Organizational Change and Innovation

For further volumes, go to

http://www.springer.com/series/8833
Davide Secchi

Extendable Rationality
Understanding Decision Making
in Organizations

123
Davide Secchi
Department of Management
University of Wisconsin-La Crosse
La Crosse, WI 54601, USA
secchi.davi@uwlax.edu

ISBN 978-1-4419-7541-6 e-ISBN 978-1-4419-7542-3

DOI 10.1007/978-1-4419-7542-3
Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2010937275

Springer Science+Business Media, LLC 2011

All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York,
NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in
connection with any form of information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are
not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject
to proprietary rights.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

A Zia Brunella e Nonno Egeo
Perch ogni decisione che prendo
inevitabilmente sottoposta, nella mia mente,
ai test di realismo e umanit che mi avete
insegnato.
Preface

Writing a book is not very much appreciated in todays academic world and it is not
popular by any means. Scholars (including myself) are oriented toward writing and
publishing journal articles, that is where the newest advancements in the field are
often found. For this reason, the number of academics that rely on books for their
research is lower compared to those that lean on articles. According to Thompson
(Chronicle of Higher Education, 2005, Vol. 51, Issue 41), in the 1970s the number
of copies per monographs that US and UK publishing companies would print was
somewhere between 2,000 and 3,000. Expectations were to sell the most part of
them. That trend has changed dramatically. Today, publishing companies print 400
or 500 copies of a monograph and hope to sell most part of them. Some publishing
companies print on-demand and make the text available as an e-book. The decline
in monograph publishing is only limitedly attributable to budget cuts in libraries,
financial independence for academic presses, or for business conglomerates. There
has been an increasing lack of support for this kind of publishing from scholars. This
is my guess on the evidence reflected in the numbers that Thompson mentions in his
book and in an article that appeared in the Chronicle of Higher Education. You may
disagree with my interpretation, but you cannot disagree about the data that show a
declining trend for monographs and on their lack of popularity in todays academia
(general management and organization behavior fields are not excluded).
Another factor that supports what is written above is that an academic career is
usually not affected by book writing as much as it is by article writing. This element
may vary depending on the discipline, but it has expanded from the hard science to
all the remaining fields.
On top of this, writing a book involves a type of activity that is very different from
what it takes to write a journal article. The latter is usually based on one (hopefully)
original and innovative idea that is empirically tested or validated through a theoret-
ical model. The former contains several ideas, organized in a web of connections.
They are two outcomes of research and creativity that require different efforts, time,
and dedication. Of course, I am not concerned with those monographs that are col-
lections or that rewrite previously published articles. These are not original and add
little to the spirit of a research effort. Everyone that earned a PhD knows the dif-
ference between writing a dissertationthat is very close to a bookand writing

vii
viii Preface

an article. The dissertation may well be a collection of articles, where the candidate
explains what the links are among them. Even in this case, dissertations and books
usually require more effort, if not more time.
To sum up, monographs (1) are not supported nor considered for ones academic
career, and (2) they require more effort and time. It seems that my decision to write
a book makes little sense. If I add to these reasons that I am an assistant professor
at the early stage of my career, this decision makes even less sense. Was it a rational
decision? Why did I write this book then? Didnt I care about my career and the
academic community I belong to?
From what I wrote, it is apparent that the reasons for this book should be found
elsewhere. In fact, this is an attempt to ask some questions on rationality and deci-
sion making. Given what I wrote above on monographs, my decision to write this
book should in itself be analyzed under the lenses of rationality. Is it rational for a
young scholar at the early stage of his career to write a monograph? Well, besides
writing that I really hope so, in the following I present the reasons that brought me to
the decision to write this book. The entire book may be used as a tool to understand,
among many others, if the reasons I provide below are rational or not.
The reason underneath this project is tied to my research plan. I needed an
outlet to present and analyze how the well-known theory of bounded rationality
changes when its assumptions are modified. This effort cannot be contained in a
single journal article, nor can several together give the breadth or the continuity
that are needed. This is the rationale for the book. Bounded rationality is the idea
that individuals have (a) limited computational capabilities (i.e., internal bound) and
(b) limited access to information (i.e., external bound). As far as I know, there are
no (recent) books that discuss foundational assumptions of this theory. Attempts to
find a different perspective or to improve the existing framework are limited. The
book aims at introducing some concepts that have the potential to redefine bounded
rationality.
For almost five years Emanuele Bardone and I have been planning to write a book
that could summarize most of our efforts in understanding rationality and develop-
ing the bounded rationality theory. Ever since I moved to the United States, this
distant cooperation worked fine with respect to articles, but we found it hard to write
a book. We still plan to do that in the future. In the moment I am writing, both of
us are about to submit books to the attention of some publishing companies. When
Emanuele and I started to discuss rationality and decision making, it took both of
us awhile before getting a common understanding of what we were saying. He is
a cognitive scientist, I am a management scholar. At first, we just couldnt under-
stand each other. Anyway, I hope that this work will be the first of many attempts
to present these ideas, and that we can soon add more writings on our common and,
we believe, very interesting and intellectually challenging projects.
Second, it happened that one semester (spring 2009) I taught an MBA class at
the University of WisconsinLa Crosse together with Thomas Krueger, professor
of finance. The course was Decision Framing II. I have no idea why somebody
decided to use the word framing instead of making; anyway, this is not rele-
vant here. What is important is that Tom was in charge of teaching students how to
Preface ix

make decisions in uncertain financial environments, and I was in charge of provid-

ing students with an introduction to decision making. In particular, as a professor of
organizational behavior, I had the idea to put all my efforts toward defining how peo-
ple make decisions inside organizations. I thought this part should have been very
easy, although I wanted to give students perspectives that were current (e.g., sense-
making, embodied cognition, distributed cognition, and more). I looked at books on
this topic, but they ended up being too narrow, too broad, or too simplistic. I wanted
something that fitted my needs. I couldnt find anything. After that, I started col-
lecting articles on the topic and soon I realized that what I wanted to teach stayed
together with difficulty. Then I tried to mix chapters of books with academic papers,
but still the material remained too heterogeneous. At that point I realized that it
was hard for students to read from all of these different sources without ending up
being confused. Problems were related to two major points: (a) academic papers are
not the easiest thing to read (sometimes even if you are in the profession!) and (b)
jumping from author to author makes it hard to follow a common line of thought.
Although the first part presents a quasi-standard reading of bounded rationality,
the second part is too explorative and far from mainstream research. If used in a
class on decision making, for example, the book needs to be associated with more
standard readingsas I personally do in the classes I teach. It is probably more
suitable for doctoral seminars.
Third, early in January 2009 I was stuck in a situation that I didnt want and that
I didnt expect. I accepted a tenure-track position at the University of WisconsinLa
Crosse on January 7, 2007, and started working for that university that same year on
August 27. While I was getting a few papers published, I found it difficult to study
and write on new topics or even to continue with my older streams of research. The
reason was that here at UW-L we teach three classes per semester and, if I include
summer sessions, in the first year I taught seven classes, most of them for the first
time. The following fall I was stuck with four classes, two of which I had never
taught before. I like teaching, but not at this rate. When my third semester at UW-L
finished the only thing I could do was to recharge my batteries. I asked myself Is
this why I moved to the States? Is this what I wanted to do in the profession? Teach?
The answer was crystal clear in my mind. I decided to enter this profession because
I have always had a passion for research and I know I cannot live well without
my usual studying, reading, thinking, opinion exchanges, and writing. I needed to
change something, and I needed to do it quickly. I decided to push hard on research,
no matter what. While the teaching load for the coming semesters hasnt changed,
my research plan for the year 2009 was to have at least one publishable outcome per
month.
The first two months of the year were particularly productive. I wrote a total
of six papers between January and February, well beyond my plans and expec-
tations. Although I have other projects going on and I know that a book is not
the best outcome for a tenure-track guy, I also thought that I have never written
for career purposes only; I write because I think I have interesting questions to
answer. Moreover, sometimes there are concepts that dont fit into one, two, or sev-
eral papers. I thought a book should have been a good product to start with in March.
x Preface

I ended the first draft in April. I was still tied to my writing and research plan; also
I built some foundational concepts for the future. I know that this is probably some-
thing you usually dont write in prefaces, but this has been a starting point in my
decision making.
Although I wrote this book having my students in mind, this is not a textbook.
There are several aspects that have potential interest for organizational behavior and
decision making scholars. In the first chapter I go into further detail and explain
what is new with the approach to decision making you are about to read, and then
conclude with a short overview of the books contents.
As far as the rationality of writing a monograph is concerned, I believe that
these few pages (and Chapter 1) help clarify why I came out with the unpopular and
improbable decision to write a book.

La Crosse, Wisconsin Davide Secchi

Acknowledgments

Multiple sclerosis is a terrible disease that hit my family very hard. For this rea-
son, all authors royalties will be paid directly to Fast Forward, a subordinate
organization of the National Multiple Sclerosis Society (USA) that focuses on
expediting the drug development process, bridging the gap between promising
discoveries and the commercial expertise and funding to move them forward
(http://www.nationalmssociety.org/fast-forward//index.aspx). I wish to thank the
publisher, Springer USespecially Nick Philipsonthat handled this with extreme
flexibility and had gentle and supportive words for me. Also, Timothy Coetzee and
Carol Miller from Fast Forward were very kind and worked over the weekend to
match the publishers deadline.
There are many people that contributed, and in many different ways, to every-
thing it took to write and publish this book. It goes without saying that, despite all
the help, I hold myself responsible for shortcomings that still are in the book.
Everybody at the Department of Management at the University of WisconsinLa
Crosse supported this effort, and I wish to thank every single colleague for slightly
different reasons. William Ross read the first draft of the manuscript and gave me
the most useful insights and suggestions that I ever received on this book. It is not
easy to find colleagues that are so much dedicated to research that it does not matter
if it is their own or somebody elses. William is one of these amazing persons.
Gail Gillis also read the original version of the manuscript and gave it back to me
with her very useful notes on it. Given her very busy teaching schedule, I really
have no idea when she found the time to read the book. For exactly this reason, I
express double appreciation and a million thanks for her help. I wish to thank John
Betton and Tom Hench for challenging discussions, carried over on a regular basis
and on an incredibly wide range of topics. Leticia Pena offered me to go to Caen,
France, where I had the idea and wrote the book. I have shared with Drew Stapleton
some of the ideas that are included in this book and benefited from his comments.
Tom Kuffel, former Chair of the Department, has done his best to let me have all
resources that the department could afford in order for me to complete the book
or whatever I was doing at that time. Also, our discussions were like fresh air in
the desert! Lori Komarek did a wonderful job simplifying everything that could be
simplified to ease my job at the university.

xi
xii Acknowledgments

My friend and colleague Thomas Krueger deserves a special mention here. We

teach together the MBA class where I talk about many of the topics that you will find
in this book. During my part of teaching, Tom always sits in the class and provides
me with useful and original insights in the form of questions. Some of the points
raised by Tom have been included in the text. Tom has an inquiring mind, and it has
been a real pleasure to work with him.
Many others helped with their thoughts, words, and other issues that I happened
to connect to the manuscript I was working on. Among the many that served this
purpose are Bruce May, Bill Colclough, Kuang-Wei Wen.
MBA students at both the IAECentre Franco Amricain, Universit de Caen
Basse Normandie, and the University of WisconsinLa Crosse were always very
open to ideas that we discussed in class. They helped me significantly understand
the strengths and weaknesses of the various topics. Among those that deserve special
thanks for sending their comments back to me are Michael Keith, Roman Yeskov,
and Chen Yi-Jui.
Many of the ideas that are in this book came out during discussions with my
friend Emanuele Bardone, mostly at lunches and afternoons when we were PhD stu-
dents at the University of Pavia, Italy. Those incredibly rich and in-depth discussions
that lasted hours are the solid base for this book. Also, I am indebted to Lorenzo
Magnani for introducing me to the world of abduction and distributed cognition; I
have been touched by his humanity and deep mastery.
My thanks and appreciation go also to the editor of the book series on
Organizational Change and Innovation, Ann Gilley, for her comments. I bene-
fited significantly from suggestions and from notes by Nicolai J. Foss, who served
as reviewer of the original manuscript. His insights on several parts of the book
certainly led to an improved version.
Nick Philipson, Executive Editor at Springer US for the area Business and
Management, has always been very supportive and active in promoting the idea of
the book within Springer and with the book series editor. I enjoyed very much our
e-mail correspondence and his willingness to keep me updated and current at every
single stage of the process. Charlotte Cusumano, Editorial Assistant at Springer US,
was always there to answer all of the (sometimes silly) questions I might have had.
Thanks so much for making this publishing experience very smooth and extremely
positive!
On a more personal basis, my fiance Claudia deserves very special thanks for
all her loving support and care for my person when I was too much involved in
reading, thinking, and writing. She even gave up some time of her favorite activity
shoppingwhen we were in Caen, France, to stay with me while I needed to
discuss some ideas. She also read every single word I wrote and carefully edited
the manuscript. I could not have had the luxury to focus on the book the way I did
if it wasnt for her.
Also, my mothernow Dean of the Faculty of Economics and Business in my
home town, Cagliarihas always supported me in everything I wanted to do. Being
herself an author of several books, her comments and suggestions on an early draft
of the manuscript have been extremely helpful. My father has been a safe place
Acknowledgments xiii

to land when I needed a break from work and other too-serious thinking. Thanks,
dad. My two brothers contributed to this book too. As a PhD student in supply chain
management, Enrico has always been very critical of whatever his older brother was
naively thinking. His challenges are always very important to me. Marco helped me
in a subtler way; his and Zio Ninnis relentlessness, passion, and dedication when
they started up what now is a very successful brewpub businessIl Birrificio di
Cagliarishowed me actual problem solving and decisions in the making. They
extended their rationality very often, especially during the first months, when uncer-
tainty was incredibly high. Notwithstanding the fact that Marco and Zio Ninni were
my guinea pigs, they pour their tasty and unique beer for me every time I visit
them. For free!
A special thanks goes to my most precious friends Patrick and Gamze Randolph,
Plator Ulkinaqu, and Raffaello Seri for our endless discussions on this and many
other subjects.
Last but not least, I want to extend my thanks to those of you that decided to buy
the book and are eventually reading it. I really hope you will enjoy it.
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
What to Expect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Book Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Part I The Limited Cognition

2 Rationalization and Rationality . . . . . . . . . . . . . . . . . . . 9
Kinds of Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
The Legacy of Herbert Simon . . . . . . . . . . . . . . . . . . . . . 13
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 Bounded Rationality . . . . . . . . . . . . . . . . . . . . . . . . . 19
What Is Bounded Rationality . . . . . . . . . . . . . . . . . . . . . . 19
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4 Maps of Bounded Rationality (I) . . . . . . . . . . . . . . . . . . . 27
Prospect Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
This Is a Biased World! . . . . . . . . . . . . . . . . . . . . . . . . . 31
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5 Maps of Bounded Rationality (II) . . . . . . . . . . . . . . . . . . 41
Heuristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Accessibility, Representation, and Framing . . . . . . . . . . . . . . 48
Two Logics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Implications of Using One or More Maps . . . . . . . . . . . . . . . 54
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Part II The Extended Brain

6 Simons Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Distributing Cognition . . . . . . . . . . . . . . . . . . . . . . . . . 64

xv
xvi Contents

How Bounded Is Rationality? . . . . . . . . . . . . . . . . . . . . . 74

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
7 Stretching the Bounds (I) . . . . . . . . . . . . . . . . . . . . . . . 81
Through Doing Decision Making . . . . . . . . . . . . . . . . . . . 82
The Rationality of Change . . . . . . . . . . . . . . . . . . . . . . . 89
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
8 Stretching the Bounds (II) . . . . . . . . . . . . . . . . . . . . . . 97
The Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Advice Taking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Passive Advice Taking . . . . . . . . . . . . . . . . . . . . . . . . . 108
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
9 The Docile Organization . . . . . . . . . . . . . . . . . . . . . . 113
The Docile Individual . . . . . . . . . . . . . . . . . . . . . . . . 113
Levels of Docility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Understanding Docility . . . . . . . . . . . . . . . . . . . . . . . . . 123
What Is a Docile Organization? . . . . . . . . . . . . . . . . . . . . 129
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
10 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
The Point on Rationality . . . . . . . . . . . . . . . . . . . . . . . . 135
What Are We Mapping? . . . . . . . . . . . . . . . . . . . . . . . . 137
The Individual and the Group . . . . . . . . . . . . . . . . . . . . . 138
A Methodological Note . . . . . . . . . . . . . . . . . . . . . . . . . 141
Extendable Rationality . . . . . . . . . . . . . . . . . . . . . . . . . 143
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Afterword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
List of Figures

8.1 Information richness scale . . . . . . . . . . . . . . . . . . . . . . 102

8.2 The shared meaning of communication . . . . . . . . . . . . . . . 103
8.3 Information richness fit . . . . . . . . . . . . . . . . . . . . . . . . 105
8.4 A low richness medium . . . . . . . . . . . . . . . . . . . . . . . . 106
8.5 A high richness medium . . . . . . . . . . . . . . . . . . . . . . . 107

xvii
List of Tables

3.1 A combination of processes and outcomes . . . . . . . . . . . . . . 20

4.1 Common errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.1 The Gang of eighteen . . . . . . . . . . . . . . . . . . . . . . . 50
8.1 Judges options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
8.2 A judge-advisor system (one type of advice) . . . . . . . . . . . . . 100
8.3 A judge-advisor system (two types of advice) . . . . . . . . . . . . 100
9.1 Characteristics of individual docility . . . . . . . . . . . . . . . . . 123

xix
Chapter 1
Introduction

The first chapter of a book should be on how and why the author (me) decided to
work on this topic, and made the decision to start writing. Ironically, it should take
a whole book to explain how decisions of this sort happen. And this is exactly the
reason for this book: explain the how and why of decisions.
The decision to write this book in particular relates to the need that, I believe, the
academic community has of it. There are four major points that are addressed in the
book:

1. In an article published in the Journal of Economic Psychology in 2003b, Nicolai

Foss suggests that the idea of bounded rationality is much cited and little used.
This implies that there is a limited interest in foundational questions among
management scholars and behavioral scientists. The book provocatively ques-
tions the nature of bounded rationality and of decision making. The need for
conducting this study is dictated by the fact that science, and especially man-
agement, has changed since Simon presented his theory of rationality more than
60 years ago (1947). Does this relatively old theory fit todays management the-
ories? Does it fit advancements in cognitive science? The book is an attempt
to answer these questions. And the answer is negative: The theory needs to be
updated. However, if the reader thinks that ideas presented in the book do not
provide a plausible answer to these questions still, the book may be useful for
raising questions on what theory of rationality suits recent advancements in sci-
ence. Going back to Foss, I believe the questions raised in the book are much
needed, little expected.
2. The book presents an application to management and organizational behavior
of the distributed cognition approach. Given the fact that this is a new and
emerging approach, this is the first book that tries to connect it to management.
What Hutchins (1995) wrote in his book Cognition in the Wild may be help-
ful for management scholars, and it is an interesting starting point. However,
Hutchins is a cognitive scholar, and this remains his approach throughout the
book. An attempt to bring distributed cognition close to management has been
recently done by Michel (Administrative Science Quarterly, 2007). This book
takes Michels study one step further, trying to answer background questions

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 1

DOI 10.1007/978-1-4419-7542-3_1,  C Springer Science+Business Media, LLC 2011
2 1 Introduction

and to provide readers with a broader picture of how this approach may change
our perspectives of managerial rationality and decision making.
3. There is a limited number of studies on docility (see Chapters 7, 8, and 9). I use
this concept to connect cognition to behavior. In doing that, the book is one of the
few to address socially based decision-making procedures to management. It is
very unusual that authors go beyond the atomistic and individualistic views of the
decision maker. In fact, those who study socially based decision making end up
studying group dynamics. One of the most significant arguments advanced in the
book is that no decision is (can be) made in complete isolation. The consideration
of advice giving and taking, through decision-making processes, and docility
is an attempt to fill this gap. The need for this gap to be filled relates to the
need for having theories that are closer to the ways individuals actually make
decisions.
4. The book tries to connect bounded rational decision making to epistemology.
While philosophers (epistemologists) are aware of psychology and manage-
rial studies, we cannot say that the relation works the other way around. I
decided to make this philosophical connection clear (e.g., logical fallacies,
abduction).
The idea of an extendable rationality emerges when the nature of bounds
is considered: What if human rationality is shaped by external resources? What
if there is no clear divide between internal and external bounds? How can we
analyze decision making if rationality is extendable? What is the role of
social resources (or channels) in decision-making processes? Are innovation
and change processes in organizations better understood through the extendable
rationality approach? What are the implications for organizations?
The approach presented in this book aims at bringing to models of human
rationality the adaptability and flexibility that can be observed when individuals
face change and, especially, innovation. One of the paragraphs of a later chapter
is dedicated to studying how high-tech innovations shape human rationality.

