeed, had there been a handbook put together even 20 years ago, chapters on the
topics of deduction, induction, problem solving and decision making would certainly have been
included, as they are staples in our field. Many of the same themes that permeate the present
volume have permeated the field for over 20 years, including debates about the nature of ration-
ality and the standards by which it should be evaluated. Less clear from a surface glance, however,
is that the content of the current chapters would be novel to readers of an earlier generation,
with the new emphasis on probability (versus deductive logic), multimodal and multidiscipli-
nary approaches to theorising and the application of mathematical modelling and neuroimaging
techniques. Researchers today also rely extensively on eye-movement tracking, mouse track-
ing, reaction-time data, confidence measures and data that derive from other process-tracing
methods. The use of multiple, converging evidence in understanding reasoning phenomena
represents possibly one of the largest single transformations of our field as we move away from
xix
Preface
established approaches in which the primary (and usually only) measurement of interest has
been a single outcome measure: the answer. Thus, instead of having to infer backwards from the
answer to the processes that produced the answer, we now have access to a variety of techniques
to examine those processes as they unfold over the time-course of thinking and reasoning.
We have also emphasised situated reasoning in our Handbook, with several chapters explic-
itly dedicated to studying reasoning in context. These chapters include, but are not limited to,
approaches emphasising the role of fast-and-frugal heuristics and naturalistic decision-making
strategies in everyday reasoning, but also extend to chapters discussing informal inference, rea-
soning by argumentation, conversational inference, expert reasoning, abductive reasoning, ana-
logical reasoning and cultural and emotional influences on reasoning. As our field has grown
more sophisticated in its theoretical and empirical approaches, we are very glad to see research-
ers expanding their vision to move from the laboratory to the real world.
In addition, we have included a large number of chapters that address emerging issues and
paradigms in reasoning research, again with an emphasis on both informal as well as formal rea-
soning.This emphasis is in keeping with the larger movement of the field away from formalisms
(or at least a single formalism) towards capturing reasoning that is nonmonotonic and probabil-
istic, and that is guided by emotion and is culturally situated. The role of intuition in thinking is
another pervasive theme in the current volume. Again, the origins of this work are historical in
nature, dating back to the original work on judgment, decision making and problem solving by
the likes of Daniel Kahneman, Amos Tversky, Herbert Simon and Janet Metcalfe, but it is clear
that intuition now plays a pervasive role in many different applications, ranging from creativity
and innovation to medical practice and metacognition. There appears to be a growing recog-
nition that thinking and reasoning, as is the case with most other cognitive activities, requires
at least as much attention to the vast cognitive underground of implicit processes as to more
explicit ones. Given that both thinking and reasoning are traditionally defined with reference to
consciousness, this represents a significant shift in the way that we think about thinking.
Note that, rather than trying to assemble the chapters into themes, we have organised them
alphabetically by author.The reason for this is that we were more or less defeated in our attempts
at other organisational structures given that many chapters could legitimately be classified in
multiple categories. In this respect our final observation on the state of thinking and reasoning
research is that the issues common to various areas of inquiry seem nowadays to be far greater
than the issues that separate these areas. Consider, for a moment, what differentiates the field of
problem solving from the field of reasoning, other than the paradigms being used to investigate
them. Both involve a combination of implicit and explicit processes and require the reasoner to
have a goal and a plan to carry out that goal; both require the transformation of a representation
from one state to another; and both may rely on stored knowledge, insight, logical reasoning,
probabilistic reasoning and the like. One runs into similar difficulties when trying to separate
the topics of reasoning and decision making, or when trying to carve up the various subtypes
of reasoning. The interconnectedness of theories, explanatory constructs and conceptual issues
makes it difficult to sort things into silos. In our view, an important step forward for the future
of our discipline is to continue to work our way out of the silos towards a more integrated view
of the processes that contribute to thinking and reasoning, broadly defined.