What to Expect

For a large part, the work on decision making has two roots: (a) formal (e.g., game
theory) and (b) behavioral (e.g., prospect theory). There are still few works that
combine the two, although cross-disciplinary research is growing. These approaches
maintain their abilities to explain decision making, to analyze how these processes
work, and to provide multiple resources to forecast the outcome of decisions.
Notwithstanding all of these significant advantages, there are many ways to enrich
these classic approaches to decision making. The objective of this book is to spec-
ify what domains can contribute to this enriching of decision making and how they
can do that.
On the one hand, the formal perspective reached its highest levels with the
subjective expected utility (SEU) theory of von Neumann and Morgenstern,
What to Expect 3

and continued to inspire and define theories of decision making after that. The
rational choice approachi is related to that first attempt. The idea of defining a men-
tal process strictly on the basis of mathematicsii has never left the decision-making
field of study. Moreover, utility continues to be the concept on which most economic
theories lean on. Decision trees, cooperative and non cooperative games, and many
other approaches and theories shape the field of decision making at a very deep
level. It is a shared belief that theories of bounded rationality also share with SEU
some of its basic assumptions and especially its methodologyiii , which has remained
unchanged for too many years.
On the other hand, the behavioral root has started with a critique of the formal
approach, trying to make decision making more human. Behavioral economics (and
especially behavioral finance) provides descriptive theories of how individuals act in
real-life settings. We now have an ever-growing number of experiments, field stud-
ies, and all kind of dataiv supporting the hypothesis that human beings are not even
close to the calculator-like metaphorv . This field of study brings psychology and
sociology into play so that the focus is on prejudices, biases, anomalies, mistakes,
heuristics, and on the many imperfections that characterize human decision-making
processes.
This manner of picturizing the whole field of decision making is probably too
simplistic. However, it offers many advantages. First, it allows us to define two
approaches to the same problem: explaining how people make decisions. Second,
although convergences can be found between the two rootsfor example, Herbert
Simon, one of the fathers of the behavioral approach, is also deeply immersed
in the formal rootstudies that focus exclusively on either the first or the sec-
ond root are different indeed. On average, mathematicians, statisticians, engineers,
cognitive scientists, and operations and production management scholars approach
decision making from a formal perspective. Organizational behavior scholars and
psychologists study the behavioral side of these processes, on average. Economists
are divided among the two: They are agreeable enough to find good company
everywhere!
The fusion of these approaches brings new subfields, new research agendas, and
new ideas. For example, neuroeconomics and social decision making are two sub-
fields that have the potential to change the ideas we have of decision making one
more time. This research allows us to think about economic processes in new ways.
We have the opportunity to found the theory on observed neural mechanisms, having
a more precise idea of what determines a specific feeling, for example, that leads to,
stays together with, or comes after a certain decision. This gives scholars the poten-
tial to have more accurate descriptions and forecasts on how thinking and behavior
are actually linked.
Although very promising and interesting, this book is not about neuroeconomics.
In recent years, the two perspectives have contributed greatly to the development of
our understanding of decision making. However, they tend to focus too much on
the way the respective frameworks support or reject specific hypotheses on human
behavior and/or theoretical analysis. There is nothing wrong about this. Quite the
contrary: This is the way science improves! However, in the last decades, there has
4 1 Introduction

been a lack of foundational questions. I refer to those questions that, for example,
lead Herbert Simon to define the concept of bounded rationality. Who questions this
concept today without falling into either the behavioral or the formal paradigm? It
is far from the purpose of this book to pretend to present a new theory that has
an impact similar to those of the past. All I can do is to ask questions. And this
is what I will be doing in this book. In so doing, I present a different approach to
decision making that integrates cognition and pro-social behavior (including neu-
roeconomics) to bounded rationality. The book presents bounded rationality and
analyzes its limitations.
The objective of this work is to show that, if we focus on bounds, we have a
very limited and inaccurate representation of human rationality. If we think of great
achievements that individuals reached in the past, what comes to our minds first?
What intrigues us?
It is a legend that when Gdel demonstrated his incompleteness theoremvi , he
started writing in silence on the board during one of the most important math con-
ferences of his time. When he was done showing his colleagues what was about
to change the history of mathematics, the tribute of the hall was a very intense
applause. Creativity and originality come together with Gdels theorem and the
way he delivered it to his colleagues. I guess we do not think of his bounds and how
limited his rationality had been in what he was able to achieve. What makes Gdels
work so important and what allowed him to prove a theorem that nobody else had
been able to prove before is based on his ability to think outside the box. Gdels
brilliant demonstration comes from the ability to challenge ones own limits, and to
go beyond them. It is not about bounds, it is about what makes an excellent mind
overcome its limitations.
Overcoming ones own limitations is not typical of excellent minds only. If you
think about it for a second, you know that everyone of us goes past his or her lim-
itations. How? Take a simple task such as to write a 1,000-word review of a book
you have just finished reading (lets say Extendable Rationality). Writing is a typical
activity that requires external support. A personal computer is something that could
help you with proper software that has a template that fits your needs, with help
functions showing you how many words you have written, and with a vocabulary
that supports your writing. Depending on the software you are using, you can acti-
vate a grammar and syntax assistant, a function that signals if the sentence you are
typing is correct in standard American or British English. If somebody is around,
you can also ask for comments and opinions on what you are about to write. What
is this about? This is a way that you enrich and expand your rationality in order to
reach a better outcome. This process is that of exploiting external resources, and it
is what increases our chances of being rational. We overcome our limitations when
we exploit external resources.
What if there is no real divide between internal and external resources, between
our brain and the tools and artifacts that we use? This book explores exactly this
hypothesis, showing that if we remove this divide we can explain a significant part
of human behavior (and thinking) and redefine rationality.
Book Structure 5

Book Structure
I do not pretend to offer a shared interpretation of the concept of bounded rationality,
nor do I think that what the reader finds in this text is close to what other scholars
are thinking. I have found very limited evidence that other scientists are thinking
about the nature of bounded rationality. The most common sporting activity is that
of taking it for granted, on the basis of an ipse dixit of medieval flavor. I strongly
believe that this is a luxury that scientists cannot be allowed to have and, for this
reason, I support the idea that any critique and attempt to better understand limits as
well as potentials of any theory are extremely welcome. I was particularly surprised
that critics of bounded rationality only came from economists who were in love with
neoclassical economics. Few behavioral scholars think that the understanding of its
limitations is what makes the theory stronger.
This work presents my interpretation of bounded rationality and some of the
problems of this theory. However, my criticism is not designed to take the field of
decision making backwards (i.e., neoclassic theory), but instead looks at the future.
This doesnt mean that I expect to be right with what you are about to read, but I hope
that a debate on the meanings and perspectives of bounded rationality could start.
The ideas presented in this book are an attempt to improve what defines bounded
rationality.
The book is divided into two parts. The first part is The Limited Cognition, from
Chapters 2 to 5, dedicated to the analysis of bounded rationality. The second, The
Extended Brain, Chapters 6 to 9, presents the idea of extendable rationality. Chapter
10 offers a summary of the second part and includes implications, a research agenda,
and concluding remarks.
What a rational process is and how it is connected to rationality is the topic
of Chapter 2. The starting point is that of defining decision making on the basis
of three types of decisions: (1) mechanical, (2) decisions that imply a choice, and
(3) creative. In short, the question that I try to answer in this chapter is How do
we make sense of our decisions? Individual processes of making sense of ones
own behavior and thinking start from there, and rationality can be analyzed as a
distinct, but associated, concept. The thesis that will be presented is that the tra-
ditional idea of rationality is a good fit for mechanical decisions and for those
that imply a choice, but creative decisions are less capable of being explained
that way.
In Chapter 3 I introduce the idea of bounded rationality in a classic fashion. The
chapter answers the question: What is bounded rationality, and why is it so impor-
tant? Concepts such as procedural and substantive rationality, satisficing versus
maximizing, and the nature of rational bounds are presented and discussed in the
chapter.
Chapters 4 and 5 present an overview of selected studies that map bounded ratio-
nality. Starting with the notorious prospect theory, I present some of the most typical
analyses, including biases, heuristics, representation, accessibility, and framing
processes.
6 1 Introduction

Chapter 6, provocatively entitled Simons Error, introduces the reader to the

concept of distributed cognition, and presents critiques of bounded rationality
based on stylized and simplistic assumptions known as the cognitive divide. The
question that I try to answer there is What are the limits of bounded rationality?
The next two Chapters (7 and 8) then explore the idea of unstable bounds of ratio-
nality, and there I present a different perspective on decision making. They are based
on the question How can we move forward? Chapter 9 continues with a concept
that is close to both distributed cognition and extendable rationality, and puts these
two into play when we analyze organizations. It is the concept of docility that is an
answer to Is there a theory of human behavior and extendable rationality?
Some chapters present an epistemological corner. This is a cognitive perspec-
tive on some of the problems I mention in the book, and helps the analysis and
understanding of the concepts presented here. It gives a multidisciplinary perspec-
tive from a field of study, that of the philosophy of science, where some of the topics
under analysis have a long tradition of study. The decision to skip and not read the
epistemological corner does not hinder the ability to understand the books theme
and follow its line of argument.
If you are a student of bounded rationality, you may read directly the second
part of this book with no prejudice to your understanding. Or, you can read the
ending paragraph (summary) of each chapter that reviews materials there presented.
However, I suggest starting from the beginning, because you may probably want to
see what my starting points are and what my reading of bounded rationality is. Have
a good trip.

Notes
i. Simon (1979) provides a review of early rational choice approaches. See also Gilboa (2010).
ii. This is true also for what concerns works on bounded rationality (Mousavi and Garrison,
1992; Patokorpi, 2008).
iii. Zelen (2001), Mousavi and Garrison (1992), Patokorpi (2008) Peng (1992) Sent (1997, 2005)
and Langley et al. (1995).
iv. See Camerer (2007) for a survey of what is the recent evolution of behavioral economics,
called neuroeconomics.
v. The computer metaphor of the brain is analyzed and discussed in Patokorpi (2008).
vi. An explanation to the theorem for non-mathematicians is offered in Nagel and Newman
(1958).
 Part I
The Limited Cognition
Chapter 2
Rationalization and Rationality

This is our starting point: Rationality. If you want to understand decision making,
the first step is to define what a rational choice is. This has always been the first task
for students of decision making.i However, it doesnt have to be like that.
Some can argue that decision making is not the same as rational decision making,
and that many of our everyday decisions are not rational at all.ii Think of when you
get into the grocery store to buy water and bread only. How could it be that every
time you get out from the supermarket you find yourself with a seemingly infinite
list of products? And, by the time you are home you realize that you didnt need
anything besides bread and water. Still, the next time you get into the store you
will make the same mistake.iii We cannot say that buying unneeded products is a
rational decision, but the process that is involved in making the decision has to deal
with rationality. This is the link that is explored in this chapter.

Kinds of Decisions
Every day, every one of us makes a number of decisions. These decisions are
not always the same; they vary depending on circumstances, importance, events,
involvements, and many other factors. The first step is that of making a distinction
between different types of decisions. It is important to discern that not all decisions
are equal, and that some of them differ from the others in terms of efforts of the
decision maker and effects deriving from their implementation. In order to have a
clear understanding of what I mean, we may find decisions that

are mechanical, automatic, or immediate;

imply a choice or are non-mechanical;
are creative.

Mechanical Decisions

The first kind of decisions is those that happen without thinking.iv These are
widely diffused in our everyday behavior and often we take them for granted. We

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 9

DOI 10.1007/978-1-4419-7542-3_2,  C Springer Science+Business Media, LLC 2011
10 2 Rationalization and Rationality

never think about these decisions in a proper and specific way because they come
out as usual. For example, I am pretty sure that you never make the decision to put
some clothes on before going to school, work, university, or wherever you want to
go. You assume that you have to dress and you do that. However, you dont actu-
ally make the conscious, deliberate decision to get dressed, you just do it. This is a
mechanical decision, where you dont go through a decision-making process that is
well-defined and distinct. In the example, the decision to get dressed has roots that
go to a period of your life that is undefined and that you dont remember. The thing
is you continue to practise the way you learned: You get dressed.
Other examples of mechanical decisions could be related to everything you do
without any deliberate decision-making process, such as eating, the way you tip,
how you write (or dont write) the first line of your e-mails, what you say to people
when you meet them, things you do everyday when you have breakfast. From a
personal point of view, these are all examples of habits.v This doesnt mean that
there is no process here, it means only that the decision-making process is embedded
in the way you behave or think.
We have many examples of mechanical decisions in organizations too. A typical
case could be that of sending invoices, or that of sending monthly emails with earn-
ing statements to employees. Everything that could be defined as an organizational
routine falls into a mechanical decision-making process.vi The way to identify it is
the same that we used at the individual level, but it has more complicated implica-
tions. A routine is something that deals with implicit knowledge and that is done
on the basis of what has been done in the past and written rules. It is a convenient
and efficient way to make quick decisions. Sometimes it is in the interest of the
company to break the routine, especially when it emerges in areas where it is not
needed. For example, routines are very welcome in call centers. Here, people dont
need to think too much,vii they need to provide information/solutions to customers
according to corporate guidelines. Nothing outside the guidelines is admitted. In
other terms, the routine (and standard operating procedures) is what makes the call
center work. Other departments of the company suffer when too many routines are
set up and when people dont step out of the so-called ordinary business. Research
and development teams as well as marketing departments are usually well aware
that too many routines are not good ways to achieve excellent performances. This is
to say that companies usually tend to avoid mechanical decision makingviii in key
areas.
Now, these are examples of mechanical decisions at both organizational and indi-
vidual levels. In summary, a mechanical decision is a particular behavior or thinking
attitude that people practice because they are used to it.

Decisions that Imply a Choice

Decisions that imply a conscious, deliberate choice are of a completely differ-

ent type. This means that we evaluate possible alternatives and then select one.
Kinds of Decisions 11

Evaluations that we bring in might range from trivial, or fairly simple to very
complicated or complex. We may face decisions that imply simple alternatives.
To stay with the example above, while the decision to get dressed is mechanical,
that of how to get dressed, i.e., what clothes to put on, is not. The latter deci-
sion implies a choice. For most of us this is a very simple choice; it may become
more complicated if we get into the domain of color matching (e.g., socks with
belt, pants with hat)where my fiance Claudia excels and I make miserable
mistakes.
The difference between the two types of decisions is not related to alternatives,
though. It is not that when we are involved in mechanical decisions we dont have
alternatives, or while in decisions that imply a choice we do. The point is that we
dont see alternatives in the first case while in the second the decision is based on
alternatives. The choice implies that we select among different options; hence, the
existence of these options is not a necessary condition for the choice to be made
while it is a sufficient condition. In other words, it is the individual that defines the
kind of decision, not vice versa. It is not that certain types of circumstances lead you
to make a mechanical rather than a choice-based decision. It is your actual involve-
ment in the process that defines what decision you are making. If you are involved
in selecting, weighing, and choosing among alternatives, then you are in the second
type (i.e., the decision implies a choice); if you dont see alternatives and you just do
what you are about to do without further thinking, that is a first type decision (i.e.,
mechanical).
There is continuity between the two types of decisions. We can switch from the
first to the second or vice versa as needed. And the point is that we need to do that,
on some occasions. Let us consider a few examples.
What is a decision that implies a choice for businesses? Goal-setting activi-
ties and hiring/firing decisions are examples of non-mechanical decisions. Or, are
they? We all like to think of them as something that implies a choice, but the
passage between mechanical and non-mechanical decision making is very weak.ix
Goals may be set up in ways that replicate the past or imitate competitorsx , and
firing may be an automatic expulsion of those with poor performance (see GE
when Welch was CEO).xi Other examples may include scheduling and budget-
ing activities where previous ones may be simply carried over into the next
period (thus treated as mechanical decisions) rather then examined carefully (thus
treated as decisions that imply choices). Of course, you have never heard of any
academic that is willing to support these activities as mechanical; and this is espe-
cially true when we are dealing with human resource management. However, this
is what happens sometimes in organizations. Now, if we want to deal with it, I
believe that the continuity rather than the opposition of concepts better serves the
scope of defining how decisions are taken and can change. Therefore, we can
assume that it is the power of individuals, groups, and organizations that take
into consideration alternatives when making decisions. Again, if they are consid-
ering those alternatives, we have a decision that implies a choice; where these
alternatives fall outside of the persons will, then we are facing a mechanical
decision.
12 2 Rationalization and Rationality

Creative Decisions
A particular kind of decision is that related to the creation of something. Creativity
may be defined in many waysxii , and I dont think the present discourse could benefit
from entering this vast domain of knowledge. For the purpose of this text, creativ-
ity is an activity (a decision) that brings something new, something that was not
experienced before.
This third type of decision is close to the one mentioned in the previous section
(e.g., choice-based); however, it is somehow different from that. What is a creative
decision? It may be (a) the ability to find a new pattern to make a decision, or (b) the
substantial newness of the decision. The latter relates to the outcome of the decision,
while the former is about the process. Although a creative process may lead to a
creative decision, this is not automatic and we cannot take it for granted. Unusual
or highly creative decision processes can also lead to a decision that is ordinary (I
have written more on this distinction in the following pages).
There are several examples of what creativity can be. This ranges from a simple
rereading of data you already know to find new insights, to a decision based on a
newly generated and unexplored set of options. Creativity in decision making is the
ability to generate alternatives that serve as a basis for your choice. The more you
generate, the wider the basis for your choice. The better you generate, the greater
your chances of success. It is not a matter of quantity, nor is it a matter of quality
only. It is a general attitude that brings together quality and quantity. Creative minds
are not limited to a single outcome of their creativity; quite the contrary. If you
look at highly creative minds you find that these persons are also very prolific. How
many books does a creative writer publish? Take Simenon, Christie, or Hemingway
as examples.
Not everyone is like Simenon, but we can see that the wave of creativity is not
limited to one single shot. It might be, but if we experience that wave once, we tend
to reproduce it for personal and group satisfaction and fulfillment. It enhances the
chances of success.

Epistemological Corner
I would like to make a very quick point on the processes that relate to each one of
the three types of decisions. What is the logic behind mechanical, non-mechanical,
and creative decisions? What are the processes that the mind carries on when we
are involved in one of the three processes? Are these processes the same, or do they
differ depending on the type of decision?
Each one of the types described above relates to a different process. What follows
here are speculations about what it could be, not on what it surely is.
The first type, mechanical, leans on previous decisions. It is a replication of some-
thing that has been generated in the past and that continues to be. I should say that
what happens is very close to induction, i.e., the ability to generalize starting from
single events. In our case, the generalization is the fact that we tend to repeat the
The Legacy of Herbert Simon 13

same decision (e.g., we get dressed) because we always did in the past. Or, we
deduce that this is the right thing to do or think. Put differently, the universal rule
emerges from repetitions of single actions.xiii Individuals extrapolate the norm from
repeated actions so that the point of reference becomes the norm, not the action.
When the norm is followed, the decision is mechanical. In the example, to get
dressed after waking from sleep and before going to work is a norm for the indi-
vidual decision making that has become such because of social habits and repeated
behavior over time.
In the second type, the non-mechanical, we have a set of alternatives among
which a choice is made. We analyze these alternatives and try to deduce what is the
best choice to make. Therefore, and contrary to the first type of decisions, the major
process involved in decision making is deduction. It is the way to infer something
from assumptions using a logical analysis. This logic is close to what most stud-
ies on decision making follow. As already stated in the previous chapter, decision
making has a significant computational and formal root that makes significant use
of deductive logic to frame decisions.
Creativity needs something different from both induction and deduction. The
process of getting something innovative has induction, since it gathers conclusions
from experience (i.e., generalizes through the collection of single events), and it
has deduction, since there is a logic that arrives at conclusions from given assump-
tions. However, we should say that this description of creativity is poor and that we
need something more sophisticated. We need a process that is able to describe how
some general principle could emerge from assumptions that could not be directly
and immediately related to it. This is what is called abduction, and it is a process
often used by philosophers of sciencexiv to describe creativity. Since deduction and
induction are part of the general scientific vocabulary while abduction is not, I think
it needs further explanation.
The term abduction was coined by Charles S. Peirce to describe inference that
involves generating and evaluating explanatory hypotheses.xv For example, take
the manager that explains the low response rate to the survey on the quality of the
working environment due to its online delivery. This is not a deduction, rather an
intuition, for many other variables could have affected the low response rate (e.g.,
insufficient efforts toward explaining its value, workers beliefs that the surveys
findings will have no impact, workers are dealing with a period of increased pro-
ductivity and extended working hours, and more). The cause (hypothesis) that the
manager finds appropriate (evaluates) is generated through a creative effort.
As shown in later chapters, the importance of abduction is not limited to creative
decisions only, but it has the potential to provide significant insights on the other
types of decisions. For now, this introduction is sufficient.

The Legacy of Herbert Simon

If we look at the past century in search for somebody who well represents the
decision-making field, that person is Herbert A. Simon. I have always been fas-
cinated by the studies and the legacy of this man, that I consider a sort of modern
14 2 Rationalization and Rationality

Leonardo da Vinci. It is very difficult to define his studies as confined to psychology,

decision making, rationality, artificial intelligence, cognition, social psychology,
sociology, organizational behavior, or economics. He made profound and insight-
ful contributions to all of these fields. I stop here with the acknowledgment of his
work because the very best way to get an idea of what I mean is to read one of his
writings.
Simons starting point has been that of criticizing the neoclassical model of
rationalityxvi and of decision making. This model is used in economic models to
describe and predict human behavior on the assumptions that, when making deci-
sions, individuals have full access to information and their cognition works as a
perfect computational device. Arguments on whether the neoclassical model is use-
ful or not and to what extent we should switch to a bounded rationality model started
more than half century ago and, believe it or not, the debate is still alive. However,
we dont need to summarize it here; for the economy of our discourse, we will recall
concepts and ideas as needed.
One of Simons major contributions has been on the understanding and analysis
of rationality. We start from a classic point, that of rationalization.xvii

Rationalization
Suppose you are asked the following question: How do you go to work? The answer
may vary between car, bike, foot, tram, bus, metro, train, airplane, helicopter (maybe
the last two apply only if you are a Stanford or Harvard student). Whatever means
of transportation you use, most of the time you dont make an active choice, but
you just use the means that you always use. If you have a monthly ticket for the
metro system in your city, you dont make the decision to take the train since this is
a typical mechanical decision-making process.
Now, what if I ask you why do you take the metro-train? Here you can recall to
your memory the reason why you do so. Here too, the hypothetical answers may
vary between, for example, (a) it is the fastest way to get there, (b) I havent a car,
and this is the only way for me to get to the university or to work, (c) it is the
cheapest means of transportation, (d) it is where I always meet my sweetheart, (e)
and the like. What are you doing when providing such answers? You are explaining
the reason(s) why you take the metro-train, or you are offering sound reasons that
support your choice.
This attitude is widely diffused, and we use it every time it is needed, no matter
if somebody else asks about our behavior or if it is ourselves asking for reasons. We
use motives to put a rational emphasis on our behavior.xviii We explain our behavior
(and choices, and way of thinking) through what is called rationalization.
Rationalizing ones behavior doesnt mean that you put efforts to always find
reasons that make sense. It means that your ability to think about your thoughts
becomes real and useful. Rationalization is a mental process that is driven by self-
awareness,xix something that lets you think about what you are doing. For example,
The Legacy of Herbert Simon 15

suppose that you provide a different answer to the question above on why you use
a specific means of transportation. You go to the university by car but prefer walk-
ing, and it isnt that you are that far from the university, it is that you are lazy.
Moreover, you dont like being lazy at all! Now, what is the rationale for taking
the car? Apparently, according to this answer, there is no rationality in taking the
car, but still the process of analyzing, explaining, and providing reasons for your
behavior falls under the rationalization phenomenon. Once again, rationalization is
the action of making sense of what you do or think, and it happens through your
own interpretation categories.
The last part of the sentence is particularly important. The fact that you or other
people ask you to rationalize is important but not relevant in the argument I want
to make. It is you that need to make sense of what you do. Of course, you can do
this through social constructs or through your personal beliefs (when does the first
end and the second start is a question that we address below) but the most important
point is that you do take time to make sense of your behavior.
When individuals fail to rationalize or find out that their rationalization is inac-
curate, then a feeling of discomfort and/or distress emerges.xx There are several
experiments and studies that point out how this mechanism works. Imagine that
you have to rank 10 different music tracks.xxi Soon after that, you are asked to
explain why you ranked tracks #5 and #6 in their respective order. To explain why
#5 is better than #6, you may over rate it. Otherwise stated, you are trying to make
sense of your choices through a rationalization process. What happens if the two
alternatives are equally likable to you? This is the case when you listened to that
music for the first time. It may happen that you do not have a preference and do not
significantly prefer #5 over #6. Your difficulty (distress and physiological arousal)
in finding a sound reason for your choice is called cognitive dissonance.xxii Put
another way, regardless of the significance of the decisions, people faced with
equally attractive alternatives tend to experience cognitive dissonance and justify
their decisions.xxiii This is exactly the case of the example for most people: You
have to make up an explanation to convince (yourself) and the person asking the
question that your choice makes sense. Although very interesting and important for
both our psychology and cognition, there is no point in continuing to analyze dis-
sonance here. The only purpose of introducing cognitive dissonance is to highlight
the fact that rationalization is a matter of utmost importance. In our everyday rea-
soning, when we (a) fail to rationalize, (b) realize that our rationalization is not
consistent with the choices we made or that we will make, or (c) rationalization
is poor for us or for other people important to us, then we experience distress and
discomfort (i.e., cognitive dissonance). The tendency to see oneself as a rational
individual that makes rational choices is an important part of the way people think
of themselves.xxiv Another way to think of rationalization may be that of defining it
as the tendency to avoid dissonance by activating sense-making processes.
It is now apparent that rationalization is one of the major activities that comes
together with decision making. We cannot analyze our decision-making attitudes
or our choices if we cannot understand the underlying process that allows us to
do so.
16 2 Rationalization and Rationality

Rationality
Now that we have dealt with rationalization, we can try to understand what ratio-
nality is. Rationality is brought into the decision-making discourse by the process
of rationalization, and it has always been a study of how good the decision (or the
underlying process) has been or should be. While rationalization defines thinking
activity in general, rationality defines its contents (both goals and procedures).
To use Simons words, rationality is concerned with the selection of preferred
behavior alternatives in terms of some system of values whereby the consequences
of behavior can be evaluated.xxv This is a very classical definition of rationality,
and many scholars dont use it anymore. However, it is a good starting point from
where we can build, modify, and add.
For a better understanding of this concept of rationality, we can divide the
definition into four parts. Rationality means

1. the selection of alternatives

2. through a system of values (i.e., weights or choice drivers)
3. that allows individual to make decisions
4. and to make evaluations on potential and actual consequences of behavior (or
actions).

According to Simon, rationality is what allows individuals to make decisions out

of a pool of alternatives that are consistent with ones values. We can also define a
rational decision when the process of rationalization reveals a system that is struc-
tured according to the definition. For example, the answer of the lazy personi.e.,
the one who prefers to use the car even when recognizing that this is not the best
choice for him/heris not rational. The choice in that case is not consistent with the
evaluation of alternatives. There, the walking alternative has more expressed value
than the driving alternative; however the former has not been chosen. Again, the
selection of alternatives does not follow the system of values and suffices to define
the decision as irrational.
Or, we can find a different explanation for the example and make it rational.
We can, for example, imagine that what that person is telling us is only a par-
tial truth. In reality, there is a trade-off between the feeling of not being lazy and
the comfort of getting to the university by car. In this case, the weighting function
(i.e., the values) is different and, since the comfort is superior to the laziness feel-
ing, the choice is consistent with the system of values. It becomes apparent from
this simple example that the most important part of the definition is point four, the
evaluations on potential and actual consequences of behavior. Put differently, the
analysis of consequences that derive from a specific action. This last idea brings into
the decision-making computation the future or expected impact of our behavior.
There are many limitations in this definition of rationality. It is what we should
call a definition of the process through which we actually make decisions. However,
what kind of decisions? Are all three types of decisionsmechanical, choice-based,
Notes 17

and creativewell-represented by this idea of rationality? When we select alterna-

tives through a system of values that allows us to try to forecast future consequences,
we are making a choice. Apparently this definition of rationality supports the second
type of decisions; those decisions are rational if they follow the process as described
in the definition. Mechanical decisions never follow explicitly a process similar to
that of the definition. They can be analyzed (rationalized) ex post, and a rational
explanation of behavior can be added to them. Creative decisions are the most diffi-
cult to define in relation to this basic concept of rationality. The whole point here is
that our decisions are rational if we follow this well-defined pattern, and that there
are certain types of decisions that are more likely to be so. Those that imply a con-
scious mental activity directed toward a decision-making effort have more chances
to fall under the rationality cap. But we are struggling between two different ideas
of rationality: One looks at the process, the other at the goal. Is a decision rational
when we find its rationality after it has been made? The next chapter addresses this
and other points.

Summary

In this chapter, we have learned of three types of decisions: (1) mechanical, (2)
choice-based, and (3) creative. Alternatives play a crucial role in these three as they
are overlooked in the first-type decisions, examined in the second, and newly gener-
ated in the third. We have also associated mental processes to each of these decisions
(respectively, inductive, deductive, and abductive reasoning). After that we started
with the analysis of how people make sense of their decisions (i.e., behaviors and
thoughts) and explained that this is what is called rationalization. A decision could
be rational or not, but individuals can always rationalize it. The chapter ends defin-
ing rationality as the selection of alternatives, filtered by each ones values, that
allows the individual to make the decision on the basis of the evaluation of potential
outcomes associated with that decision.

Notes
i. The work of H.A. Simon (1947, 1955) affected decision sciences greatly, and this may be
considered an outcome of his work. However, I suspect that this link between rationality
and decision making is more than old. Philosophers track down to Plato the first connection
between decision making and rationality, and to Descartes for what concerns a modern
approach to rationality.
ii. This is the point that many behavioral economists make crystal clear; for example, see
Ariely (2008).
iii. This behavior is studied by marketing scholars and it is not something where I cannot be
said to have any sort of expertise. However, an interesting reading can be Dickson and
Sawyer (1990), in relation to the (secondary) role of price for grocery store shoppers.
iv. The literature on decision making prefers the word intuition to define these decisions. A
distinction between intuitive and controlled mode is offered by Daniel Kahneman (2003).
18 2 Rationalization and Rationality

v. This is what Verplanken et al. (2005) study in their paper on the measurement of habits.
According to these authors, given the prevalence of repeated over new behavior, there
is good reason to pay more systematic attention to constructs like past behavior, repeti-
tive choices, experience, routines, and habit (p. 231). I agree with them and believe that
mechanical decision making is a significant part of human behavior and that more attention
should be directed to it by scientists of all disciplines. Unfortunately, it has been largely
overlooked and has had only occasional scientific investigation. It is not in the economy
of this book to explore habits or mechanical decision making in particular although I will
refer often to this type of decisions.
vi. Whether it is called tacit knowledge (Polanyi, 1966) or routines (Nelson and Winter,
1973). Although the notion of a tacit knowledge has potential to lead to creativity when
made explicit, here we consider its connection to routines, as analyzed by Nelson and
Winter (1973) (see also Foss, 2003a).
vii. At least, this is what managers think is an effective way to conduct that business. As I show
later in the book, excessive limitations on human discretional decision making is often not
counterproductive for organizations.
viii. There are several types of routines. In particular, some of them are policies or actions
that prevent the organization from experiencing pain or threat and simultaneously prevent
learning how to correct the causes of the threat in the first place (Argyris, 1986, p. 541).
These are called defensive routines; organizations are not immune from creating these sorts
of pain-killers but tend to avoid them when they become apparent obstacles to managerial
goals.
ix. Betsch and Haberstroh (2005), analyze routines in decision making.
x. This is exactly what studies on bandwagons in innovation diffusion show (Abrahamson
and Rosenkopf, 1993).
xi. David Olive (2001) explains how Welch represents the bright and the dark side of an era.
With his job-cutting strategy, he forced workers to better perform through the fear of losing
their job more than commitment to the same cause, goals, values, or any other thing that
motivational theories should suggest. It seems that Welch himself recognizes nowwhen
it is too latethat it was a dumb idea for executives to focus so heavily on quarterly
profits and share price gains (Guerrera, 2009).
xii. See Sternberg (Ed.), 1999.
xiii. I am borrowing the definition of induction from Poppers Logic of Scientific Discovery
(1935/2002).
xiv. Introduced by Peirce (1955), abduction has been put at the center of philosophic and
scientific inquiries (see Magnani, 2001).
xv. Magnani (2007, p. 224).
xvi. This is what can be found in Von Neumann and Morgenstern (1944), Friedman (1953) and
summarized by Wallister (2008, pp. 5254).
xvii. Rationalization is the process that leads people to explain their own actions in terms of
their alternatives and the consequences of those alternatives for their preferences. Similarly,
they explain the actions of others by imagining a set of expectations and preferences that
would make the action rational (March, 1994, p. 3).
xviii. This is what can be found in Festinger (1957); see also Kunda (1999, p. 216f).
xix. Self-affirmation is also a by-product of this mental process; Kunda (1999, p. 220f).
xx. Zanna and Cooper (1974).
xxi. This is a variation of Heine and Lehmans experiment (1997).
xxii. Festinger (1957) has been the first to point out the importance and implications of cognitive
dissonance. After his seminal work, a significant amount of studies has been conducted (see
Kunda, 1999, Chapters 6 and 11).
xxiii. Hoshino-Browne et al. (2005, p. 294).
xxiv. As Heine and Lehman (1997) show there is a cultural difference in the way people
experience cultural dissonance and use rationalization processes as a reduction mechanism.
xxv. Simon (1997, p. 84).
Chapter 3
Bounded Rationality

The study of rationality has always dealt with mathematics. The analysis of and
the solution to a problem have always been connected to the logical ability of the
individual so that tests like IQs and GMATs or GREs are supposed to predict how
successful a person will be in the real world.i
Unfortunately, all of these measures fall short of such predictionsii and raise inter-
esting questions on the limits of individual rationality. To make a long story short,
one of the most interesting debates in decision making is the one that relates to
why and how rationality is limited. I postpone discussing another basic question, if
rationality is limited, until the last chapters of this book.iii
An interesting idea on rationality has been introduced by Herbert Simon in 1947
and then again in 1955iv . It is the idea that rationality is bounded. This chapter is
dedicated to the analysis of bounded rationality (BR), according to its father.

What Is Bounded Rationality

The most important strength of the BR modelv is that it contributed immensely
to understanding and overcoming the limits of the theory of a fully rational or
unboundedly rational individual. In short, this theory assumes that individuals have
perfect computational capabilities and make decisions on the basis of informa-
tion that is complete, available, and unambiguous. This idea was (and still is)
widespread in economics, finance, and in certain management domainsvi (e.g., strat-
egy). Problems with this theory relate to the fact that theoretical models lacked
empirical validationvii , especially when tested at the individual level. This is prob-
ably due to the nature of the assumptions and to the historical period in which they
emerged. Models of the fully rational individual were theoretical first, and only years
later have they been empirically tested.viii
Students of bounded rationality, on the contrary, maintain that behavior and
observation of behavior come first. They tend to define theory on the basis of
observed evidence instead of offering the theory firstix . For this reason, the approach
and the outcome (i.e., bounded rationality) come as descriptions of how people
actually behave.

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 19

DOI 10.1007/978-1-4419-7542-3_3,  C Springer Science+Business Media, LLC 2011
20 3 Bounded Rationality

This distinction may seem close to the older one, based on descriptive and norma-
tive theories. The full-rationality theories explain how human behavior should be,
while the theories of bounded rationality describe the world as it isx . The question is
what behavior happens to be prescribed in a world of fully rational individuals. And,
has this prescription any relation at all to a world of boundedly rational agents? We
see in the following that the world of boundedly rational agents has its own rules
that could prescribe as well as describe individual behavior.
Nevertheless, these points are hardly definitive. There is an interesting debate
on human rationality that still goes on, and what you read here is my take on it.
I believe that there is enough evidence to support the statement that a completely
rational individual has no room in our world, not a normative or a descriptive role.
Any theory of rationality and decision making needs to start from something close
to actual behavior and thinking. The theory of bounded rationality describes with
reasonable accuracy how individuals make decisions in the real world rather than
in some idealized condition. And this is the reason why it still offers an excellent
starting point to any study on decision making. Therefore, there is no treatment of
neoclassical theories in this book. The following pages are dedicated to a description
of some of BRs most significant contributions.

Substantive and Procedural Rationality

Bounded rationality is concerned with the process, first and foremost, and with
the organization of resources. Assuming that the goal of the firm is profit
maximizationit cannot be, according to BR, but this is an easy way to startthere
are multiple ways to get to the same point. We dont have an optimal configuration
of resources here; instead we have different patterns that lead to a cluster of simi-
lar goals. This means, basically, that the same profit goal can be achieved through
operational cost cuts, layoffs, supply chain optimization, improved quality control,
outsourcing, tax shields, increased sales, and more. The world of bounded rational-
ity is more complex than that of full rationality; it approximates the real world.
Of course, when the process and the outcome are separated, we may have mul-
tiple combinations then. Table 3.1 presents these possibilities. Only cases one and
three are consistent with the full-rationality theory: The use of resources leads to a
result that is perfectly in line with it.

Table 3.1 A combination of Process Outcome

processes and outcomes 1 Rational Rational

2 Rational Irrational

3 Irrational Irrational

4 Irrational Rational
What Is Bounded Rationality 21

Broadly speaking, all results may be considered consistent with a broader idea
of rationality. The most rational use of resources can relate to an irrational out-
come. I realize that this is an understatement, but the process through which the
Inquisition executed thousands of women (witch-hunt) over the past centuries fol-
lows a rational process that shows consistency between alternatives, values, and
evaluation of consequences. However, we cannot see any rationality in the outcome
of killing individuals based on prejudices, no matter if we are God or Goddess
believers or not. On the other hand, irrational processes may lead to rational out-
comes. The process though which a company installs and promotes the use of a
new software among employees may be sloppy and inefficient. The outcome may
be that of improving internal information processing anyway. Another example may
be that of considering emotional choices or decisions we make when short of time.
Even though we skip some of the logical passages we should follow, decisions are
rational.
In discussing decision-making processes, Simon placed great emphasis on the
distinction between substantive and procedural rationalityxi . He described that dif-
ference stating, We must give an account not only of substantive rationalitythe
extent to which appropriate courses of action are chosenbut also procedural
rationalitythe effectiveness, in light of human cognitive powers and limitations,
of the procedures used to choose actions.xii
According to substantive rationality, the rational character of decision making is
concerned with the result one could get following the appropriate actions, whereas
procedural rationality points out the process by which people make decisions.
According to Simon, bounded rationality belongs to the latter category because
it does not look only at the result one could get, but at the way people make
decisions.xiii The definition of rationality presented and discussed in the first chapter
is exactly of this kind.
The difference between substantial and procedural rationality is fundamental
since it gives to the domain the same width that efficiency and effectiveness give
to management. In a fully rational model these two aspects are mixed up, since
there is no possibility that the same goal (substantial r.) could be obtained from
an alternative use of resources (procedural r.). Otherwise stated, in a perfect world
the process leads to the only possible optimal outcome. If the objective of a firm
is that of profit maximization, this means that resources are allocated according to
this goal. There are no different configurations of resources since the only possible
outcome is the optimal level of profits for that firm. Therefore it doesnt matter what
is inside the firm because the rational exploitation of resources leads to the optimal
result. Here is where the idea of the firm as a black box comes from.
Table 3.1 presents only two levels of analysis(1) process, (2) outcomeand I
have voluntarily left unclear the relation between these two, and (3) goals. As far as
the neoclassical model is concerned, there is no difference between outcomes and
goals. The former is what results from the process, while the latter is something
one wants to achieve, or the aim at which action is oriented. When there is one best
solution, the way resources are organized lead to the outcome that is the goal. In the
case where process and outcome vary, as in the case of BR, the outcome may be far
22 3 Bounded Rationality

away from the goal. And this gap is unavoidable, due to how human beings process
information.
What I am trying to define relates to the nature of bounds since rationality is
bounded when it falls short of omniscience. And the failures of omniscience are
largely failures of knowing all the alternatives, uncertainty about relevant exogenous
events, and inability to calculate consequences.xiv In short, we can argue that there
are external (informational) and internal (computational) limitations.

External and Internal Limitations

Our knowledge of the external world is not perfect. We dont know all possible
alternatives to a given problem. This is the reason why, for example, in the problem
with corporate profits as mentioned above, there are multiple ways to get similar
results.
Uncertainty is the word that describes how the external world affects our atti-
tude to make sound decisions. It is very unlikely that managers and top executives
spend their time collecting information on all possible variables that could poten-
tially affect decisions and their consequences. It is more likely that these individuals
get what they consider the most relevant information and then make a decision in a
timely fashion. There, a second and internal limitation occurs: computation.
This second is the limit to compute large amounts of data. That is to say that
even if all possible data and variables related to a given problem were known, we
shouldnt be able to use them. This is because of the cognitive internal bounds that
characterize our rationality. Take the case of competition. Suppose that you are
looking for a convenient home insurance. You start surfing the web in search for
information, then you start visiting companies. Imagine that the market is not per-
fect (this is easy even if your imagination is not that vivid!), so disequilibria occur,
there are frictions and transaction costs. What is the difference if you get information
from 50 or 100 companies?xv Unless you do the market price tester as a full-time
job, you dont have the skills or the time to handle all of this information. You would
probably give up much sooner than needed. I believe that even price collectors and
analysts need particularly sophisticated tools to compare market prices. Tools and
skills that are not available to everybody. On average, for the individual it doesnt
make any difference if the information out there is complete and available because
he/she cannot handle it.

Satisficing
Internal and external limits define bounded rationality in terms of the procedure
that is the most appropriate to make a decision, should a problem be given. This
procedure is shaped by the limits to get the knowledge that individuals show in
terms of cognition and access to information. The result is that people dont get the
What Is Bounded Rationality 23

optimal result but only suboptimal ones. Therefore they do not maximize but obtain
results that are only satisficing.
There are important differences here from the traditional economic model of
rationality and decision making. First, the result is not one but multiple outcomes to
any given problem. Going back to the example of profits, this is quite understandable
when we deal with a single goal since we can get to that point through many differ-
ent pathways: There are so many ways to get to a given amount of profit. However,
what a company gets is never the best result possible, the maximum amount of profit
possible. There are multiple levels of analysis here, though. Lets try to take a clearer
approach to the problem. In relation to profits, our managers may have to deal with
(a) defining the level of profit that is sustainable for the company (i.e., that is satis-
ficing); (b) understanding the possible/viable (satisficing) alternatives between cost
cutting, tax shielding, or sales incrementing strategies; and (c) making a choice that
satisfies the companys needs at best. The latter point is close to the maximum only
by chance, since not all of the variables are known. As you can see, satisficing can
be used in many ways when we deal with a problem.
Some scholarsxvi suggest that satisficing can also be a search strategy. In this par-
ticular meaning, decision makers adjust their behavior according to whether the goal
is getting closer or not. If the companys managers set the goal to sell N units of the
new product by the end of the sixth month of its introduction, they can increase
their efforts toward the goal (e.g., ads, marketing, put pressure on retailers, TV
commercials) if after the third month they sold only N/4 units. Otherwise, they can
decrease their efforts if the units sold by the third month are more than N/2 units.
This mechanism defines satisficing as a decision and search strategy.
Broadly speaking, satisficing defines results and it could, on certain occasions,
specify the process by which individuals achieve these results.

Bounds or Limits?
Although Simon writes about bounds of rationality, the worldwide translation of
this concept may indicate a different aspect of the problem. Translations in many
languages use the term limited to describe the same idea as bounds. Here are a
few examples: The French expression is rationalit limite; the Italian is razionalit
limitata; the German is begrenzten Rationalitt; in Spanish, it is racionalidad limi-
tada. Why do translations put such an emphasis on limits instead of bounds? Well, I
dont have the answer to this question, but I think that we should focus a little bit on
the differences between the use of different words to describe the same idea/theory
and try to understand which one of the two does a better job describing the theory.
Some scholarsxvii studied the genesis of BR and highlighted the fact that the pro-
cess through which Simon came to this terminology had been a refinement process.
In fact, he passed through approximate rationality and limited rationality before
settling for what became the label for his idea.
24 3 Bounded Rationality

James March, once very close to Simon and part of the so-called Carnegie
schoolxviii , in his book A Primer on Decision Makingxix , uses the terms limited
rationality to define the theory presented in this chapter. However, I dont know if
this is just a recognition of what is widely accepted outside the U.S. or if this is a
readaptation to what the theory really means. As the inventor of this terminology
puts it,

You have to realize about the bounded rationality terminology that I began to use this as a
label for the things that economists needed to pay attention toand were not. It was never
intended as a theory in any sense.xx

This came as a reading of what happened 40 years after the first time the words
were used. It really seems that the choice of words has been made to state something
different from what one should think, and it is not clear whether the two words
bounds versus limits reflect any change in Simons original idea.
The point is that the theory of bounded rationality has always been intended as a
theory of limits and limited rationality. While bounds give the impression that some-
thing is tied up, restricted to a specific territory, limits are something stricter. Bounds
can be modified while limits cannot. Think of the fact that we call boundaries those
of a state, not limits; and we can eventually pass these boundaries. Limits give the
impression of being stable; we have limits of a mathematical function, for example,
and we know that these are fixed. Is this theory of rationality closer to limits or to
bounds? This is what we will see in the coming pages.