We were very fortunate in this endeavour to have had cooperation and contributions from
such a wide range of authors. Many are established experts in their fields and others are midca-
reer researchers with a reputation for excellent work, whilst a few are “rising stars” who are still
at an early career stage but who are nevertheless already being recognised for producing out-
standing research. It has been a genuine pleasure to interact with such an exceptional group of
authors, and the quality of the current Handbook is, of course, entirely due to the high-quality,
xx
Preface
accessible chapters that they have produced. We are also grateful to our authors for serving as
each other’s peer reviewers so as to enable further enhancements to the research presented in
this volume. We also thank four external reviewers – Jamie Campbell, Niall Galbraith, Ed Stup-
ple and Dries Trippas – who graciously gave up their time to provide additional expert reviews
of certain chapters. Finally, we thank Ceri Griffiths, our wonderful editor at Routledge, who was
instrumental in initiating this Handbook and who has provided us with considerable support
and encouragement throughout the whole project. We are now happy to bring this endeavour
to its fruition and very much hope that you will enjoy the experience of reading these chapters
as much as we did.
Valerie A. Thompson
University of Saskatchewan
Saskatoon, Canada
Linden J. Ball
University of Central Lancashire
Preston, UK
April 2017
xxi
1
META-REASONING
Shedding metacognitive light on
reasoning research
In this chapter, we argue that understanding the processes that underlie reasoning, problem
solving, and decision-making1 can be informed by understanding the metacognitive processes
that monitor and control them. Our goal is to show that a metacognitive analysis applies to
a wide range of reasoning tasks and theoretical perspectives, including Dual Process Theo-
ries, Mental Models Theory (Johnson-Laird & Byrne, 1991), Fast and Frugal Heuristics (e.g.,
Gigerenzer, Todd, & the ABC Group, 1999), probabilistic models of reasoning (Oaksford &
Chater, 2007), and a wide variety of problem-solving paradigms. We hope that the range of
examples that we provide will allow the reader to usefully extend these principles even further,
to theories of analogy, induction, causal inference, and so on.
Metacognition is often defined as “thinking about thinking”, which implies a reflective,
introspective set of processes by which we evaluate and alter our approach to the world. In
contrast, most theorists conceive of metacognitive processes as those that are responsible for
monitoring and controlling our ongoing cognitive processes (Nelson & Narens, 1990). They
are thought to be running in the background and monitoring the success of ongoing cogni-
tive processes (such as reading, remembering, reasoning) in much the same way as a thermostat
monitors the temperature of the air. Like the thermostat, which can send a signal to the furnace
to start or terminate functioning, metacognitive processes are assumed to have an analogous
control function over the initiation or cessation of mental effort.
From the point of view of reasoning research, understanding these processes is important
because there is compelling evidence from other domains (e.g., memorisation by Metcalfe &
Finn, 2011 and reading comprehension by Thiede, Anderson, & Therriault, 2003) that these
monitoring processes are the input to control processes, which then allocate attentional and
working memory resources. By extension, therefore, we would expect analogous control
processes in reasoning, problem-solving, and decision-making tasks. Equally important is the
evidence that the processes that monitor performance are often based on aspects of said per-
formance that may be irrelevant to achieving good outcomes. Thus, control processes may
misdirect or prematurely terminate processing based on poorly calibrated input cues. We
argue that understanding the factors that inform these monitoring processes is necessary to
understanding the outcome of any reasoning endeavor and for improving reasoners’ allocation
of resources.
1
Rakefet Ackerman and Valerie A. Thompson
2
Meta-Reasoning
means that there is potential for a reasoner to change their initial answer via deliberate analy-
sis. It does not matter whether one assumes that the fast and the slow answer are delivered by
qualitatively different processes – the fact that the initial answer may be changed invites the
question of when and under what circumstances the answer is changed or kept. It also invites an
explanation for the length of time the reasoner spends deliberating as well as the variables that
determine how satisfied she is with the final answer. Again, these are fundamentally metacogni-
tive questions.
A case in point is the well-known Mental Models theory (Johnson-Laird & Byrne, 2002;
Johnson-Laird, Goodwin, & Khemlani, this volume). This is a theory of how people represent
information in a problem space, and how those representations afford inferences about that
information. A key assumption of the theory is that people often form an incomplete represen-
tation of the problem space, which can be, but which is not always, fleshed out. For example,
consider the following pair of premises:
What follows?