Summary
In this chapter we have defined bounded rationality in a classic fashion. The distinc-
tion between substantive and procedural rationality is useful to make the difference
between goals (outcomes) and resources employed in the process, respectively.
Individuals are boundedly rational in that (a) they have computational and analyti-
cal limitations (internal limits) and (b) they also cannot get all possible information
from the outside environment (external limits). The combination of these limitations
leads us to satisficing rather than maximizing decisions. Moreover, satisficing can
be a useful search mechanism that proceeds through feedbacks that adjust the search
toward the achievement of a given goal.

Notes
i. In The Intelligence Controversy, Eysenck and Kamin (1981) exchange opinions on the
nature versus nurture debate on intelligence. They discuss also intelligence tests and go
over their history and limits. See also Richardson (2000).
ii. A very interesting way to look at this is Gladwells Outliers (2008).
iii. In his writing Why bounded rationality? Conlisk (1996) offers significant evidence of
what are bounds of rationality.
Notes 25

iv. According to Klaes and Sent (2005) the first time the terms bounded rationality appeared
were in Simons Models of Man, in 1957, page 198. Herbert Simon, very likely the first
person to have used bounded rationality, tentatively applied a number of expressions that
seemed available on the basis of associations with what he sought to express. In a series
of writings between 1947 and 1957, he consciously refined and replaced concepts such as
approximate rationality and limited rationality until settling for bounded rationality
(Klaes and Sent, 2005, p. 37). I refer to these attempts, since the meaning Simon intended
to give to these expressions is the same; what changed was the label, not the content.
v. I am not sure whether to call this model, theory or, more generally, approach to bounded
rationality since there is no one theory nor there is one single model. However, the major
assumptions of all theorization are clear and remain very similar for every scholar (Foss,
2003b). There are many models of bounded rationality (e.g., Rubinstein, 1998) but there is
no single theory. Maybe, we can define it as a general approach to rationality that shares
the belief that human beings have limits.
vi. I recently read a very vivid debate on strategic management and objectives of the firm.
Whether authors never mention or explicit any link with bounded rationality, I believe
that Sundaram and Inkpen (2004) clearly refer to full rationality when they advocate for
maximization of profits as a single-goal function for the firm while Freeman et al. (2004)
implicitly refer to the limits of rationality when considering the stakeholder approach and
multiple goals for the firm.
vii. I will explain this later in this text; see Conlisk (1996).
viii. The debate around Tversky and Kahneman (1974), reveals very little evidence of full
rationality.
ix. There are two streams of research in bounded rationality (BR). The old school, that of
Simon (1955), started from the neoclassical theory of the fully rational homo economicus
to build the theory of BR. The new school, that of neo-Simonians (e.g., Kahneman and
Tversky, 1979; Gigerenzer and Selten, 2001), started from empirical tests on neoclassical
theory to show limits in human rationality.
x. In the words of Herbert Simon, theories of full rationality explain how people ought to
behave, not how they do behave (Simon, 1959, p. 254).
xi. See also Munier, Selten, et al. (1999, p. 234).
xii. Simon (1978, p. 9); italics in the original text.
xii. Simon (1978).
xiv. Simon (1979, p. 502).
xv. I use this argument in a paper (Secchi, 2009; unpublished paper) that criticizes the theory of
perfect competition as Block et al. (2008) apply it to the banking and insurance industries.
xvi. See for example March, 1994, Chapter 1.
xvii. Klaes and Sent (2005, p. 37). See also Augier (2000).
xviii. Augier (2004).
xix. March (1994).
xx. Simon (1999, p. 23); quotation retrieved in Klaes and Sent (2005).
Chapter 4
Maps of Bounded Rationality (I)

In the late 1970s, scholars started to study if and how bounded rationality works in
practice. Experiments were set up and there started the ever growing literature on
the subject. The first evidences were related to falsifying neoclassical approaches to
rationality, e.g., the theory of expected utility (see below), while the latter efforts
cover a wider terrain, focusing on heuristics, ethics, cooperation, and altruism
(discussed in a later chapter).
This chapter offers an overview of some of these studies. The objective is to
show that the work on bounded rationality is well-alive among behavioral scholars,
and that it has recently gained momentum. A general study of the effects of these
theories on organizations has yet to come. I do not try to provide any organizational
transposition of these theories in this chapter, but attempt to offer hints on how
they can be utilized and what meanings we can get.

Prospect Theory
With a series of experiments, Daniel Kahneman and Amos Tverskyi introduced a
test of rationality directed at falsifying the assumptions of expected utility theory.
This is the iconic neoclassical theory where individuals are supposed to make deci-
sions on the basis of statistical calculations.ii Before getting into a foundational and
theoretical analysis of their work, I believe that examples will serve the reader better.
As Kahneman himself wrote, [o]ur research attempted to obtain a map of bounded
rationality, by exploring the systematic biases that separate the beliefs that people
have and the choices they make from the optimal beliefs and choices assumed in
rational-agent models.iii
Truth to be told, most of the constituents of this map of bounded rational-
ity were introduced in some of their original pieces in 1974 and 1979. The topics
reproduced below have been selected from the 1979 article.iv

The Certainty Effect

Among behavioral studies, one of the most important developments of the last few
decades has been prospect theory. This is the analysis of decisions under uncertainty

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 27

DOI 10.1007/978-1-4419-7542-3_4,  C Springer Science+Business Media, LLC 2011
28 4 Maps of Bounded Rationality (I)

where people choose between contracts, where each alternative could materialize
depending on its probability. For example, suppose you have an 80% chance to get
$4,000 or that you can have $3,000 for sure (i.e., with probability 100%). Or, more
formally stated,
A: ($4,000, .80) or B: ($3,000, 1.00)

What do you choose? Well, I can tell you that most of the people choose to get
$3,000 right away instead of betting on the higher sum of money. Now, consider the
following two alternatives:
C: ($4,000, .20) or D: ($3,000, .25)

Is your choice C or D? If you were a strictly rational (logical-mathematical) person

you should have chosen D, the only choice that goes well along with your previous
choice, i.e., B. If you were not using this mindset, mostly based on expected utility,
you have probably contemplated C among your choices and gone for it!
This is what the two scientists found in many of their experiments. People tend to
prefer what is certain compared to what is not (this is why choice B is preferred to A)
despite the fact that the expected value of the two choices differs. Choice A is worth
$3,200 while choice B is only $3,000. The more risk-seeking individuals would have
chosen A while more conservatives would have chosen B. The majority of people
that took part in the experiments preferred B to A. However, this is not the most
interesting part of the experiment. When presented with the second set of choices,
where quantities stay the same and the relation between the two probabilities also
remains the same, people are not consistent with that first choice. It is worth noting
that 80/100 is the same as 20/25. On a strictly logical basis, if you prefer 100 to 80,
you should also prefer 25 to 20. Or, dont you? Well, people usually dont use this
strictly logical way of reasoning, they prefer a gain that is for sure to something that
is uncertain. This is what is called the certainty effect.
Another interesting way to look at this is to focus on the value of free goods.v
Imagine you are attracted by two bowls full of chocolate. In one bowl you have
Hersheys Kisses while in the other you have Lindt Truffles. Once you get close to
the bowls, you read that the price of Hersheys is 1 while Lindt costs 15. You
make your choice and eat the chocolate.
Now that you have just finished with your chocolate, you go back to the same
place and see that the two bowls are still there. You decide that you want another
chocolate but, as soon as you get closer, you realize that something has changed. In
fact, the price of Hersheys is now 0 while that of Lindt is 14. I dont know what
you are thinking, but I know what I should do: Take the free chocolate!
This experiment has been conducted among students/customers, and results were
aimed at defining the value of a free good. It resulted that customers preferred
Lindt (36%) over Hersheys (14%) in the first round, i.e., when Lindts price was
15. In the second round (0 and 14), the preference was reversed and customers
chose Hersheys (42% vs. 19%). When researchers tried to lower the price of the
higher-quality product to 10 maintaining the other chocolate on 0, the result didnt
change: Customers continued to choose Hersheys (40%; Lindt 12%). The benefit of
Prospect Theory 29

choosing the free product overcomes the value of paying any money for a superior
taste.
Together with the authors of this experiment, I believe that the mechanism here
is very similar to the certainty effect defined above.vi People exhibit different behav-
iors when exposed to something that is given for sure or that is available without any
cost. A free good causes people to make decisions on that basis. The availability of
a product makes it certain in the sense that it is given for sure, you can take it with
no deprivation of any sort. This is what the experiment shows very clearly.

The Possibility Violation

Another significant violation of the assumptions of a mathematically based mind
(expected utility) is what can be expressed by the domain of the possible confronted
with the domain of the probable. The following prospect is particularly helpful.
Choose between
A: ($6,000, .45) or B: ($3,000, .90)

Here, most of us choose prospect B where the probability to get the money is the
highest. Note that the expected utility of the two prospects is identical since 45% of
6,000 is 2,700 and this happens to be the same number we obtain from the second
prospect. Of course, the fact that the probability is greater in prospect B makes the
difference.
Now, consider the following problems:
C: ($6,000, .001) or D: ($3,000, .002)

You may face a dilemma here since the probability to gain something is very close
to zero for both prospects. However, the probability to get $3,000 is double that
of getting $6,000; still you dont think it really makes a difference whether you
choose the first or the second. Your decision making is impaired: What is to be
done? These two prospects maintain the same characteristics of the previous ones
in that the expected utilities are identical for C and D respectively, and the relation
between the probabilities of A, B and C, D is 1/2 in both cases. However, when the
probability becomes a mere possibility, people go for the highest amount (i.e., they
take prospect C) because they think that probability doesnt make any difference
and if they have to take a risk. . . well, it is better to risk for the highest amount!
This logic finds application in a lottery, for example. There, the probability of
winning something is very low, so low that an economist once called it the fools
gamble.vii People usually choose to buy tickets of lotteries where the jackpot is
the highest compared to the price of the ticket. This choice is consistent with the
possibility violation. It is called a violation since it violates what an individual with
perfect rational preferences should have chosen, according to expected utility theory.
Of course it is not a violation of common sense and of what is a rule for everyday
survival.
30 4 Maps of Bounded Rationality (I)

The Reflection Effect

What happens when we lose instead of gain something? Do we react the same way
when we are about to lose or when we are about to gain a certain amount of money?
Imagine that your personal bankerviii just called saying that your portfolio with
a market value of $120,000 needs to be renegotiated. Here is her proposal. You
leave the money invested in the same stocks and, according to financial forecasts,
have 80% probability to gain 20% in the near future, i.e., you get $24,000 on your
original capital which makes a total of $144,000.ix This implies that you face 20%
probability of getting nothing more than what you have right now. Or, you can invest
nothing and get your money back with an $18,000 bonus on the $120,000 that you
already have, i.e., $138,000. What is your choice?
According to a simpler choice made on prospects, most of us choose certainty
over the risk of getting nothing out of an investment. It is better to have something
right now than to face the risk of having nothing (no gain, in this example you still
have the principal). This is the basic certainty effect that we explored above. Now,
try to picture the following scenario.
Your personal banker proposes to keep your money invested in the same stocks,
but financial forecasts outline that you have 80% probability to lose 20% of your
investment, i.e., to have a loss of $24,000 that brings your capital to $96,000
total. However, you also have 20% probability to get your capital intact at that
same near future date (which means $120,000). The alternative is to have your
money back right now, but the bank should retain $18,000 from your account to
pay for losses and various fees so that your money is $102,000 in this second
option.
The first alternative is uncertain: You dont know if you are going to lose or not,
and you still have the chance not to lose a single cent. The second alternative is well-
defined so that your capital decreases by $18,000 for sure. The two options bring
you the possibility of ending up with $120,000 still. At the end you decide to risk
and try to get all of your money back. And, this is what most of the people should
have chosen according to prospect theory.
What is interesting here is that it seems that the certainty effect works on the
positive domain while it doesnt in the negative. When people face the chance to
lose money they tend to do whatever it takes to retain what they have earned. A
certain loss is still a loss, while a potentially avoidable loss is not a loss yet! Even
when the probability that the loss will not materialize is low (20% in our example)
people decide to take the risk. The two behaviors are contradictory, and this is the
reason why this is called the reflection effect.
Here is another example. Think of the people playing pool for money. Who is
more likely to say double or nothing? after the first game? The player who lost!
He is advocating a risky strategy in an attempt to eliminate losses.x
From the reflection effect, it follows that people are risk averse when they oper-
ate in the positive domain, while they become risk seekers when they operate with
losses.xi
This Is a Biased World! 31

The Point on Bounded Rationality

This last paragraph brings into our discourse the fact that individuals are not cold
computational machines, but they perceive differently gains and losses and/or they
value differently gains and losses. To this respect, being rationally bounded means
that when confronted with the same sort of variations that, for example, go from
the positive to the negative domain, individuals arent perfect calculators. They are
bounded in the sense that they cannot treat the same way alternatives that are fun-
damentally equal; the limits are internal computational limits in all of these cases.
From this perspective, they are not consistent.
They probably dont have a clear idea of how to use statistics either.xii The pos-
sibility violation and the certainty effect point out specifically this, i.e., the fact that
individuals are not able to explore to the fullest extent the alternatives that they face.
On the contrary, they show limits in computational abilities; luckily, these limita-
tions happen to be particularly useful to individuals everyday life and survival. The
following pages and the next chapter explain why.

This Is a Biased World!

If we were to find a mechanism that defines our rational bounds, this would be
a bias. Psychologistsxiii isolate a list of biases so long that this topic cannot be
completely described and analyzed here. An entire book is needed for that pur-
pose. Nevertheless, biases and prejudices are what characterize many decisions, and
strategies leading to their avoidance are among the most studied.
Although I cannot analyze all biases and prejudices, I have decided to present a
selection of those that represent and support the idea of bounded rationality more
clearly than others. In the following pages you will find some of the most studied
biases in economics, psychology, sociology, and decision making.

The Endowment Effect

A thief took your wallet. You are desperate because you had placed in it (temporar-
ily) your wedding ring. What is the value of what was in the wallet, sir? asks the
policeman. The point is that you dont know. It is not that you dont know how much
that ring cost (you remember that very well indeed); the fact is that you attribute to
that good a value far higher that its market price. That is something you will never
even consider selling. Its value is too high.
This is a very specific and particular case of what is called the endowment
effect,xiv the fact that something gains value only because of ownership. It is this
effect that explains why certain goods have higher prices when possession is exer-
cised on them. This is a pattern of behavior typical of one phase of the game
Monopoly. At a certain stage of the game, close to the beginning, players can
32 4 Maps of Bounded Rationality (I)

negotiate on the cards they have bought in the very first part of the game. Suddenly,
through the negotiation process you learn that the price of those contracts is never
close to its book value (i.e., to what is written on the card). In fact, the selling
price can be very high and far from its original. Of course, we should consider
the economic trade-off here since, once sold, the property can become a treat to
the seller that could land there and pay the sometimes expensive rent. A mix of
contractual abilities, risk aversion, and the endowment effect make that price so
high.xv
This effect can be found within organizations too. For example, the information
that you gained on a specific quality test is particularly relevant once it becomes
of your knowledge. And, in many cases you sell it at a price higher than the one
that should be. Another example could be that of acquisitions. When one company
makes an offer to the other, the first price is never the right one (for many reasons,
of course) but I guess that a not so small part of it is played by the endowment
effect. Moreover, I believe that in negotiation, arbitration,xvi compensation, benefit
administration,xvii and in many more areas of management we have the potential to
find this bias at work.

The Status Quo Bias

When you repeat mental or behavioral patterns that have an anchor to how things
have been set up in the past (and/or still are in the present), you are following a
status quo.xviii A status quo bias becomes apparent when individuals tend to make
the same decisions that they have made in the past.
This is very likely to happen when past decisions have been successful. Business
managers tend to repeat successful decisions, realizing afterwards that the circum-
stances were not the same. The repetition, in Europe, of the same dynamics that were
extremely successful in the U.S. lead Jack Welch to lose the opportunity to complete
one of the most important mergers in history, the one between General Electric and
Honeywell.xix The negotiation tactic that worked with the US Justice Departments
Anti-Trust Division miserably failed with the Directorate-General for Competition
of the European Commission. Without the European market, the merger made no
sense.
Now, if one of the most acclaimed and successful managers can fall under a
status quo bias, what about you and me? What about every one of us? The reality is
that this bias is one of the most powerful and widespread. It is very easy to become
overconfident if we have been successful with a tactic in the past. If you have always
beaten your opponent with the barbers opening, why change it next time you play
chess? This is basically what happens when we lean on past experience. The more
we have been successful with that formula in the past, the more we will lean on it in
future occasions. However, circumstances change all the time, and this is the only
thing we know for sure.
A second application of this bias can be found when people dont want to change
the state of things. It is not that we apply our mental frames to any situation that
This Is a Biased World! 33

has something in common with past situations, but sometimes we do not want to
change. This has very interesting effects when it transfers into organizations.xx
People find themselves stuck in their ways of managing their everyday routines
and elaborate sophisticated arguments to support this status quo. Universities are
no exception. These years are crucial for universities to have students get online
courses. Times are changing at a very fast pace, and the number of students taking
online classes is growing dramatically in the U.S. and everywhere in the world.xxi
The fact that universities need to provide at least a few online classes is becom-
ing a requirement rather than an option. We can discuss, as academicians, overall
value, educational outcomes, and teaching quality of online classes compared to
traditional classes. However, there are more than sound arguments that leave no
doubt that it should be better to act first (i.e., go online) instead of catching up
later. The risk of losing terrain to competing universities is becoming higher and
higher. Having said that, I know that everybody has valid personal opinions (and
supporting data) on online classes, but people who argue against them are usually
falling under the so-called status quo bias. Those who advance arguments such as
the value of traditional education or people come to our university because of
the class experience are particularly falling under this bias. If we, as educators,
had to follow that line of argument in the past, I think that we would be teaching
Socratic maieuticsxxii instead of having modern classes. Or, we would have small
and elitist universities like Pavia, Italy, around year 1000 AD (what is more tradi-
tional than that?). To avoid this bias, the point should not be that of saying yes or
no to online classes, but that of asking how to get the best from this new learning
medium.
The status quo bias works in many different ways; however, an effective way to
summarize what this is about can be the following: We face a status quo bias when
something changes and our ideas do not.

Anchor Bias
Remember the experiment with Lindt and Hersheys chocolates? The price of the
first was 15, while the second was sold at 1. Now, lets make a thought experiment.
If I tell you that you can take the first or the second chocolate and you can name
a price, what is this price? There is a significant probability that you will name a
price that is close to 1 or to 15. If something like this happens then the numbers
that I just recalled serve as anchors, and your rationality has been affected by the
so-called anchoring effectxxiii or anchor bias.
This bias has a huge impact in the way managers make their decisions. Think
about how a meeting is usually set up. Especially important meetings, those where
you are supposed to make decisions, are set up with fancy tables, handouts with tons
of numbers, many words and effective presentations in small professional rooms
set up with food and drinks. Well, not real drinks but coffee and water abound.
These numbers have the specific function to constrain your reasoning and to let you
34 4 Maps of Bounded Rationality (I)

focus on something that will remain close to that. Of course, this is not something
you do consciously. You always have the feeling that you are being very creative
indeed. Unfortunately, you are not. Numbers that come out of those meetings arent
really creative or innovative. The point is not that you can do better without those
numbers. The point is that managers need to be aware that sometimes they are too
close to original ideas, points of view, numbers, data, statistics, etc. available at the
moment of the meeting. They are being subjected to the anchoring effect. You can
stay anchored, but at least let it be your choice!
A second aspect of anchoring happens when people use their own beliefs
and perceptions as a judgmental anchor from which they adjustusually
insufficientlyto accommodate differences between themselves and others.xxiv
Studies in social psychologyxxv provide strong evidence that people anchor inter-
actions with other people to their own assumptions and beliefs. In this respect there
is a fundamental bias that operates when we approach and start a discussion with
other people for the first time. When uncertainty on how that person will react to
our comments is high, we tend to think that there must be something we share and
interpret their thinking accordingly.xxvi Sometimes, when there is evidence that oth-
ers do not react as we expect, we do not change our behavior or way of thinking and
stay anchored to what we pretend to know for sure: how we are. This knowledge
and identity-related anchoring has potential to explain misunderstandings, cultural
unawareness, and some aspects of misbehavior.

Bandwagon Effect
Have you ever jumped on a bandwagon? I am sure you have. It is easy to check
this. If you answer positively to at least one of the following questions, you have.
Have you ever (a) bought something (a pair of shoes, t-shirt, watch, car, book, etc.,)
because you saw it on somebody else, (b) expressed an opinion that was someone
elses, (c) cited or quoted an idea, an opinion, a statement, a book, an article, a
chapter because everybody knows it (or everybody cites it), (d) started listening to
the music that everybody likes, (e) spoke out because somebody else did the same,
(f) looked up because somebody in front of you was doing that, (g) clapped your
hands when you realized that everyone around you was doing so, (h) laughed after
others started doing it, (i) read this list because somebody told you to do so? The
list could continue indefinitely, but these are very simple events and I am pretty sure
that you answered yes to at least one of the questions.
If you mindlessly followed somebodys behavior then you have been part
of a bandwagon.xxvii This is a peculiar phenomenon that attracts sociologists,
economists, and management and marketing scholars.xxviii There are few studies on
bandwagons within organizationsxxix , but there is a general understanding of what
triggers that behavior and that mindset.
Bandwagon is a bias since (1) it could prevent a decision maker from thinking
in situations where this is exactly what is needed, and because (2) it is based on the
This Is a Biased World! 35

prejudice that what other people are doing is correct (worth imitating). Take how a
financial crisis could materialize: A senior financial analyst of a well-known bank
tends not to give importance to certain ratios, he overlooks some trends and focuses
on others. Then junior analysts, still in the first part of their learning curves, tend to
repeat the same mistake. Other analysts that try to make sense of those critical and
overlooked trends are told that those data show only temporary anomalies that will
be reabsorbed by the economy. Reports from this important and well-trusted bank
become public, and analysts from other banks jump on this bandwagon, diminishing
the value of data that nobody analyzes. Who wants to look foolish pointing a finger
at something that is trivial to everybody? The answer is a quick and easy one. When
the crisis is widespread, with a sad ex post feeling, people realize that those numbers
were not so trivial indeed. Instead of jumping on a bandwagon, next time they will
go through every single ratio that makes sense to them (we would hope!). Of course,
short memory and positive feelings tend to delete the feeling of what happened. A
new bandwagon could then arise.

Prejudices
Prejudices are biases of a particular kind. The word means that your judgmental
ability is given before even facing the situation where that judgment is required,
i.e., your judgment operates a priori. Although the meaning of what a prejudice
is could be very close to that of bias, it is usually associated with diversity and
discrimination issues.xxx
For example, during the apartheid regime in South Africa, you could say that
white people were biased toward black people. However, this is not what you usu-
ally hear. What you hear is that people there operated on the basis of prejudices.
This does not relate to racist regimes only, since our societies and organizations
face problems of discriminationxxxi (or, you can use the politically correct word
diversity, if you like) on a daily basis. What is different, if anything, between a
prejudice and a bias? First of all, we are writing about a very specific decision mak-
ing, the judgmental activity.xxxii This means that you are providing your opinion
implying that something is right or wrong, according to your belief. This doesnt
happen every time you make a decision. Sometimes you make a decision without
being judgmental (maybe you are so only at a very deep level). For example, when
you make the decision to wear a yellow shirt, you are not being judgmental. When
you express your opinions on how polluted the environment might be because of
chemical agents that are in the yellow ink used by the clothing industry, this is the
case when you are being judgmental. The outcome is that you probably will not
wear (or buy) the yellow shirt. The difference to make, then, is that a prejudicial
judgment is necessarily a biased judgment, but we cannot state the opposite (i.e.,
that a bias is always a prejudice). The bias could come up at the exact time you are
making the decision while the prejudice is a priori. The former depends on experi-
ence and framing effects, while the latter needs certain circumstances to emerge but
affects your judgment.xxxiii
36 4 Maps of Bounded Rationality (I)

Groups of students at Stanford, Penn-State, and Berkeley have been exposed to a

series of pictures. Some of these pictures were about primates, some of these were
about colored and non-colored human beings. After being exposed to the pictures,
students were asked questions where it was possible to map their cognitive asso-
ciations. Researchersxxxiv found that black people were more often associated with
apes by non-colored students. This surprising result brings to our knowledge that
prejudices could be very deeply embedded in peoples thoughts, something very
well rooted in the hidden and unconscious beliefs. However, they could come out
unexpectedly. Isnt this an a priori judgment? Is this a different type of bias?

Errors

Biases are nothing but errors, mistakes that we make while making decisions or
analyzing specific problems. This is the reason why in the following I consider some
of the most typical errors.xxxv As for biases, these errors are supposed to provide
effective examples of individual bounds.
Imagine that your company has an opening for a junior supply chain managers
position. Your boss asked you to classify and analyze resumes so that you can eval-
uate candidates work experience and skills on the basis of a form that needs to be
filled in each of its six sections for every candidate. You start your job, and when
you are close to the end you suddenly realize that you are rating very high in every
section of the questionnaire as those candidates that have no experience in the field.
This is probably because when you entered the company you had no experience in
that field, so you think that somebody with your characteristics may be a good col-
league and be highly productive, as you are. But, you also remember your college
studies and know that these kinds of errors are very typical and need to be avoided.
In fact, you are making a halo and a clone error (also known as similar-to-me
error). The former is the tendency to base a high evaluation on a single excellent
factor, independent of the other characteristics. The latter is the attitude to prefer
something which has the same characteristics as the decision maker. Romans rec-
ognized very well what this clone factor was all about when they stated simil cum
similis, meaning that people that are similar to each other tend to get together (birds
of a feather flock together). Of course, this is not only limited to the hiring process,
but can extend to any kind of decision-making process.
Now, let us assume that you redo your ranking to avoid those errors. When you
are done with it something strange happens and you realize that you have fallen
under the opposite error. You have penalized candidates that have not had previous
work experience in the field. Moreover, you notice that together with this your rat-
ings are all extremely low, nobody scored excellent, none are even close to well.
Here we are, again!
You have made two kinds of errors. One is the opposite of the halo error and it
is called the horns error. It is the tendency to make poor evaluations on the basis of
one single factor. The other one is the severity or harshness error, and it happens
This Is a Biased World! 37

when the decision maker consistently evaluates something or somebody below what
is deserved. While the horn error is the opposite of the halo error, the severity is the
opposite of the leniency error.
There are other common errors based on biases. When you make a decision on
something based on the impulse of the very first idea that comes up in your mind,
this is known as the first impression error, also called the primacy effect. Many of
the mechanisms that underlie certain shopping behaviors are based on these kinds
of errors. You see something in a shop and you suddenly think that you cannot leave
without it! You buy it based on this first idea, but you will realize later if the money
spent was or was not worth the price. In other words, you will learn later if your
first impression-based behavior was an error or not. This error has its opposite, the
recency error, and it happens when your decision is based on the last impression
you get from something or somebody.
The tendency to systematically avoid extremes causes the central tendency error.
It is generated when individuals never take into consideration extreme alternatives
so that what they get is always on average. Again, a buying behavior directed toward
the consistent avoidance of the cheapest and most expensive products is the effect
of this error. The opposite error could be dangerous, especially when the extreme is
made of the most expensive products. This is the extreme tendency error. Last but
not least, when you decide not to do (or think of) something because of problems
that it caused when you first did it, you are committing a spillover error. This seems
to be generated by a status quo bias since it is anchored to something that happened
in the past, i.e., it is a spillover of past memories. Table 4.1 provides a summary of
all of these errors.

Table 4.1 Common errors

The error. . . . . . And its opposite!

Halo error Horn error

Central tendency error Extreme tendency error
First impression error Recency error
Leniency error Severity error
Spillover error

The simple fact that we are not omniscient leads to the implication that we make
mistakes. The point here is that these mistakes are part of the common way to any
reasoning activity. Of course, sometimes we could be better off without these errors,
but it is crystal clear that we can never avoid them completely. They are part of the
way our rationality is organized, and research shows that we are bounded. We can
also use the opposite argument writing that we are bounded because of the mistakes
we make. These studies on errors provide additional evidence that individuals are
boundedly rational.
38 4 Maps of Bounded Rationality (I)

Summary
There are many ways to show how individuals are rationally bounded. The chapter
has started the process of drawing a map of bounded rationality. For the moment,
we have explored bounds that deal with the failure to use mathematics and prob-
ability mechanisms (prospect theory), and the limits that derive from biases and
errors. The chapter presented a selected review of the findings of prospect theory: (a)
the tendency of individuals to prefer certain alternatives over uncertainty (certainty
effect), (b) the attitude to divide what is possible from what is probable (possibility
violation), and (c) the idea that risky choices are different when the positive or the
negative domain is considered (reflection effect). Associated with these effects, there
are several biases. We have presented those biases related to ownership (endowment
effect), past experience (status quo), external resources (anchoring), and social imi-
tation (bandwagon). Prejudices have been defined as a priori judgments and errors
classified on the basis of their formation. We have tried to list many of the fallacies
that can be found in organizational settings. The map is only half complete. Keep
reading!

Notes
i. Kahneman and Tversky (1979).
ii. A foundational study is Morgenstern and von Neumans Theory of Games and Economic
Behavior, 1944.
iii. Kahneman (2003, p. 1449).
iv. I prefer the vague expression behavioral studies to indicate all behavioral contributions,
including psychology, marketing, finance, and economics.
v. Examined in Shampanier et al. (2007).
vi. Shampanier et al. (2007, p. 743).
vii. This economist is Luigi Einaudi.
viii. Contrary to prospects mentioned earlier in the chapter, this example is not taken from
Kahneman and Tversky (1979).
ix. A finance professor could argue that $20,000 in the future is different from $20,000 now.
This is right but still, it is more than the $10,000 the banker offers you right away.
x. I owe this example to William Ross, Professor of Management at the University of
WisconsinLa Crosse.
xi. The literature on risk aversion and its anomalies is very extensive. For further readings, see
Rabin and Thaler (2001), Post et al. (2008), Benartzi and Thaler (1999), and Thaler and
Sunstein (2008).
xii. This point has been made clear in many papers but it is very well described in Thaler and
Sunstein (2008), and in Ariely (2008).
xii. For example, see Bazerman (1994).
xiv. Kahneman et al. (1990).
xv. Better examples could be found in Neale and Bazerman (1991).
xvi. Neale and Bazerman (1991).
xvii. See Milkovich and Newman (2008).
xviii. Silver and Mitchell (1990); Kahneman et al. (1991).
xix. Chapter 4 of the book The United States of Europe by Reid provides an interesting sum-
mary of what happens to Jack Welchs overconfidence. The title of that chapter is sarcastic
and appropriate: Welchs Waterloo.
Notes 39

xx. Change management is the subfield that studies determinants and implications of variabil-
ity in organizations. From those studies, we understand that change is first and foremost
a state of the mind. Many examples are provided by Martin (2007), in his book The
Opposable Mind where he shows that one of the most important characters of successful
leadership is related to the way these managers think about alternatives.
xxi. In the U.S. [o]ver 3.9 million students were taking at least one online course during
the fall 2007 term; a 12 percent increase over the number reported the previous year
(Allen and Seaman, 2008, p. 1). The point with online education is that it also changes the
boundaries of where a student is located. If teaching methods are corrected to fit the vir-
tual environment, graduate and undergraduate programs can reach anybody, independent
of the location. I am not impressed by the argument of the implicit inferiority of online
classes compared to the face-to-face experience. Chapters 69 offer some ideas to think
(or rethink) on these topics.
xxii. This method is based on dialogue and conversation between teacher and students, where
the role of the teacher is that of knowledge and learning facilitator. It is a learner-centered
approach that is not based on lectures where all that the student happens to know comes
from dialogue and confrontation. Plato wrote several dialogues that offer an idea of what
this art of teaching looks like.
xxii. The anchoring heuristic has been isolated and studied by Tversky and Kahneman (1974).
In the book Nudge, Thaler and Sunstein (2008) exploit this effect a number of times and in
many different domains.
xxiv. Goldman and Mason (2007, p. 284).
xxv. For example Nickerson (1999).
xxvi. This is close to the curse of knowledge in Camerer et al. (1989).
xxvii. See Leibenstein (1950); Granovetter (1978). A cognitive approach to bandwagons is
offered in Bardones PhD dissertation (and forthcoming book).
xxviii. Recently there has been a proliferation of studies on the bandwagon effect. Among the
others, see Abrahamson and Rosenkopf (1990, 1993, 1997); Chiang (2007); Deephouse
(1996); Laland (2001); Staw and Epstein (2000).
xxix. An exception is Fiol and OConnor (2003) and Secchi and Bardone (2009a).
xxx. This is what social cognitive studies tend to show (Pennington, 2000).
xxxi. Every management textbook has a section or chapter dedicated to diversity (e.g., Jones and
George, 2009).
xxxii. Usually judgment and decision making are considered together (e.g., Weber and Johnson,
2009). In the text I try to make a distinction instead.
xxxiii. It is apparent that I am not making a distinction between conscious (or implicit) and uncon-
scious (or explicit) cognition here (see Chao and Willaby, 2007), for prejudices may be
both explicit or implicit. Although this is an interesting difference to consider, I believe
that the latter is the most intriguing for organizational behavior since it is the most difficult
to analyze and govern.
xxxiv. Goff et al. (2008).
xxxv. What follows is taken from an analysis of errors in performance appraisal, in Milkovich
and Newman (2008, p. 333f).
Chapter 5
Maps of Bounded Rationality (II)

Heuristics
What is the mechanism that leads us to a biased decision? How can we be so aware
of the fact that something will happen when we have no evidence of that? How can
we be so convinced that ours is the solution to a given problem when we know it is
only a guess?
Heuristicsi are mental shortcuts. They are the mechanism that let us jump to
conclusions instead of going through detailed logical reasoning. The process that
lets us cut corners is defined this way; but the conclusion may or may not be biased
(or an error). Broadly speaking, the bias falls under the domain of substantive ratio-
nality, while heuristics is an analysis of procedural rationality. The former emerges
in relation to the objective of our decision making, the latter is the underlying
process.
We can have a closer look at the way the status quo bias works. Suppose that
a pharmaceutical company has a new drug that needs a few additional tests to be
safely put on shelves. These tests are not standard for the company, and they are
necessary for this product only, R&D managers suggest. The product is innovative
and has the potential to be a blockbuster sale item. There is a 0.07 probability that
the drug will end up having significantly heavy side effects for patients that use it
above the average dosage; this suggests the need for further testing and analysis.
However, the price per share of company stock is plummeting, and the market for
the companys existing products is decreasing. The choice is hard: take the risk and
eventually start a new period of high sales and profits for the company, or have a
break in the standard procedures and be sure that the drug is safe. The risk is very
low, and the procedures the company will follow require extra legislation, too. The
decision maker may not consider all of these implications but says a straight, Yes,
go for the market, since this is what they have always done. This describes a status
quo tendency, but also puts evidence on the fact that the decision lies on a logical
flaw. Every new drug (or new product in general) is different from previous ones and
needs to be treated accordingly. This is the logic behind innovation and, in our very
short summary of the case, the decision maker jumps to a conclusion with no sound
logic. The process that doesnt help you in making sense of alternatives, situations,
variables, etc., is a heuristic, while the result is a biased decision.

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 41

DOI 10.1007/978-1-4419-7542-3_5,  C Springer Science+Business Media, LLC 2011
42 5 Maps of Bounded Rationality (II)

Nevertheless, not all heuristics end up being bad decisions. Many times when
we use heuristics we end up with successful (or at least satisficing) solutions to
problems. In the following pages I present a few studies that bring to our attention
the fact that not being logical is not necessarily bad. On the contrary, it is what we
do most of the time.

The Fast and the . . .Frugal!

Gigerenzer and many others at the Max Planck Institute in Berlinii put incredible
effort in to studying simple heuristics. According to their perspective, these mecha-
nisms are very diffuse in our brain and make us smart. The point is that we have
an adaptive toolbox of simple heuristics that is used depending on external condi-
tions and internal dispositions. Bounded rationality is adaptive, and it is a toolbox.
The idea that we have (or develop) many tools that help in our decision-making pro-
cesses is not new. Studies on memory, capabilities, intelligence, traits, skills, etc.,iii
are directed toward the understanding and mapping of cognitive tools of the human
brain. What is new here is the focus on simple tools and not complex abilities (such
as mathematics and logics). What is simple could be used more often and, as a
matter of fact, is what helps our everyday decision making.
There is something even more important than this. The idea that our brain adapts
to the external situation is very well analyzed in evolutionary studiesiv , and it seems
to be at the basis of the emergence of the homo sapiens.v Considering this attitude
of change as typical of our brain is more than important; it is the core. How can
we add this important new feature, i.e., adaptation, to bounded rationality? I follow
the arguments that Gigerenzer and other co-authors present in their extensive and
insightful publications.vi
Simple heuristics are thought to be fast and frugal (called the fast and frugal
heuristics, FFH). They economize in the time dedicated to the making of the deci-
sion and in the quantity (and quality) of information they use. A fast and frugal
heuristic is, for example, what makes us think we prefer one piece of cake instead
of the other when we see an indefinite number in a bakery. Think about this for a
moment: You see an indefinite number of cakes. You dont have the time to ask what
each cake is made of; you lean on past experience and the appearance of the cake.
Moreover, the bakery is overcrowded so you feel like you dont have time. What
will your choice be? You have very limited visual information (maybe a few words
from the seller) and limited time.vii Your choice will be based on a fast and frugal
heuristic. Mine doesnt work very well because I always choose the wrong cake; it
looks beautiful and tastes awful!
Another example of a simple heuristic that helps us all the time is how language
works. This may seem counterintuitive, but the activities of speaking and writing
are made possible exactly because we use these sorts of mechanisms. The words
we choose and their sequence are only partly dictated by a rational selection of
choices. In fact, our speech and writing follow a well-established syntactical order
Heuristics 43

(when available), and a selection process that relies on which word is close to what
we have in mind at that moment. While I am writing, I do not spend the same
amount of time making decisions on every single word that goes in this book (I
probably should!). Instead, I write sentences as they come, and I seldom think about
the right word. This is faster because the choice is made in limited time, and the
information I use is also limited. Sometimes a dictionary or thesaurus helps me find
a better word, but I am never sure that my exploration and analysis of alternatives
has been fully explored. This example becomes more significant when we think
of a foreign language. What happens when we use a foreign language? There are
different strategies that our brain uses, but the most common are two: (a) when
you use sentences that you have already learned that come out without any effort
(i.e., How are you? Im alright, What time is it?, and the other survival-like
sentences) and (b) when something unexpected happens and you have to quickly
come up with something to say. This last case is more complicated and, in many
cases, you have to think about which words to use and in what sequence. Suppose
you are in France and ask for two tickets you are about to visit the Muse du
Louvre. Everything seems to go very smoothly but, at a certain point, the ticket
seller asks you something regarding an audio guide. You were thinking about the
Mona Lisa you are about to meet and are not prepared; you are not even sure that
this is exactly what the seller just said. You have to make up an answer, saying
something that is closer to what you should have said in your language. Of course,
the alternatives of your mother tongue are quite infinite since you can choose among
hundreds of expressions. You may have a polite answer such as That is very kind
of you but no thanks, I have my guide. You should have told her a sentence trying
to use humor, like Wow, thanks. . . and do you have it with Dolby-surround? Or,
there is the quick answer that you never choose: No, thank you. In this case, the
Non, merci bien seems very likely to occur, but the very first of your thoughts
were on which are the right French words for the dolby-surround sentence. Time
is extremely limited, the question is simple, people in line are waiting and you dont
have access to relevant information (i.e., you do not speak fluent French). Your
fast and frugal heuristic helps you find an appropriate answer. And the result is
successful: You dont pay more money for an audio guide you dont really need.
This last example is particularly powerful regarding what we can learn on FFH.
Studies on these mechanisms are recent and many questions remain unanswered.
Research is ongoing, and we are confident we can get more from it soon; however,
the idea that something so basic and widely used such as spoken language uses
FFH suggests that these mechanisms are extensively used and are a good starting
point for the study of decision making. These mechanisms are explicitly structured
to let us make decisions in a timely fashion and without a significant amount of
information. It seems that our knowledge structure and the mental frames we use
rely dramatically on these heuristics.
One last example that I can use is a fascinating experiment on bounded ratio-
nality. I have even tried this experiment in one of my classes. My results are not
significant because I use it as a teaching tool, but students of FFH run similar exper-
iments on the same subject matter. I report my experience relying on their data for
44 5 Maps of Bounded Rationality (II)

significance.viii I asked students which is the larger city, San Diego or San Antonio.
Most of the students answer San Diego, based on their knowledge. When asked
about this knowledge of theirs, they tend to raise interesting points. They say that
they have been there once and they know how big the city is, they know somebody
who lives in that city and that person told them how big the city is, they say that
they dont know the other city and they answered from what they knew. Basically,
the answer is based on ignorance. I then asked if they knew of any data on popula-
tion in the two cities; it became apparent that nobody really knew the answer to the
question. What happened? The students used FFH to make a decision. They tried
to retrieve information on the two cities from their memory and, when this retrieval
ended up being related to one of these two alternatives, they tended to rely on the
only data they had. This led to a discount effect on information they didnt have. If
you dont know it, it must not be relevant! This is fast and frugal heuristics at work.
What emerges from this experiment, that was not evident from previous examples,
is that FFH are not always right. We can summarize that they are not accurate and
uncertain in the appropriateness of their outcome. FFH can bring you to inaccurate
answers that end up being not useful or not appropriate for the problem you are
facing.ix
FFH have not been fully analyzed in organizations. How do they affect man-
agerial choices? How could personal experience, skills, capabilities, and other
variables affect the development of fast and frugal heuristics? Under which con-
ditions do organizations favor the emergence of FFH? Are FFH always successful
in organizational settings? This is a list of questions that could and should be
answered. Organizations are social environmentsx where individuals spend a signif-
icant amount of their lives. This means that organizations are shaped by individuals
that work in them. Therefore, we need to consider if and how individuals using FFH
affect organizations. In the following pages I present a few hypotheses (or thought
experiments) on how FFH (a) emerge, (b) are embedded, and (c) are favored in
organizations.