According to Mental Model theory, reasoners construct a model that integrates the premise
information and then draw a conclusion from it. For example, the following notation describes
a mental representation of the premises that support the conclusion that all of the beekeepers
are chemists:
There is, however, another way to represent the premises that undermines that conclusion:
beekeeper
chemist artist beekeeper
chemist artist beekeeper
chemist artist
Consequently, the conclusions that reasoners draw are determined, at least in part, by whether
they are content with their initial answer or whether they search for an alternative. Thus, moni-
toring and control processes are key for understanding why and when reasoners are content
with their initial representations, and the conditions that lead them to expend the necessary
effort to supplement it.
Similarly, several theories of reasoning posit that the search for counter-examples to a puta-
tive conclusion plays a crucial role in reasoning outcomes (e.g., Cummins, Lubart, Alksnis, &
Rist, 1991; Markovits, 1986; Thompson, 2000); again, the goal is to understand when and why
reasoners initiate or fail to initiate such a search. Across a wide spectrum of tasks and para-
digms, therefore, a complete understanding of the processes that operate therein requires us to
3
Rakefet Ackerman and Valerie A. Thompson
Reasoning Meta-Reasoning
Metacognive Metacognive
monitoring control
Time
line Idenfy components Judgments of Trigger a
and goal knowledge and memory
strategies search?
Inial Judgment Engage in solving?
of Solvability Give up?
Generate an inial,
autonomous, response Feeling of Provide the inial
Rightness response?
Analyc processing Reconsider?
Figure 1.1 Main reasoning and Meta-Reasoning components presented on a schematic time line
understand when an initial response is deemed adequate and when further reasoning is under-
taken. In sum, we argue that no theory of reasoning is complete without understanding the
metacognitive processes by which reasoning processes are engaged, formulated, and terminated.
4
Meta-Reasoning
Many studies have demonstrated that metacognitive monitoring of all types is sensitive to
heuristic cues (Koriat, 1997). For instance, retrieval fluency – the ease with which an answer
is retrieved – creates a subjective sense of confidence, even when actual performance may be
either unrelated or inversely related to it (e.g., Benjamin, Bjork, Schwartz, 1998; Rhodes &
Castel, 2008). In addition to relying on heuristic cues that may be misleading, people also
underweight factors that consistently affect performance. We know, for example, that repeated
rehearsal enhances learning; however, a more effective strategy is repeated testing. In addition,
instructions may guide learners to use effective strategies (e.g., imagining words and the relation
between them). In all these cases, the pronounced benefit in performance is underestimated in
learners’ JOLs and confidence judgments (e.g., Karpicke & Roediger, 2008; Koriat, Sheffer, &
Ma’ayan, 2002; Rabinowitz, Ackerman, Craik, & Hinchley, 1982). Similarly, people underes-
timate the impact of forgetting on their ability to recall information, predicting similar recall
performance for a test to be taken immediately after studying and one that would take place
after a week of retention (Koriat, Bjork, Sheffer, & Bar, 2004).
Recently, we proposed a Meta-Reasoning framework (Ackerman & Thompson, 2015), which
extends the metacognitive conceptualization to the domain of reasoning (see Figure 1.1, 2017).
In particular, we outlined several ways in which Meta-Reasoning processes share features with
their Meta-Memory counterparts. For example, the monitoring of reasoning processes relies
on some of the same heuristic cues shared with memory processes. Specifically, reasoners have
been found to rely on answer fluency; that is, the ease with which a solution comes to mind,
and the familiarity of the material being reasoned about (e.g., Markovits, 2015; Thompson
et al., 2011;Thompson, Prowse Turner et al., 2013). FORs are weakened by variables such as the
presence of conflicting answers (Thompson & Johnson, 2014) and by the presence of unfamiliar
terms in the problems (Markovits et al., 2015). The JOS, in contrast, was postulated to occur
before solving begins and to reflect an estimate of the probability of successfully finding a solu-
tion; it is most relevant to situations where an immediate answer to a problem does not come
to mind. Much less is known about JOS than about FOR. The initial indication is that they are
prone, like other monitoring processes, to relying on heuristic cues (Ackerman & Beller, in press;
Topolinski, 2014). We speculate that the initial Judgment of Solvability is related to the decision
to attempt solving at the first place, as well as to the length of time that the respondent persists
with solving attempts, once the process is initiated.
5
Rakefet Ackerman and Valerie A. Thompson
intermediate confidence monitors Type 2 processes. Final confidence reflects the assessment of
the chance of the chosen solution to be correct.