Organizational Heuristics
The starting point is that individuals use FFH. I have described how individuals use
FFH for personal decisions. Is this individual tendency also reflected in the work-
ing environment? Yes, these heuristics emerge in organizations as well as in other
environments. Some decisions need to be quick and made with limited information
available. There, individuals lean on what they know more than on what they should
have known. The decision to set up a goal, select and recruit, launch a new prod-
uct, or open a new branch or subsidiary leans on simple heuristics.xi Think of one
of the processes mentioned above, recruitment and selection. There is no need to
look at the entire processxii , but only at the beginning of it. The fact that the depart-
ment has an independent budget doesnt mean that every team is allowed to recruit
without authorization. The process of getting authorization is complex, most of the
times, since the head of the department needs to look at available funding and get
Heuristics 45

a first informal approval from the human resource department. During this process,
whether the answer is yes or no, she will hear many motives supporting the
choice. The fact is that, at the very end of the process, when recruitment is approved
it could be for any reason. I am not saying that it is not rational, quite the opposite.
It is rational because the choice relies on simple heuristics that make a decision
possible. If for every position opening the process should be that of fully evaluat-
ing the case, no position will be easily opened in a few weeks. Shortcuts are used
instead. The idea to let those departments or subsidiaries open positions may not be
well-justified in terms of needs, but will be in terms of reward, for example, if that
team usually shows an above average performance. Observing common and diffuse
organizational facts like this one, we can assume that there is emergence of FFH.
Some of these heuristics may also be embedded in organizational processes.
Organizations work through formal and informal structures.xiii The informal (also
called behavioral) is made of individual recurrent behavior, norms interpretations,
and culture, while the formal structure (also defined as the normative structure) is
defined by role expectations, values, and norms. It is this last form that offers the
potential to incorporate simple heuristics. The logical passage is simple: If individ-
uals use FFH, then it is probable that the social environments where they spend a
vast amount of time (e.g., organizations) develop a structured way to facilitate the
individuals to operate through FFH. When a behavior or a mental attitude becomes
commonplace, then it becomes a social behavior or something which is favored by
the community. This is the way FFH enters the organization. The next step is that it
can be formalized and become a norm. This probably changes the nature of what
is discussed here; however, this means that further work is needed in this particular
domain of organizational behavior.
Or, there might be a second way for FFH to become part of the organization.
Instead of being integrated as part of a way to make decisions, simple heuristics
may be accepted and favored. A comment pointing at a specific direction, for exam-
ple, receives more interesting and supportive feedback than a comment pointing at
another, less probable direction. The team supports and encourages your thoughts
on that Return on Investments (ROI)even if you dont know the exact numbers
that define itbecause it is commonplace to discuss about that. When you take
under consideration the Return on Assets (ROA), people dont care about what you
are saying even if you do know how this is generated in detail. Heuristics work
in the first and in the second cases; however, the process/team supports heuris-
tics when using ROI, while it does not when using ROA. Besides these financial
measures, the point is that FFH sometimes needs the support of the team or orga-
nization where it emerges (more on this social dynamic in the second part of the
book).
When simple heuristics are considered, I believe that the most interesting element
is that they can be used to explain a vast amount of mental processes. Can we cat-
egorize FFH? Can we classify them? How can we get more from these processes?
Organizational behaviors that rely on simple heuristics have been overlooked in
the literature but maintain a significant potential to explain individual thinking and
behavior.
46 5 Maps of Bounded Rationality (II)

Emotionally Bounded
In one of their works, Human Problem Solving, Newell and Simon offer reasons
why they decided not to include emotions in their analysis:
Our final choice is to exclude motivational and personality variables what Abelson (1963)
covered in part with the term hot cognition. We omit them by reason of convictions,
not about the importance or unimportance of the phenomena, but about the order in which
theory should develop. Many motivational and emotional phenomena operate through the
lens of the cognitive system [. . .]. A plausible scientific strategy is to put our cognitive
models in order before moving to these other phenomena. Our one explanatory foray in the
direction of motivation and emotion was precisely in this vein [. . .].xiv

Emotions have always been excluded from the rationality discourse. For a long time
decision-making scholars thought that feelings and emotions weaken and impair
decision making. Emotional was opposed to rational decision making. This is
not to criticize rationality scholars, since Platoxv stays at the very origin of this
dichotomy with his myth of the black and the white horses. This is to say that the
way these two worlds are opposed is the way the entire western philosophy and
cognitive science have developed. And Newell and Simon are no exception.
This ideal led thinkers of the past and present to believe that everything rational
must be under mindful control and it can be computed by the human brain. The
Leibnizian nihil est sine ratione (nothing is and happens without a rational expla-
nation) served to define exactly this idea. According to that classical tradition of
thought, if you can have an algorithm that describes any event after it happens, i.e.,
ex post, you can have one that explains the same phenomenon before it happens, i.e.,
ex ante. Even the most complex and random event can be reduced to a cause-effect
relation using an ad hoc and ex post explanation. Mathematics is a very useful tool
that serves this end, and it has been used many times to uncover patterns once
the event has ended. Financial crises, economic collapses, rising stock prices, social
trends, et altera, have been defined by econometric models that analyze how those
events happened. However, this is not what we need to deal with here. It is not my
intention to offer a literary review on advancements of mathematical modelingxvi in
economics, decision making, or management.xvii Although this is a very stimulating
debate, I would like to suggest a different perspective.
My point is that the explanation of everything throughout the logic of mathemat-
ics (and reason for those thinkers who used it) fulfils the expectation that the scientist
is willing to find in the observed phenomenon:xviii a pattern of relations. And this, in
turn, supports the impression that you can find that same relation ex ante. Of course,
if scientists were able to do that, we would be in a world of certainty; fewer biases
and heuristics would be needed.
What the (old ideas of the) cold world of mathematics also suggests is that anal-
ysis must be objective and it can be done in vacuum (i.e., in complete isolation).xix
Instead, neurological studiesxx have found that emotions are associated with the
process of making decisions. This means that we are not capable of making any
decision without them. The example of mathematics is also interesting from this
perspective. Every mathematician, or anybody who uses mathematics to find a
Heuristics 47

solution to a problem, enjoys the fact that she is close to a solution (when she is).
These positive emotions are associated with different configurations of the brain and
the release of chemical substances that help the search for a solution.xxi If we were
rational in the sense that Leibniz (or Descartes) thought we were, we would never
be able to find any solution.xxii Emotions trigger the ability to make decisions; thus
they facilitate the solution.
The approach taken here is narrower than the one mentioned above, though. I
want to point out what emotions deal with bounded rationality. As Hanoch puts
it,xxiii emotions work with mechanisms different from rationality, but they can help
the way individuals process information. He isolates the areas where emotions con-
tribute to rationality by restricting the range of options considered (reducing the
load on short and long term memory), [by] focusing on certain variables (cer-
tain stimuli receive higher ranking order), and [by] initiating and terminating the
evaluation process (working as a satisficing mechanism).xxiv There are two major
contributions to rationality in Hanochs statement. The first is to consider emotions
as a prioritizing mechanism: They help us to rank the alternatives operating within
our bounded rationality. The second is a help focus on particular situations, events,
facts, and circumstances.
Emotions are studied within organizationsxxv with a growing emphasis in recent
years. The outcome of many of these studies is that too much emotion is harmful
in that it prevents social relations from forming, detrimental to productivity, impairs
the ability to focus on specific issues, lets the worker misrepresent reality, etc. Few
studies have been conducted on how emotions relate to bounded rationalityxxvi in
organizational settings. We can speculate that it is very hard to find such evidence
due to problems relating to experimental and field analyses. One solution to these
problems may reside in the field of neuroeconomics.xxvii
This new field studies how various parts of the brain are activated when indi-
viduals are confronted with economic decisions. When we apply this to emotions,
we should be able to determine how emotional states are associated with decision
making. Technology allows us to study this outside the laboratory.
The fact that we use emotions to make quicker decisions is apparent. One of the
most-used examples is that of the snake. Suppose you are walking in a forest and you
see a snake close to your right foot. What should you do? I am sure you do not stay
there and rigorously calculate which alternative is better, following the definition of
rationality exposed in one of the previous chapters. What you do is jump as far as
you can from it. You dont even know if it is a dangerous snake or not! But, this
is what you do: A quick solution that could potentially have saved you from a very
painful bite. This is what emotions do: They help you prioritize among behavior
alternatives.xxviii Now that we have defined a prioritizing mechanism, I would like
to writeas a marginal notethat this example could not be very appropriate. If this
is an emotional response, what is an intuitive one? What is the difference between
emotion and intuition? Are intuitive decisions always emotional? And when can we
say that a decision is more intuitive than emotional?
With the next chapter I start answering these questions, but it seems to me that
the task is not easy. For now, I suggest that we look at what intuition is. It is a
48 5 Maps of Bounded Rationality (II)

decision that comes out apparently from nowhere and uses knowledge already in
our brains. Intuition is typical of improvisation for musicians; patterns are known,
and the right notes come at the right time according to what the musician is able and
wants to frame. Past knowledge is relevant and emotional involvement is impor-
tant, too. Intuition can be defined as a set of fast and frugal heuristics that is often
triggered by emotions. Isnt this a bridge between the two areas?

Accessibility, Representation, and Framing

Psychologists isolated three issues that define how people approach problems. These
are of utter importance when it comes to defining our rationality bounds.
The first is related to accessibilityxxix of data. How we are able to get information
from a given data set is related to how it is accessible to us. Accessibility deals with
the ease with which we get the data. A data set could be more or less accessible to
us depending on its configuration and contents.
Then comes the issue of representation of data. The way data are structured
and the relations made between variables deals with the way we make sense of
information.
The third element is the mental framing of information. How we make sense of
information depends on how we organize and analyze data. This, in turn, depends
on our past knowledge, conditions at the moment the analysis is performed, and on
our abilities.
Think of a corporate report from a public company. This report is usually avail-
able online, which makes it accessible to most of us; almost anybody can download
and get easy access to it. When you download an annual report, for example, you
start with the letter from the CEO or sometimes from the chairman of the company.
This is commonplace. The next section introduces you to the world of numbers
with highlights from the balance sheet, cash flows analysis, and the profit and loss
account for the last three fiscal years. If you are not an expert, it is hard to make
sense of these numbers. This means that the representation of those data does not fit
your understanding, and you cannot (or have a hard time) make(ing) sense of them.
When you turn pages, you see figures of data where histograms and pie-charts rep-
resent the data under an easier and clearer perspective. There you have an immediate
idea of how sales have grown, which costs have been cut, what the weight of labor
cost is on total sales, how profits are stagnating, and the like. This representation of
the same data is more helpful and actually makes the difference. Also, the way you
frame information is different in the latter case. This is not the first time you see
these kinds of charts, and this is the reason why data are more accessible here: They
are represented in a way that fits your mental frame.
These manipulations of data and information are often used by companies to trick
their stakeholders understanding. Suppose you find the following statement written
in the website, bills, annual report, of an energy company: HIJ company is the first
provider of clean energy in the nation. The trick here is subtle and well-played
Accessibility, Representation, and Framing 49

since you are induced to believe that HIJ produces most of the clean energy they
sell. However, they dont state that; they write that they provide clean energy, but it
is not clear whether they produce it or not. Of course, there is a significant difference
here. A company that produces clean energy and is the first provider in the nation
invests heavily in renewable sources and is willing to take the risks associated with
innovation. Also, in-house investment plans may indicate that the company believes
that becoming more environmentally friendly and socially responsible is part of
what the highly pollutant energy industry should do. Instead, a company that buys
clean energy is not taking any significant risk, nor is it showing that renewable
energy sources are strategic for them.xxx However, if the stakeholders (customers,
for example) limit their knowledge to what is accessible to them through the bill or
the home page of the companys website, they will be fooled by the tricky wording.
Representation of reality is flawed, and framing somehow impaired. I believe this is
an example of how companies PR play their cards, knowing that stakeholders deal
with accessibility, representation, and framing effects.
I believe that what emerges from the examples and from a more accurate analy-
sis of the three factors of accessibility, representation, and framing is that they are
intertwined so that it is not clear where the effects of the first end and the ones of
the second and third start. Despite all attempts to study these factors in separation, I
believe that a better job could be done with the study of their interrelations.

Epistemological Corner
Heuristics, biases, and errors can be defined through classic logic as fallacies. A first
and general definition of fallacy is that it is
a common misconception, that is, a false proposition which is widely and confidently held,
and which carries for the holder the strong appearance of truth. For a long time, logi-
cians [. . .] bring to their investigations a somewhat more restricted conception, according
to which, a fallacy is a common mistake, not about belief, but about reasoning.xxxi

Therefore, a fallacy is a procedural error in reasoning that gives the illusion of being
right to the decision maker. To understand why they are relevant to our discourse, I
suggest giving an explanation of the most relevant fallacies and then proceeding to
link these to our analysis of decision making and rationality.
One of the most used list of fallacies is known as the Gang of Eighteenxxxii
(Table 5.1.)
I found in Gabbay and Woods work that fallacies with asterisks appeared in a list
that dates back to Aristotle. These fallacies support arguments that are (a) erroneous,
(b) attractive, because they look correct, (c) universal, because they are employed
very often, and (d) incorrigible, i.e., they tend to be repeated again and again.xxxiii
This leads to the acronym EAUI, often used in that discipline.
It is not my intention to define each one of the 18 members of the gangxxxiv ,
because many of them are not relevant in the economy of our discourse. However, a
few of them are useful and strictly relate to concepts such as heuristics and biases.
50 5 Maps of Bounded Rationality (II)

Table 5.1 The Gang of

eighteen 1. Ad populum
2. Ad hominem
3. Ad baculum
4. Ad misericordiam
5. Ad verecundiam
6. Ad ignorantiam
7. Secundum quid ( )
8. Hasty generalization
9. Affirming the consequent and denying the
consequent
10. Begging the question ( )
11. Many questions ( )
12. Equivocation and emphibody ( )
13. Post hoc, ergo, propter hoc ( )
14. Insufficient and biased statistics
15. Composition and division ( )
16. Faulty analogy
17. Gamblers fallacy
18. Ignoratio elenchi ( )

Note: Fallacies with ( ) were isolated by Aristotle

(see Gabbay and Woods, 2007).

For example, the ad populum fallacy occurs when individuals support or accept
an argument because it is what most people believe. Sometimes academicians are
not exempt from falling under this logical trap; many papers and books are cited
because they are popular, because everybody cites them. Of course, this does not
say anything about the arguments presented in those books and papers. However,
it seems that they gain their power on the basis of how many times colleagues cite
those works. In the academia, this process is particularly harmful because it leads the
attention of scholars far from the contents of our work and to the typical shortcut
of leaning on a popular work. This is close to the way the bandwagon effect was
defined in the previous chapter. Examples of this fallacy abound. In marketing, for
example, sentences such as This is the best-selling brand! tell nothing about the
product, only that it is popular.
Close to the ad populum, there is the ad verecundiam (appeal to authority). This
can also be defined as the ipse dixit of medieval flavor (trans. he said himself;
used to support propositions that needed no proof because hethen, Aristotle
said himself). It works every time individuals advocate and take an argument as
true because of the supposed knowledge of the one that first stated it. This is partic-
ularly diffuse among believers in all religions, and it emerges every time you believe
that something is true because of the source from where it came. The argument is
fallacious because you are not making any evaluation of its contents; what you are
stating is that it must be true because of the position of the person who is saying that
(e.g., Harvard professor, the President, the Pope, the Dean). Once again, marketing
techniques seem to exploit this fallacy very well when advertising is based on the
fact that a product is the best among certain categories of people (e.g., #1 for the
American Dental Association).
Two Logics 51

An ad hominem proposition, in one of its forms, is based on the person and not
on the argument. When individuals disregard information because of the reputation
of the person instead of looking at its content, they are operating on the basis of a
logical fallacy. Another way to explain how this works is that of using characteriza-
tions. It is like being against the suggestion to read this book because of the person
who suggests it, the author for example, advancing the argument that there is a bla-
tant conflict of interests. This argument has nothing to do with the content of the
suggestion but it is a powerful way to invite people not to deal with it.
The hasty generalization is the action of inducing a general assumption from one
or a small number of events. The fact that you have visited the campus at Stanford
University doesnt mean that all American universities are that beautiful. Even when
people visit more than one campus, lets say five from east to west, they cannot make
generalized assumptions based on what they have seen. This is not a significant
sample, out of hundreds of US universities. Every time we make a poor use of
induction and generalize on the basis of limited information, we are falling into
this particular category of fallacies. Scientific works are not exempt from showing
signs of hasty generalizations; think about how many studies are based on a limited
number of students! We may argue that in this latter case, it is more abduction than
induction at work (see Chapter 2).
The fallacy composition and division is also very common in decision making.
It is the faulty logic that leads you to the attribution of characteristics of the parts
to the whole (composition) or, vice versa, to the attribution of characteristics of the
whole to the parts (division). The latter happens when you deduce that the goal of
each member of the A team is quality improvement, since that is the mission of that
team as a whole. This lets you overlook that they should have divided quality into
different aspects so that every member might have a specific subgoal. The division
fallacy is very diffuse among neoclassical economics. The assumption that the goal
of the bank is, for example, to maximize its profits makes each banker share the
same goal.xxxv This has to be demonstrated. It is a false relation (a logical fallacy)
unless somebody shows some correlation between the two utility functions, that of
the bank and that of its employees. The former (composition) becomes apparent
when you think that the team is a living creature because all of its members are
living human beings.xxxvi
What is interesting in the way the idea of fallacies is portrayed by logicians is that
it could offer a sound basis to studies on heuristics, biases, and errors. More than
2,000 years of study is a significant tradition that can be exploited when analyzing
dynamics and inherent logic of faulty mechanisms.

Two Logics
How do all of these rational bounds work within organizations? This question is
not easy to answer and it probably needs more than the end of this chapter to be
answered. However, we try here a first explanation based on what Marchxxxvii calls
52 5 Maps of Bounded Rationality (II)

the two logics. Decision making can be thought of as a process where individuals
answer questions and follow specific patterns.
Marchs idea is that we can identify a first rationalistic logic that follows a
strict and well-defined pattern. This is the logic of consequence and it is very close
to a decision maker that operates in vacuum, i.e., without an environment. The sec-
ond is a logic of appropriateness, where the individual is embedded in a situation.
In the following, I explain briefly what these two logics are about, posing a partic-
ular emphasis on the second which is more likely to explain why individuals make
mistakes, use heuristics, fail in many circumstances or, in two words, why they are
so boundedly rational. The goal is to provide a framework for everything presented
above that maps BR (Chapters 4 and 5).

The Logic of Consequence

In a strictly formal logic such as this, there are four questions that individuals (and
theorists) ask. I consider these four very briefly below.
The first question is that of alternatives. Followers of this logic need to define
which are the courses of action that are possible and map them as best as they can.
Only if you have this map you can start thinking of expectations, i.e., consequences
associated with each alternative. This process is close to what Simon defines as
rationalityxxxviii (step 4, see Chapter 2). The third element is that of preferences
associated with the decision maker. Values and principles affect the way the decision
is made; in many cases, values drive the decision-making process so that individuals
choose according to what they believe is right. Since time is often more precious
than the effort put on getting the best possible outcome, we need a decision rule.
This is what allows us to stop the search, focus on alternatives and consequences
that make sense to us (value), and decide.
This logic well describes the original idea of bounded rationality that was orig-
inally introduced by Simon. Individuals make their choices selecting alternatives
through a system of values. These preferences are evaluated in the potential out-
comes they can provide to the decision maker. The fact that rational agents use
heuristics, have problems with data accessibility and representation, are biased,
and emotionally bounded, is part of how this process is shaped but does not
change it in its major components. Still, with bounded rationality the logic of
consequence works as well. Alternatives generated could be less than expected,
preferences could be biased, decision rules subjected to emotions, and expecta-
tions defined on the basis of heuristics. This is to say that this logic works with
individuals like you and me despite its appearance of perfect chain of events
(alternatives-expectations-preferences-decision rule).
Although the logic of consequence is powerful in what it defines, March believes
that there is something missing. And this something, as you will see in the coming
pages and chapters has the potential to change the way we understand and analyze
decision making.
Two Logics 53

The Logic of Appropriateness

The second logic is presented through the analysis of three questions. When
decision making happens in organizations or in any social setting, individuals
need to answer questions that pertain to those specific circumstances. This is
the question of recognition that allows the decision maker to understand what
situation he or she is facing. It is the analysis of conditions influencing the cir-
cumstances that affect or may be affected by the decision that is about to be
taken.
Together with this the decision maker needs to understand the role that he or
she is playing in that situation. This is the problem of identity, and it is crucial
when players dont want to lose track of their potential to achieve the goal. Of
course, when you are describing a situation where the decision is to be made in
an organization, then the role and position of the decision maker may make the
difference. For example, if you are about to make a decision on a new produc-
tion line, your decision needs to be defined depending on the fact that you are
in charge of that new line of production, you are an R&D manager, you are the
CFO, or you are a middle manager. This is going to affect the way the decision is
shaped.
Roles and positions are also related to the rules that apply to the situation.
Different roles often lead to different takes on the same problem. This may depend
on multiple variables such as (a) personal attitudes, (b) abilities, and (c) possibilities
of making the decision. This latter point depends on the power associated with the
position, sometimes more than with the willingness and boldness of the decision
maker.
These three simple factorsrecognition, identity, and rulesare supposed to
define how the individual makes decisions in organizational settings. We can use
an incremental logic to consider these two logics together. I dont believe that
the second excludes the first; that is, if you follow the second you cannot use the
first. I believe that they depict different perspectives so that the logic of appro-
priateness includes that of consequence. The three factors posit questions that are
inherently different from those used in the first logic. Of course, the problems that
individuals face when they operate in social settings need to be framed through
a more socialized context so that the decision maker that faces a problem may
not be able to actually solve it because his/her position is not appropriate. The
fact that a shop manager at a fast food restaurant recognizes that deliveries of
meals at the drive-through window are not fast enough and finds a way to make
them faster does not allow him to change procedures. The decision-making pro-
cess follows a standard operating procedure that has to be the same for each shop
in the restaurant chain. The decision is not appropriate according to the role of
the shop manager. It is a direction that is more centralized and depends on some-
body that makes that decision out of the power he or she has. The process can
also be evaluated through expectations, alternatives, preferences and past the deci-
sion rules, and nevertheless it cannot be followed, according to the appropriateness
logic.
54 5 Maps of Bounded Rationality (II)

Implications of Using One or More Maps

In the last two chapters, I have offered an overview of those theories, models,
and approaches that contributed to the legacy of Herbert Simon and to the idea
of bounded rationality. What you have found in these pages is far from being an
exhaustive and complete review of these studies, rather it offers a selection of some
of the most influential works. Some concluding remarks on the impact of these
works are probably needed.
Besides the two logics that offer a framework through which decisions may be
analyzed, there are two paradigms that seem to emerge. One was originated by the
work of Kahneman and Tversky, and we may call it the biases paradigm; another is
based on that of Gigerenzer and his colleagues, the toolbox paradigm. Although it
may be implicitly or explicitly stated in this and in the previous chapter, in the fol-
lowing pages I describe what is the most significant import that these two paradigms
add to the original idea of bounded rationality.

The Biases Paradigm

First of all, I am not fully satisfied with the choice of the label used to define this
paradigm. Of course, biases are not the only interest for this subfield. However,
the choice falls to that word since this is what the two pioneers that originated the
enormous amount of studies focused on. Having written that, as far as this book is
concerned, I can summarize the most important contributions of this paradigm in
two points. The first is that it fueled a revival of behavioral studies. The second is
that it helped analyze and understand the computational limits of rationality.
Behavioral studies. The biases paradigm gives to behavioral studies a signif-
icantly different origination compared to that of Simon. The father of BR used a
deductive approach, founding his idea on critiques to a previously existent theory.
Instead, the biases paradigm is based on empirical findings that reject assump-
tions of that same neoclassical theory from where Simon started. This appears
as a tremendous change from the origins of BR, since it founds behavioral stud-
ies on empirical research instead of theoretical elaborations. Therefore, we have
two traditions (now merged). The old behavioral research is concerned with the
study of theories that explain how people think and behave. The new behavioral
studies found their research on empirical findings and eventually build theory
out of experimental results. The latter approach has become the standard for
behavioral research, contributing enormously to find consistent empirical evidence
of BR.
Computational limits. What type of BR are these studies looking at? In order to
make a significant contribution, the biases paradigm focuses on a particular aspect of
bounded rationality. The connection between the idea of BR and the paradigm is that
violations of neoclassical economics axioms are connected to computational limits
of the human mind. And this is exactly one of the two sets of bounds that define
Implications of Using One or More Maps 55

rationality. Due to a wide range of studies, we are now able to define computational
limits of the brain, have practical examples and applications, analyze implications
stemming from these limits, and measure these internal bounds. This provides BR
with a solid and measurable base.

The Toolbox Paradigm

This second and more recent paradigm contributedamong the many other
achievementsto defining two points that relate and improve the original BR. One
is the concept of ecological rationality, a second is the focus on external information
limits to rationality, and the last one is the corollary less-is-more.
Information limits. Studies from this paradigm take an approach that is slightly
different from what we have seen in the biases paradigm. The starting assump-
tion for FFH is that [f]or the most part, however, theories of human inference
have focused exclusively on the cognitive side, equating the notion of bounded
rationality with the statement that humans are limited information processors,
period.xxxix This is why they decided to start from the second set of bounds that
relate to limited access to information. In so doing, the toolbox paradigm has been
able to develop significant insights on how individuals analyze, frame, and collect
information.
Ecological rationality. Although Gigerenzer and others always see themselves as
loyal followers of the work of Herbert Simon,xl they came out with a redefinition of
rationality on the basis of its ecology. In short, their studies on heuristics (FFH) and
human cognitive processes led them to highlight one of the features of BR, defined
as dependent on availability of external information. The study of heuristics is a
study of matching between these mechanisms and the environment,xli where infor-
mation is located. There are adaptation and dependency patterns that these studies
highlight.
Less-is-more. This is a corollary that Gigerenzer and others have been able to
derive from the major assumptions of the toolbox paradigm. The heuristics mech-
anism is based on the interaction between external information and the blink that
allows individuals to make a decision. There are circumstances where the more
information people get, the less accurate and efficient their heuristics become. This
intriguing finding has many implications, one of them being, for example, the fact
that it gives us a scientific explanation of the reason why, after hours of search on
the Internet, we sometimes come out with answers that are not worth the time spent
in the search.

Final Remarks
It is now apparent that the two paradigms should be considered together since
biases explain how computational limits work while the toolbox focuses on external
bounds. And we know that these two together define BR.
56 5 Maps of Bounded Rationality (II)

The most important factor associated with using a map is that we do not get lost,
and it allows us to find our way once we get lost. To some extent, this is what maps
of BR are for. They help scholars and students of rationality find their ways to the
bounds when analyzing and understanding human cognition. A map of BR points
to these bounds and helps find them. Truth to be told, for the most part, mapping
studies show exactly this, and highlight how limited our rationality is. However,
maps are also sensitive to the tools at hand. If you happen to look at the map of
a European city drawn during the Middle Age, you will find it very different from
what we consider a reliable and valid map today. Even those parts of the city that
have not changed that much are not easy to recognize. The same thing happens if
you look at a geographical map of any area of the (known) world as it was seen in
the twelfth century. It certainly does not look like the one we are used to.
This analogy is useful to introduce two elementary concepts. The first is that
we do not know how accurate these maps of BR are; so far they seem reliable, but
how sure are we? What if our assumptions on cognition change as, in the analogy,
the assumptions on the shape of the world changed? Accuracy of tools also plays a
role in this process. The second is that doubt should always be at the basis of every
scientific inquiry; this is how we can improve and advance. The next chapter starts
instilling some doubts on the influential ideas reviewed in these pages.

Summary
The maps of bounded rationality are now completed. In the chapter we have
learned that heuristics allow us to make decisions. Especially simple heuristics,
those concerned with limited time and information (FFH), seem to be the most
widely diffused and an easy mechanism for our brain. Emotions are also an aid
to bounded rationality because they function as prioritization and focusing mech-
anisms. Accessibility, representation, and mental framing define how individuals
make sense of information. The epistemological corner opens a window on a very
long tradition of studies that share something in common with it, and is particularly
useful to analyze errors, heuristics, and biases. We have also pointed out that this
theory of bounded rationality works under the logic of consequence, while orga-
nizational behavior needs to be tied to the logic of appropriateness. Finally, some
of the most far-reaching studies that have been published in the last 60 years can
be divided in the biases paradigm and the toolbox paradigm. Differences with the
original idea of BR are then analyzed. With the next chapter I start telling a different
story.

Notes
i. Heuristics are usually defined as rules-of-thumb that people often rely on rather than
using more appropriate statistical rules (Kunda, 1999, p. 53).
ii. Gigerenzer et al. (1999), Gigerenzer and Selten (2001), Todd and Gigerenzer (2003),
Goldstein and Gigerenzer (1996), and Gigerenzer and Brighton (2009).
Notes 57

iii. Evidence is provided in Gigerenzer and Selten (2001), and also discussed, on a different
and cognitive perspective, in Thagard (2007).
iv. See, for example, Humphrey (1976), and Odling-Smee et al. (2003).
v. Gazzaniga (2008).
vi. Payne et al., (1993) study the adaptive decision maker in the tradition of Herbert Simon.
Their work has inspired Gigerenzers studies on heuristics (Gigerenzer and Brighton,
2009).
vii. Todd (2001), offers a more detailed explanation of time and informational constraints
leading to fast and frugal heuristics.
viii. The example is taken from Gigerenzer et al. (1999).
ix. A quick note on the population of the two cities. It is about the same; the last census
(2000) showed San Diego leading with 1.22 million people and San Antonio with 1.14.
However, 2007 estimates show San Antonio ahead with 1.32 and San Diego stuck at
1.26. Tricky question!
x. Part V of Gigerenzer et al. (1999), is dedicated to social intelligence; we can consider
this as a premise, although insufficient, for an organizational analysis of FFH.
xi. Textbooks are usually optimistic and provide recipes of what managers should do
to set up goals, plan, hire personnel, and perform other important managerial activ-
ities. However, there are too many cases when these processes are not followed
thoroughly. Grove (1999) at Intel and Semler (1994) offer insightful revisitations of
classical approaches to how managersespecially top managers and executivesmake
decisions.
xii. For the full story, see Mathis and Jackson (2008).
xiii. Scott (2003) offers a description of the social structure of organizations in the first
chapter of his book Organizations. Rational, Natural, and Open Systems.
xiv. Newell and Simon (1972, p. 8).
xv. There are many translations of Platos Phaedrus; I used Penguins 2005 translation by
C. Rowe.
xvi. Books on mathematical modeling for social sciences abound because the needs of
researchers can be very different in terms of what model can be used. I have recently
used Shier and Wallenius (2000), Luce and Raiffa (1989), Meerschaert (1999), and
Dreyer (1993).
xvii. Formal models in mainstream management are not significantly diffused. It is not until
recently that the Academy of Management Review published a special issue on formal
mathematical modeling in management (vol. 34, No 2, 2009). Interesting insights on
formal models of managerial and organizational behaviors are usually published in the
journal Computational and Mathematical Organization Theory.
xviii. Not exactly related to this is the problem of observer-phenomenon relations and the
so-called second order cybernetics (von Foerster, 2002).
xix. Ilya Prigogine and Isabelle Stengers would not agree with this statement. In their
work (1986) they show that complexity is what drives scientific discovery and that the
simplicity of relations is an illusion that we carry forward from the old approach to
science.
xx. I refer in particular to the study of Damasio (1994) and Bechara and Damasio (2005).
xxi. This is close to what Antonio Damasio presents in his book Descartes Error and calls it
the somatic marker hypothesis. He explains it using the following example: Imagine
yourself as the owner of a large business, faced with the prospect of meeting or not with
a possible client who can bring valuable business but also happens to be the archen-
emy of your best friend, and proceeding or not with a particular deal. [. . .] When the
bad outcome connected with a given response option comes into mind, however fleet-
ingly, you experience an unpleasant gut feeling. Because the feeling is about the body, I
gave the phenomenon the technical term somatic state (soma is Greek for body); and
because it marks an image, I called it a marker. [. . .] What does the somatic marker
58 5 Maps of Bounded Rationality (II)

achieve? It forces the attention on the negative outcome to which a given action may
lead, and functions as an automated alarm signal which says: Beware of the danger
ahead if you choose the option which leads to this outcome (1994, pp. 170, 173).
Needless to say, the somatic marker hypothesis works for both negative and positive
prospects. One of the strengths of this approach to decision making is that it finds a
way to reconciliation for positions since [t]he partnership between so-called cognitive
processes and processes usually called emotional should be apparent (p. 175). As
Damasio puts it, understanding neurobiological mechanisms behind some aspects of
cognition and behavior does not diminish the value, beauty, or dignity of that cognition
or behavior (p. 176).
xxii. This is clear from Damasios work (1994) where he describes patients with neurological
impairments.
xxiii. Hanoch (2002).
xxiv. Hanoch (2002, p. 7).
xxv. Almost 10 years have passed since the Journal of Organizational Behavior (2000, vol.
21) dedicated a special issue to the emerging role of emotions in work life. Both empiri-
cal and theoretical studies continue to emerge in this field. For a review, see Hrtel et al.,
2005.
xxvi. An exception is that of Ashkanasy et al. (2005).
xxvii. See Politser (2008) and Camerer (2007).
xxviii. Hanoch (2002), Maramatzu and Hanoch (2005). In one of their works, Hanoch et al.
(2007), link emotions to bounded rationality and to fast and frugal heuristics and suggest
that emotions are a mechanism that help individuals (older people, in the study) to come
to a decision when information and time are scarce.
xxix. I take the three factors from Kahneman (2003).
xxx. This example is taken from a real case. Please contact me if you want to know more.
xxxi. Gabbay and Woods (2007).
xxxii. The Gang of Eighteen has been introduced by Woods, 2004. Here is a short explana-
tion of the meaning of those fallacies with Latin names that are not considered in the
text. The argumentum ad baculum fallacy refers to the use of force or to the menace of
using external events to persuade one person to agree (e.g., eat this soup or something
bad will happen to you). Baculum means stick or cane. The argumentum ad miseri-
cordiam is the fallacy of bringing emotional factors in praise of the argument (e.g., we
must give him tenure because of his precarious health conditions). The ad ignorantiam
fallacy is the assumption that a proposition is true because we have no proof that it is
wrong and vice versa, a proposition is wrong because we have no evidence that it is true
(e.g., this is typical of statistical procedures when we reject a hypothesis because we
cannot prove that there is a relation). Secundum quid is the generalization of a particular
part of a proposition even when circumstances have changed. It can be thought of as a
particular type of hasty generalization. We have the fallacy post hoc, ergo, propter hoc
when something happens after an event, therefore it must be because of that event (e.g.,
every time it rains in Minneapolis, the Vikings lose the game!). Ignoratio elenchi,
red herring, or straw man occurs when a successful argument is held against ones
opponent, being the opponent not committed to that point. This is only an approximate
explanation; therefore, for a full treatment of fallacies, I recommend reading Woods
(2004) and Gabbay and Woods (2007).
xxxiii. Gabbay and Woods (2007, p. 68).
xxxiv. See Woods (2004) for a complete analysis of the Gang or Gabbay and Woods (2007),
Chapter 1, appendix for a summary.
xxxv. This analogy could be found in Block et al. (2008).
xxxvi. I found this fallacy the core concept of a management book, The Living Firm (Vicari,
1991); it may also fall under this same fallacy when adjectives such as vital are used
to denote firms or organizational characteristics (Golinelli, 2005).
Notes 59

xxxvii. March (1994), Chapters 1 and 2.

xxxviii. Simon (1997).
xxxix. Goldstein and Gigerenzer (1996, p. 651).
xl. Goldstein and Gigerenzer (1996).
xli. Gigerenzer and Brighton (2009).
 Part II
The Extended Brain
Chapter 6
Simons Error

In the previous pages I offered a vision of rationality and decision making based on
bounded rationality and theories, models, and approaches that support its assump-
tions. All of these theories, and the original model of BR itself, explain how we have
limits. Decision making, according to this perspective, is defined through the under-
standing of the limits that characterize our rational world. From this perspective, a
map of all of our bounds is particularly helpful since it defines what limits (more
than improves) our way of thinking.
A theory of bounds is a negative theory in that it defines shortcomings, anoma-
lies, or fallacies of individual cognition. The outcome is that rationality and decision
making are more defined by what cannot be done than by what can be done.
Approaches, biases, heuristics, emotions, etc., are mechanisms that limit what is
intended to be the full potential of a brain.
Bounded rationality is also negative because this is the way it was generated.
Simon thought of it as an alternative to the subjective expected utility (SEU) model
of full rationality,i widely used by economists. If we are not fully rational, we might
be something-related-to-rational instead of irrational. The theory then opted for a
somewhat rational individual, using the SEU model still as a benchmark. Today,
there are papers that try to find room for boundedly rational agents in economic
theoriesii and compare it to the mainstream model of full rationality. However, the
point is that bounded rationality has been defined as a modification of an exist-
ing theory. As shown in the previous chapters, understanding and addressing the
shortcomings of the SEU theory has always been a major concern for students of
bounded rationality. These theories have been successful in their criticisms of the
SEU paradigm.
What is covered in this chapter is based on two different perspectives of ratio-
nality and decision making. First, the focus on the how of rationality is particularly
useful for understanding and defining a context in which individuals make decisions.
These theories show how individuals use bounded rationality to make decisions.
However, the great absentee of this inquiry is the why. Why do people think the way
they do?
I voluntarily put the question in a vague format without including bounded ratio-
nality in it. This is because BR is a hypothesis and I prefer to treat it as such. It
is very difficult to get too far away from the idea that rationality has bounds, but

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 63

DOI 10.1007/978-1-4419-7542-3_6,  C Springer Science+Business Media, LLC 2011
64 6 Simons Error

still I believe that to fully understand the implications of a methodological con-

struct it is better to consider it the way it is: a scientific theory. And scientific
theories can be falsifiediii or supported. Raising the question of why our cognitive
system is the way it is may reveal overlooked sides of our rationality. This is exactly
where I intend to go in this chapter: How can we improve the theory of bounded
rationality? Probably, we should start asking questions that seem to have simple
answers.
Second, I think it is time to get rid of the negative. Putting too much emphasis
on the limits of rationality has prevented us from focusing on the study of poten-
tial of the human brain. Scholars were too involved in studying how the previous
paradigm of the homo economicus could be falsified. Even though the old paradigm
is in more and more trouble, we are still overlooking what we should call the
bright side of rationality.
The chapter offers a summary of the so-called distributed cognition (DC)
approach that has been introduced by Hutchins in 1995 and has improved dramati-
cally since then.iv The ideas that support this approach are provocative and do not
constitute the mainstream in cognitive science. This is to state that what follows here
is highly speculative, but if you follow the hypotheses that I am about to explain
here, you will find a different rationality and a very challenging decision-making
model. This is where you can also find Simons error.

Distributing Cognition
How does our cognition work? This question, or variations of it, has been asked for
thousands of years by western thinkers and philosophers, and many answers have
been given. However, in recent times the answer has started to depart from usual
philosophical frameworks. This is what we are considering here and, of course, it
is well and deeply rooted in the same western philosophical traditions, although
reinterpreted on a completely different basis.v
Richardsonvi writes that there are three groups of theories that analyze intelli-
gence, rationality, and cognition: (1) the computational group of theories includes
thosevii who explain the human brain through the metaphor of the computer, where
the study of symbols and algorithms mimics the functioning of the human and ani-
mal mind; (2) the connectionism movement is concerned with the study of neural
relations, where symbols and algorithms have a secondary role compared to the
way the elementary parts of the brain interactviii ; (3) the third groupix takes both
computational and connectionist approaches one step further to include overlooked
variables, especially those that belong to the social side of human cognition. Studies
from this perspective put the human brain on an evolutionary track and explain that
huge cognitive developments happen due to the web of social interactions that shape
the life of human beings. Both the environment and human relations become part of
the cognitive equation.
Distributing Cognition 65

The goal of theories from these three groups is to understand and analyze how
the brain works. However, what all three perspectives share is the axiom that cogni-
tion is located inside the brain. In a recent work with Emanuele Bardone, we look at
recent trends in cognitive science and suggest that there is a fourth emerging cate-
gory that consistently increases the explanatory power of previous theories.x Once
again, these theories take one step farther from previous ones and consider some of
the distributed nature of cognitive resources. The resulting picture sees human cog-
nitive processes shaped by external resources.xi Within this framework, the social
environment provides a significant source for cognitive materials, although it is only
one among many other sources. This idea of a distributed cognition (DC) approach
can be related to a sort of ecology of the mind,xii but I maintain that there are sig-
nificant differences between this and several other studies on the human brain and
on rationality.
The following pages are dedicated to what I believe constitutes the newness of
this approach, and explore the extent to which it successfully explains the individual
cognitive system in a way that seems to be particularly helpful for organizational
studies.

An Irreverent Hypothesis on Cognition

After the trauma of birth, our mother and father, together with people in the childcare
unit of the hospital operate to further our knowledge and understanding of the world.
In a few days everything will be different forever. You are not able to remember
when you spoke your first word and you dont know what that word was. For most
of us it is mom or maybe just moo. . . That sound or, better, the recognition
of that sound by somebody who was close to us in that moment started a process
that our minds will never get rid of. We started to understand that some sounds
are different from others. Up to that point we only heard strange sounds from our
parents, relatives, and close friends. But the fact that the child can interact using
the same sounds is amazing! The brain starts a process it is set up for. We have the
largest brains compared to other primates and to the relative size of our body, and
we need more activity to cope with our social environment.xiii Size and activity are
related to the fact that we are social beings.
Size is not all.xiv The interesting fact is that when we realize that those sounds
are so important, we start to learn more of them. At the beginning our vocabulary
is pretty poor and our use of grammar and syntax is awful, but we get better every
day, and fast up to a point where we do not understand (or remember) how life was
without words. Was it possible? Words, our language,xv become part of us so that
we think we have always had them. We cannot live without words, we cannot think
without them. Now, lets run a mental experiment. Imagine that you have to think
of something without words. What does your thinking look like? How is it formed?
By what is it defined? How can you express what you feel, need, or want? It isnt
easy. The point is that language defines you and the way you think, and this started
66 6 Simons Error

at a specific moment of your life. Where does language (or the words you have been
taught) end and start your real rationality? Is language something really yours? Or,
has it been imported from somebody outside of you? Your language was external
when you were a baby. You, and your parents, made it internal. But where is the
internal divided from the external?
Here is the idea. When you are learning words, your cognition is defined by this
process. Learning your first words depends on the meaning you attribute to them,
on reactions that you see on people around you, on the sound of your voice that
you hear for the first time that way, from the meaning that is associated with the
word you pronounce. Not only is your cognition shaped by this, but your behavior
will also be soon affected. As a child you learn that there are words associated with
specific actions so that, for example, eat or hungry means that you are about
to be fed. Your cognition defines itself on the basis of these mental and behavioral
processes. Again, where is the distinction between internal and external? When do
words start to become internal facts of your cognition? Especially when they are
associated with action, do they become internal?
Child behavior and mental development are very peculiar, and I dont intend to
be so naive as to pretend that those processes can be associated or assimilated with
those of adults. However, language provides a very interesting tool of analysis for
adult minds too. When we speak, we are using something that is supposed to lie
inside of us and we put it outside. Now, I would like to consider another example,
an extension in the use of language. Everyone who has tried to write something
knows very well that many of the ideas and the exact shape of your thinking come
together with the act of writing. However, if your ideas come out when you write
them, are they really inside your brain? Once again the same question: Where do
they lie?
The idea of rationality that Simon and others use lies on a separation between
the brain and what stays outside of it so that, for example, individual capabilities are
separated from external information.xvi Data stays outside while abilities stay inside.
The assumption that stays at the basis of the distributed cognition approach elimi-
nates this divide. Internal and external cognitive resources stay together in a smart
interplay,xvii where cognition is dependent and shaped by these set of relations. A
child that learns the first words and an adult that writes some text are common exam-
ples of how distributed cognition works. The written word is not inside the brain, but
continues to play with it; there is a continuous shaping activity that goes on between
internal and external.xviii The cognitive process includes the written words in such
a way that it is very difficult to define where and how this external tool can be con-
sidered only an outside source. What we know is that it is a resource for the brain
and, in many cases, cognitive processes change depending on the kind of resources
we use. This is the reason why spoken words are different from written words, and
the same written or spoken word has a different meaning in the cognitive process it
enhances.
This point can be framed with what we have learned from the previous chapters.
Data representation allows individuals to give more or less sense to a given problem.
However, what proponents of the distributed cognition argue is that this is only part
Distributing Cognition 67