FOR. A strong FOR is a sign that further analysis is not needed (Thompson et al., 2011;
Thompson, Evans, & Campbell, 2013). Evidence for this assertion is obtained using the two-
response paradigm developed by Thompson et al. (2011): reasoners provide an initial intuitive
answer to a problem, often under time pressure, followed by an answer without time pressure.
The amount of time spent rethinking the answer, as well as the probability that the reasoner
changes the initial answer, varies with the strength of the FOR for the initial answer. This rela-
tionship has now been observed in a large number of reasoning tasks, ranging from simple to
complex.
Moreover, because the FOR is cue-based, the depth of subsequent analysis may not be
related to item difficulty. For example, answers that come to mind fluently may do so because
they are based on a heuristic process, which produces a strong sense of confidence in incorrect
answers (Thompson et al., 2011; Thompson & Johnson, 2014). Similarly, familiarity is a strong
cue to confidence, but is not necessarily diagnostic of problem difficulty (Markovits et al., 2015;
Shynkaruk & Thompson, 2006). In other cases, FOR may rightly signal problem difficulty, as
when the problem cues conflicting answers (De Neys, Cromheeke, & Osman, 2011; Thompson
et al., 2011).The point is that analytic processing is engaged in response to a monitoring process
that may or may not reliably track problem difficulty.
Whereas the preceding work was undertaken in a Dual Process context, here we demonstrate
the utility of extending a metacognitive analysis to two types of “fast and frugal” decision strate-
gies (e.g., Gigerenzer, Todd, & the ABC Research Group). The first one comes from a theory of
syllogistic reasoning developed by Chater and Oaksford (1999). Syllogistic reasoning is a type of
logical reasoning in which reasoners are asked to generate (or evaluate) conclusions from two
quantified premises, as the “beekeeper” problem above illustrates. Chater and Oaksford proposed
that rather than using logical rules or mental models to solve these problems, people rely on
simple heuristics based on information gain.The primary heuristic used to generate conclusions
to problems such as the “beekeeper” problem is called the “min” heuristic, which mandates rea-
soners to choose the quantifier of the conclusion to be the same as the least informative premise.
In the beekeeper example, above, reasoners would look at the two quantifiers (all and some)
and choose “some” because is it the least informative. Thus, they would conclude that “some
beekeepers are chemists”. In their meta-analysis of past data and in a new experiment designed
to test this theory, Chater and Oaksford (1999) demonstrated that reasoners did just that.
Whilst Chater and Oaksford (1999) demonstrated the importance of min for conclusion
generation, Thompson and colleagues demonstrated that the min principle also played a role
in how reasoners evaluate their confidence in a conclusion (Thompson et al., 2011). Using
the two-response paradigm described above, reasoners were asked to evaluate conclusions that
were either consistent or inconsistent with the min rule. They found that min conclusions were
processed more fluently (i.e., more quickly) than their non-min counterparts and consequently
gave rise to stronger FORs. The difference in FORs also had the expected down-stream effect
on analytic thinking: problems with min conclusions were subject to less re-analysis and fewer
answer changes than their non-min counterparts. In other words, using a fast-and-frugal strategy
such as min may give rise to confidently held responses that were less likely to be re-considered,
regardless of whether they were, in fact, logically valid.
Another potential extension of the FOR to fast and frugal heuristics is in experiments that
focus on information search. Many termination decisions are posited to be non-compensatory,
in that processing is supposed to be terminated whenever a satisfactory answer is found (see
Bröder & Newell, 2008; Hilbig, 2010, for reviews). An example is the “take-the-best” heuristic
6
Another random document with
no related content on Scribd:
sentiva spuntarsi una lacrima. Non poteva a meno di rievocare il
tempo in cui que’ suoi cari rallegravano il tetto domestico e
scherzavano sulle ginocchia materne. Ahimè, ormai la madre era
morta da un pezzo.... Involontariamente l’occhio del colonnello si
posava sulla parete ove sotto i ritratti di Vittorio Emanuele, di
Umberto, di Garibaldi, di Lamarmora, di Napoleone III, ecc., ecc.,
c’era un gruppo di fotografie di famiglia. La più antica e sbiadita era
appunto quella di sua moglie, una donna esile, dall’aria stanca e
sofferente. A fianco di lei Bedeschi in persona, in uniforme, con la
sua medaglia al valor militare sul petto, con la sua mano bravamente
piantata sull’elsa della sciabola. Qualche linea più sotto l’effigie di tre
giovinotti, Vittorio, Augusto, ed un terzo, minore di tutti.
Sicuro, c’era un terzo figliuolo, Federico, ed egli solo non s’era fatto
vivo in quel giorno, e da Londra, dove si trovava, non aveva spedito
nè una lettera, nè un dispaccio. Quando Bedeschi pensava a questo
ragazzo ch’era stato il suo preferito egli si doleva di aver ceduto una
volta tanto alle preghiere di sua moglie, la quale, impuntatasi
nell’idea che Federico fosse di salute cagionevole, aveva, tra gemiti
e singhiozzi, scongiurato il marito di non fargli abbracciar la carriera
militare come i fratelli, e strappatagliene un giorno a malincuore la
promessa, se l’era fatta rinnovare solennemente al letto di morte.
Federico era quindi rimasto in casa, aveva frequentate le scuole
pubbliche, ed era giunto senza gloria fino all’università. Non gli
mancava nè cuore nè ingegno, ma aveva uno spirito indisciplinato,
ripugnante a studi regolari, turbato piuttosto da vaghe inquietudini
d’artista. Onde nel bel mezzo del corso di legge gli era saltato il
ghiribizzo di darsi alla pittura, con grande sdegno del colonnello, il
quale nè amava l’arte, nè credeva a questa vocazione improvvisa.
N’eran seguite scene violente, per merito delle quali Federico aveva
finito col non studiar nè pittura nè legge e col menare una vita oziosa
e dissipata. Allora il padre gli aveva posto un dilemma. O mettersi in
grado di prendere la laurea entro un anno, o partire subito per
Londra, ove un antico compagno di cospirazioni del colonnello,
arricchitosi nel commercio, impiegava volentieri dei giovani italiani
per mandarli, dopo un tirocinio più o meno lungo, presso le sue case
filiali di San Francisco o di Sidney. Federico che della laurea non
voleva saperne accettò la seconda proposta; meglio far il minatore in
California o il pastore in Australia che incretinirsi su una scranna di
giudice o assottigliare il cervello nei cavilli avvocateschi.
E partì con una cert’aria spavalda che il colonnello, cattivo psicologo,
attribuì a perversità d’animo, mentre Federico, dal canto suo,
risentiva profondamente l’affettata indifferenza del padre. Come
avviene quando c’è un equivoco che non si chiarisce subito, la
freddezza reciproca andò a mano a mano crescendo; padre e
figliuolo si accusavano in silenzio di poco cuore e non si
scambiavano che lettere brevi, fredde e insignificanti.
Frattanto entrò in casa la Bice portando nella dimora solitaria un
nuovo alito di giovinezza, togliendole quell’aspetto triste e desolato
ch’essa aveva nelle prime settimane dell’assenza di Federico. Senza
volerlo, senza saperlo, la fanciulla nuoceva al cugino. Una frase
dello zio lo mise in guardia. — Non parliamo di quello scapato. —
egli le disse. — Ora sei tu che ne tieni il posto.
Tenere il posto di Federico? No, ciò non poteva, non doveva essere.
Ed ella dichiarò allo zio che prima che accadesse una cosa simile
sarebbe tornata in collegio, sicura di farvisi accettare dalla direttrice
come assistente.
Alla lunga si calmò, ma fermando il proposito di esercitar tutta la
propria influenza per sopire quel dissidio domestico. Pur non tardò
ad accorgersi che l’impresa era ardua ed esigeva infinite cautele.
Non le fu difficile mettersi in relazione con Federico, avendola lo zio
stesso incaricata talora di scrivergli in vece sua. E Federico le
rispose in principio diffidente e guardingo, poi, via via, più sciolto ed
espansivo. A lei rivelava la tristezza del suo esilio, l’acuta nostalgia
da cui era sovente assalito, la sua ripugnanza ad allontanarsi ancora
di più dall’Italia, la sua sfiducia assoluta di far buona prova nella
mercatura. Ma soprattutto le discorreva dell’arte, ch’egli aveva
ripreso ad amar con passione, che coltivava in segreto, e nella quale
avrebbe potuto forse non esser degli ultimi se gli fosse stato
permesso di dedicarvisi intero.
Il colonnello Bedeschi aveva tempra di despota, non d’inquisitore, e
avrebbe stimato inferiore alla sua dignità lo spiar le corrispondenze
della nipote. Delle lettere ch’ella riceveva da Federico egli sapeva
quel tanto che a lei piaceva di dirgliene, ed è naturale ch’ella gliene
presentasse un’edizione riveduta e corretta. Accennava alla
condotta regolare del giovine, al desiderio ch’egli manifestava di
riacquistare l’affetto e la stima del padre.... soggiungendo
timidamente che a parer suo non c’era ragione di tenerlo più oltre in
castigo a Londra, e meno che mai di spedirlo in capo al mondo.
La prima volta che la Bice toccò questo tasto, Bedeschi montò su
tutte le furie. — O ch’ella pretendeva di dargli lezioni? Ella, una
bambina, con quell’esperienza che aveva? Badasse ai casi suoi e
non s’impicciasse di ciò che non la riguardava. Se Federico le
scriveva delle sciocchezze, padrone; e padrona lei di rispondergliene
altrettante, ma non venisse a far la saccente. Aveva capito?
La fanciulla non si smarrì d’animo per questo rabbuffo nè perdette di
vista la sua meta. A ogni occasione opportuna ella tornava alla
carica, sopportando in santa pace le sfuriate dello zio, il quale, in
cuor suo, non si rammaricava troppo ch’ella difendesse il cugino. Ma
il colonnello aveva riputazione d’uomo forte, d’uomo inflessibile, e
certe riputazioni sono come un patrimonio da conservare. Bedeschi
non voleva che si dicesse ch’egli s’infemminiva cogli anni. Accadeva
poi un fatto curioso. Quantunque egli non osasse confessarlo a sè
stesso, la Bice gli diventava più cara per la sua generosità nel
prender le parti di Federico, e appunto col diventargli più cara gli
rendeva meno sensibile la mancanza del figlio.
S’era sbagliata strada. La Bice lo riconobbe e mutò tattica. Da due o
tre mesi ella non parlava di Federico che quand’era strettamente
necessario il parlarne, pareva rassegnata non solo alla relegazione
del cugino a Londra, ma anche alla sua partenza per Sidney o San
Francisco.
— È frivola e obliosa come tutte le donne, — pensava il vecchio
soldato. — Que’ suoi grandi ardori battaglieri sono sbolliti.
E non le sapeva grado della sua docilità. Era meno sicuro di aver
ragione dacchè nessuno gli dava torto.
Ella intanto ne pesava le parole, ne scrutava i silenzi, i gesti,
l’espressione della fisonomia, arrischiando di tratto in tratto con finta
ingenuità una frase, una domanda, come un generale che spinge
innanzi i suoi esploratori per esaminare il terreno.
— Se Federico deve lasciar l’Europa, — ella disse una mattina, —
suppongo che verrà prima a salutarci.
Bedeschi levò il capo con un movimento brusco. — Non so....
Forse.... Vedremo.... — E sentendo lo sguardo della nipote fisso
sopra di lui, si alzò da sedere e uscì dalla stanza.
La corrispondenza fra i due cugini durava non interrotta e non
vigilata. Federico tradiva spesso la sua impazienza, accusava la
Bice di non spiegar sufficiente energia per agevolargli il ritorno in
patria, dichiarava che assolutamente a Londra non ci poteva stare e
che avrebbe finito col fare un colpo di testa.... Poi, nella medesima
lettera, chiedeva scusa della sua petulanza e prometteva di seguire
a occhi chiusi i consigli della sua savia cuginetta, ch’egli si ricordava
in vestito da collegiale e che aveva giudizio da vendere a lui e a
molti altri meglio di lui.
E la savia cuginetta gli aveva scritto un giorno con gravità di esperta
diplomatica: — Un colpo di testa può anche esser necessario, ma
bisogna saper scegliere il momento di farlo. Il momento lo sceglierò
io.
II.
· · · · · · · · · · · · · · · ·
· · · · · · · · · · · · · · · ·