of the process. It is not that the brain, with its well-defined mental abilities, makes
sense of something, but that it is shaped by what stays outside of it.xix The cognitive
process changes and sets different limits depending on the kind of external resource
(a single word in our example) and the medium used to transfer that information
(e.g., spoken or written resource). Our ability is not given a priori then, but it modi-
fies together with the interaction that we have with the external resource. Our mind
has plasticityxx because it is flexible and adaptable.
This interaction between internal and external resources can be observed in many
processes that define our everyday life. Anything that includes resources that are
located outside of our brains can be analyzed through the distributed cognition
approach. When you speak, write, or perform an action, your brain is shaped by
these processes. Nevertheless, if I accept words such as to shape, define, or
structure, I face the risk of falling under the same old internal-external dichotomy.
I believe that we need to consider what the distributed cognition approach could con-
tribute on a wider basis. Neural processes are affected by external resources in the
sense that synapses (the way neurons connect to each other) start to emerge only
when this external resource is considered.xxi Hence, there is no (or limited, or dif-
ferent) cognitive activity without external resources interacting with us. Even when
we dream we picture these interactions. This means that the cognitive process is
activated, maintained, or lost depending on this web of internal-external relations.
This is what students of DC and I mean with verbs such as to shape or to define.
The described process is so powerful that even if there is no actual handling of
or contact with external resources, our synapses start to function. The so-called mir-
ror neuronsxxii are supposed to mimic the actual activities that we perform, and in
doing that they mimic the interaction between the brain and the external resources.
For example, when we watch a basketball player, an ice skater, or a dancer that
executes an incredible performance, our brains mimic that person, activating areas
that are usually associated with movement. This is, neurologists tell us, the reason
why we enjoy sports, music, films, and all other apparently brain-based activities. In
short, we see ourselves playing, skating, dancing just the same as people that we are
only watching. Isnt it curious that to enjoy something, our brain needs to function
like we are actually doing it? Using our vocabulary, the brain mimics an active dis-
tributed cognitive process.xxiii The point is that this looks like a distribution within
the distribution since the player, the skater, and the dancer are external resources to
our brain (distribution #1), and we make sense of them using a defined schema that
sees us doing the same thing (distribution #2).
One of the corollaries of this approach is that individuals perform differently
when they exploit diverse external resources. Suppose that somebody asks you to
draw and analyze a mathematical equation. You have two options: (a) use paper
and pencil or (b) use a computer. If you choose option (a) you will probably need
more time and, depending on your math skills, you will need to apply rules to study
the function and get the solution; otherwise it shall take forever. Choosing option
(b) is easier, it usually takes less time, and you can use software that helps you
in the analysis (and I am not referring to the option to Google it). However, this
works only if you know how to let a computer study an equation and solve the
68 6 Simons Error

problem for you. Otherwise, learning new software could be more painful than the
paper-and-pencil option. What counts most is that independent of what the choice
is, the cognitive processes associated with each option are different. Also, the two
activities bring together mental patterns so different that the ideas and analyses they
can suggest may be different. Ideas on what is most interesting in the function may
vary together with the outcome of the study. Of course, a computer may be more
accurate, while the drawing may suggest (stress) areas of interest in the function.
Again, results have the potential to be different. The way cognition distributes these
external resources defines the process.
This also suggests that there is something that the previous model is not
able to seriously take into consideration: the quality of external resources.xxiv
Bounded rationality models are concerned with the quantity of information that
can be handled. Those who attempt to map its bounds also fall under the same
category.xxv

Boundaries of the Mind: The Through Doing Logic

In real life, the choice between the two optionspaper and pencil or computer
does not happen very often. People do not choose between painful or less painful
approaches to a problem. They usually dont like painful processes at all, unless they
are forced or are masochists, so the choice goes in the preferred direction first. Let
us hypothesize that the first medium chosen (external resource) is the computer. You
start with your study function using the computer and draw the equation through a
dedicated software. The answer to the question is immediate, but soon after that you
try to study the equation, setting its maxima and minima. You need a couple of math
passages and the computer wont help. But, you want to go further with that: What
do you do? Well, one should take paper and pencil and start with a partial derivative
analysis. A few passages may be sufficient, but while you write a new world is open
to you, so that you start this strange interaction between paper, computer, and your
brain: What happens if I change this coefficient so that it equals 1, and what if it
equals 0? And what if you transform the equation into two new ones, where the old
one is just the sum of these two? Does this help your study? While you work on the
paper, you check passages on the computer software to make sure you are using the
math correctly. This is a work in progress.
As you can see, in this example you are not using one single source as an aid to
your work, you are using multiple sources at one time and modifying your cogni-
tion constantly. The point is that you are definingand redefiningyour cognitive
processes through doing.xxvi This brings us to at least two implications.
Implication 1. Your rationality is not well-defined in a specific moment in time,
and neither are your capabilities. They change according to what you do; they are
dynamic. A rational decision maker is one who adapts the cognitive processes at a
fast pace. A rigid one-resource person has more limits that one who jumps from
one to another. Yes, to solve a mathematical problem with paper and pencil is
Distributing Cognition 69

more complicated, and it looks like it indicates your potential. But, what about
solving a mathematical problem with no external resources at all? No speech, no
paper, no computer? Brain activity only. That might probably be a true muscle
showing. However, things change when you have, as usual, access to multiple exter-
nal resources. For example, suppose that (a) you have access to a computer with
(b) an internet connection. Moreover, imagine that (c) a mathematician friend of
yours is available and willing to help you, and (d) a specific software that analyzes
mathematical problems has been installed on your computer. Now, what is the real
muscle showing when its time to evaluate if you are being rational? Is it rational
to do without the external resources available? Or, what muscle are you showing?
Consider this second task: A manager of an important software company has
been asked to make a decision on a new product. Financial and marketing managers
think that the software can be sold as it is, since they know the needs of the com-
pany and that eventual bugs can be adjusted by offering free updates to customers.
Software developers and engineers keep saying that there are too many bugs and
that the company cannot afford to sell a defective product. The decision here comes
out of multiple and contrasting resources. When the manager faces the alternative
(if it exists) to make a decision with or without an appropriate consideration of these
resources/opinions, he is interacting with these resources anyway. But let us take the
first option. What if the manager goes for it: What should he expect? The point is
that, even before the scenario-drawing activity, the manager knows that whatever
decision is made will be redefined and fine-tuned once made. It is not new that,
especially in uncertain and complex environments, managers rarely make decisions
once and for all. Part of the usual cycle of any strategic decision-making hand-
book is double-checking that the action remains in line with what is planned. These
feedbacksxxvii are (or should be) the essence of organizational decision making.
They suggest that some decisions are interactive decisions, in that they change
and develop together with their action, i.e., they develop through doing. The deci-
sion to come out with an unfinished product needs further decisions. It is like in a
decision tree when you know that option A implies a second and maybe a third
subsequent choice. In this case, the option implies corollary decisions that could end
up modifying or changing or reverting, in the case of the recall, the original deci-
sion. The decision is taken with a high degree of uncertainty, where the manager is
not willing to take any definitive action; that of version waves of the same soft-
ware product seems to be one of the main strategies in that industry.xxviii Basically
the software company starts selling the product while engineers and developers con-
tinue working on it. As soon as a new and improved version is ready, they sell it to
the market without giving adequate communication so that the product is perceived
as the same, even if it is not. When customers have problems with the first, the com-
pany can give them the new version for free or sell it as a software update. Another
strategy is to push the innovative side of the company, having the product sold with
progressive numbers where version 1.0 is usually the first, followed by 1.1, 1.2 or
1.01, 1.02. It is apparent that whatever the choice, the decision-making process hap-
pens in a time continuum, and that organizations (and their products) provide the
basis for this continuity.
70 6 Simons Error

The second option, that of not selling the product, leads also to a through doing
logic. The fact that the manager needs to routinely check developments of quality
tests and software debugging is something that may lead him/her to a different deci-
sion day by day as well as to reinforce the initial decision. This process is based on
the influence that external resources have on the manager, and the process can move
these bounds far away from where they started. The ability to see a solution where
there are none is typical of this function of our intellect.xxix We can think of all of
these decisions as separate from the other but this is just a thought experiment that
we can run for educational purposes since we know very well that the continuity
of organizational processes provides a significant help to managers (and all of the
participants). This way managers can track their (and others) decisions where the
next step is related to the previous ones. It is this way that people extend their limits
and their rationality. Isnt it true that we learn constantly? Isnt it true that learning
implies the fact that we can move our rational bounds?
Implication 2. We learn from our mistakes as well as try to modify our behav-
ior according to past experiences and, sometimes, creative and innovative thinking.
These factors set variable boundaries to our minds. It is true that sometimes individ-
uals are not successful in overcoming their mistakes but they do try to do that. The
processes that lead to the through doing or working decision making follow this
pattern. Managers, like every human being, make bad decisions sometimes, this is
undeniable. Does this happen when they do not use their distributive cognition prop-
erly? Does this happen when they fail to exploit external resources? When they fail
to learn? Or, is it just the nature of their rational bounds? More analysis is needed
here.
This is just to say that sometimes the boundaries of the mind are the ones that we
put on it not the ones that it has. At the very end, the distributed cognition approach
is something that makes this point apparent. The richness of a cognitive process lies
on its interactions with multiple external resources at one time instead of believing
that rational muscles are fixed and stable or that they can be defined by studying
individuals that perform one simple task at a time.xxx Now, the question is this:
Are organizations places where people perform one single task at a time or we are
moving far away from this Taylorist conception?

Externalizing and Reprojecting

We have seen that the divide between internal and external resources is not
necessarynot even particularly helpfulto understand why our cognition works
the way it does. We need to find a different explanation. We have also stressed that
interactions (or interplay) between internal and external is worth studying. How
does this work? Which are the components and determinants of the distribution
process?
Individuals have two ways to exploit external resources. The first is a process
known as externalization.xxxi This is the tendency to disembodyxxxii part of ones
Distributing Cognition 71

cognition to resources other than the brain.xxxiii For example, listening to the music,
reading a book, writing, singing, or solving a logical problem come from cognitive
processes directed toward making sense of our reality.
Externalization can be active or passive. This is one of the most important dis-
tinctions to notice since it allows us to separate acts of will (voluntary behaviors)
from involuntary behaviors. The latter includes making sense of music, for exam-
ple. The music is something we listen to and it appears that processing information
from the outside is internally based; however if we indulge on how this happens,
we find that interaction stays at its basis. We need to make sense of this external
activity (the sound) located outside. And we do it and process it with emotional
reactions, among others. It is counterintuitive but this is what stays at the basis
of the DC approach: Everything is not located here or there, it is the interaction
that explains the cognition behind it. If we are musicians, then our externalization
becomes a mix of active and passive. It becomes active when we start writing music
or playing an instrument, i.e., when it is clear that we are putting things outside our
internally defined brain. The idea that the mind needs something outside to work
properly is apparent to anthropologists that explain our ancestors mural paintings
as a big jump ahead in the evolution of the brain.xxxiv If this could be thought of
as a first example of distributed cognition I leave it to experts but, the fact that
there is a tendency to externalize is out there. The list of examples can be end-
less and facts of organizational life fall into this endless category. Financial, annual,
social, environmental reports are a well-defined and structured way to externalize:
meetings, standard operating procedures, contracts, negotiation and arbitration pro-
cedures, as also hiring or firing procedures, routines, roles expectations, and the like.
Moreover, corporate culturexxxv can be included in these organizational externaliza-
tions since it is defined by behaviors that emerge from a group of people that take
part in organizational activities.
This process comes with another one that is not less important. What happens
when people engage in an active externalization? What happens when the above
mentioned musician writes the music or plays an instrument? This person is doing
two things at the same time: She is disembodying part of her knowledge over an
outside tool and making sense of it constantly. This means that once this music
is externalized on a piece of paper, for example, the process is not over yet. The
cognition continues to lean on what comes to be an external resource now and this
fact changes the perspective together with the process that is ongoing. That note is
not the abstract idea that the musician had in her mind anymore, now it is something
written on a piece of paper that makes sense together with its context (other notes)
or that assumes a different meaning when it is there. It could be that the single
note sounds better in a different context or that this is no more what the musician
intends to do with that music. In other words, the externalization leads to a re-
projecting activity.xxxvi This is essentially what happens every time we externalize
and every time we subsequently exploit an external resource. The cognitive process
is made of the combination of these two processes. Again, where the external and
the internal part end and/or start is not relevant; the interaction is what cognition is
all about.
72 6 Simons Error

We can read the above-mentioned distinction stating that representations are

external, as they are described by the externalization process, and they are inter-
nal, as they appear in the reprojecting phase, but we may fall into the same old
cognitive divide. An example may help clarify these points.
How is the annual social report of a company created? Companies define and
report their socially responsible behaviors in so-called social reports. There are
many guidelines on how to write effective social reportsxxxvii , and it is not the
purpose of this book to present or analyze them. However, there is usually an organi-
zational motto or mission with colorful (and reassuring or careful) pictures that
leads to a table of contents. After this, there is a letter from the CEO, the President,
or the Chairman of the Board of Directors. In the next pages, the report starts with all
of its sections. The distributed cognition works from the side of the reader, as well as
from that of the writers. The letter from the chairman, for example, has been drafted
many times and it has been modified from that written by the original writer and
many other individuals from her team, as well as PR specialists. Sometimes, external
service companies are used to write and contribute to the writing of the report. These
function as another source of fine-tuning of the original message. All of these activ-
ities of writing, reading, changing, rewriting, etc., are externalizing and reprojecting
processes that shape the cognition of individuals involved and the decision-making
process too. Without these processes, there will probably be no letter from the chair-
man. I know that some of you are thinking that this should probably be a better
outcome, but lets stay focused on the DC process. First, when the chairman writes
the message, he experiences externalizing and reprojecting activities together. The
choice of words may come from a stream of consciousness at first, but it can also
be a very complex activity made of writing-reading-rereading-changing-writing-etc.
When the chairman puts words into word-processing software, she is externalizing;
when she is evaluating, thinking of, and choosing words, the chairman is repro-
jecting. As you can see, these two processes are intertwined, one affects the other.
All other processes are made of passive externalizations (other people that read the
draft) and reprojecting activities. When we read a word, we make sense of it depend-
ing on the fact that we have that word in our memoryxxxviii and on the basis of the
way it relates to the other words in the sentence. This is a very common and widely
studied phenomenon. Take the sentence she is very eclectic since she is interested
in many different activities. The sentence can be understood even if you dont know
what the word eclectic means because the other words suggest a meaning for it.
The reprojecting happens all the time, and we have to make sense of things. The fact
that people read the report means that they have to attribute to the signs that form
the message a meaning that may be different from what they have learned before.
Moreover, when reading something from a given source, e.g., when you know that
you are reading something from your boss, it also implies that you frame the role,
ideas of that person, beliefs, and all that you think of that person and of the situation
you are facing (logic of appropriateness). Therefore the reprojecting may go well
beyond the simple word-related meaning of words. This last point, for example, is
particularly meaningful if we look at the reader of the report that is not part of the
organization. The process this individual goes through is similar to that of an insider,
Distributing Cognition 73

except that she/he is reading an externalization that she/he perceives as being related
to the company as a whole.
How far can we go with these concepts when we analyze decision making in
organizations? What can we get from the use of a DC framework to organizational
decision-making processes? The next two chapters (especially Chapter 8) offer some
analytical hypotheses.

The Extended Mind

What I have tried to present here is apparent in the work of many students of the
distributed cognition approach. In summary, the most relevant points that emerge
from this approach are different from the old paradigm on which bounded rationality
is based. Some of the achievements are that (a) the divide between external and
internal cognitive resources has a limited explanatory power, and (b) we can have a
better understanding of human cognition if we abandon that paradigm.
The idea is that to understand cognition, we need not focus only on its internal
bounds. As Wilson suggests,
Accepting the extended mind thesis means holding that the mind is not physically bounded
by the body but extends into the environment of the organism. [. . .] The extended mind
version of externalism represents a stronger and more striking view of the mind than do
earlier forms of externalism. It is a view that embraces the claim that technological and
cultural artifacts may be physically constitutive of cognition.xxxix

This is exactly the take I have supported in this chapter: Our mind extends to external
aids that define and shape cognition. To this respect, the mind is extended. And
this is the reason why that in order to understand cognitionand rationality as part
of itwe must take into consideration all the constituents of our cognitive activity.
These constituents are not only internal, nor are they entirely separable from the
outside variables. In a part of his work dedicated to learning, Hutchins writes that
[t]his heavy interaction of internal and external structure suggests that the boundary between
inside and outside, or between individual and the context, should be softened.xl

This is very close to the idea of the soft self that Clarkxli explains in his work:
There is no self, if by self we mean some central cognitive essence that makes me who
and what I am. In its place there is just the soft self: a rough-and-tumble, control-sharing
coalition of processessome neural, some bodily, some technologicaland an ongoing
drive to tell a story, to paint a picture in which I am the central player.xlii

The DC approach is not a complete break with past tradition of thoughts. Wilsons
discussion of collective memory is particularly helpful:
For between the individualistic approach and the collectivist approach are extended mind
views that, in some sense, borrow from both. From individualistic approaches, they accept
that remembering is an activity that is done by individuals, and from collectivist approaches
they take the idea that this activity is not bounded by what goes on in the head of the
individual, and so encompasses commemorative objects and practices, mnemonic devices
and strategies, external symbols and structures.xliii
74 6 Simons Error

We can go on with many other examples of what I have tried to evoke in this chapter:
Bounds are not what bounded rationality students think they are. For the economy
of the discourse, I believe I should stop here with examples from the literature and
move forward. Here comes the need to spend a few words on bounded rationality.

How Bounded Is Rationality?

Now, this is the question: How bounded is rationality? What the distributed cog-
nition approach suggests is that a theory of the limits of our cognitive activities is
helpful, but it falls short when it has to define the dynamic and plastic functions of
human intellect. Do we solve problems within our rational bounds, or do we solve
them because we are able to move these bounds? Do we make decisions depend-
ing on the way we are bounded, or do we make decisions based on a variation of
these bounds? I believe these are not trivial questions and they need to be addressed.
The first question is about successful decision making (problem solving), while the
second is broader since it includes all kinds of decisions.

Cognition and Rationality

Before addressing these questions, there is one point that needs to be considered.
What is the relation between cognition and rationality? Until now I have used these
two as synonyms with a certain degree of vagueness. Although there is a common
ground between the two words, they maintain their differences. While cognition
refers to every mental process,xliv rationality gets involved when these processes
follow sound and consistent procedures and have effective outcomes (see previous
chapters). It is clear that rationality is a specific type of cognitive process. However,
the study of rationality in the last decades has emphasized irrational behavior and
flawed decision making.xlv Moreover, many studies on cognition and decision mak-
ing have appeared to explain and make sense of rationality. Right now, rationality is
grounded in cognitive science for decision-making students, and different meanings
of it have started to appear and be used by the scientific community. The inter-
changeability of rationality and cognition misrepresents the complexity of the issue,
but, since there is no understanding of rationality without cognition, I believe that we
need to define which cognitive hypotheses stay at the basis of any rational thought.
This short note is useful for providing guidance in the use of the terms and in fact
the answer to the question that entitles this section can be answered only if we look
at cognitive theories. And the answer is clear, if DC is preferred to other approaches.

Problem Solving
The starting point is the first, which is the easiest question: Do we solve prob-
lems within our rational bounds, or do we solve them because we are able to
move these bounds? Bounds to human cognition depend on how individuals exploit
external resources. One point here may be of stating that this is still a problem
How Bounded Is Rationality? 75

that pertains to individual cognitive capabilities. I am not arguing, nor are scholars
and proponents of the DC, that all human beings have equal potential or the same
capabilities. Individuals are different and they sometimes react differently to similar
threats or problems; however, this is but a secondary point. This is because every-
body is different, their minds work in a slightly different way so that an external
resource can affect their cognition differently. This deals in part with the representa-
tion, accessibility, and framing processes. Why do they matter? They are important
because of the way cognition interacts with external resources. This is not an effort
to falsify previous attempts to make sense of rationality; the aim is to improve and
redefine them on the basis of what I believe is a more accurate and broader approach
to the problem at hand.
The externalization process explains how bounds move. When a manager finds
a solution to a problem, what are the factors most important to his or her cogni-
tions? They are related to how the manager has been successful in moving his or
her bounds, how he or she overcame the limits that impaired his or her decision-
making process. These were limits of cognition that could move when the manager
uses an externalization strategy (without even knowing what he or she is doing from
the cognitive perspective). Of course, not every externalization process leads to a
successful decision. Externalizations describe the ways our cognition works so that
it shows how dynamic the process is and not how limited it is instead. Every suc-
cessful decision passes through a process of externalization and reprojecting where
interplay of resources causes our cognition to work more smoothly. Therefore, the
answer to the question cannot be a straightforward statement where we deny the
first part of the question and affirm the second. Human rationality is bounded, but
these bounds move consistently with the nature of the problem and with the type
of resources that are exploited. Are we still bounded then? Yes. Do we move these
bounds? Yes. The point is that the way rational bounds move redefines cognition
and rationality. Its plasticity and dynamism should be used to define human prob-
lem solving. Think of the way individuals adapt themselves (reset their limits) when
asked to make decisions and solve problems in different social environments (e.g.,
office, family, church), or when different behaviors are in play (e.g., shopping, play-
ing sports, studying). The individual is the same, but his cognition varies. Different
problems call into play diverse bounds and cognitive processes, and the same prob-
lem may call for that bound adapting to the solution. What happens when you solve a
problem because of new data found online, of advice from a colleague, of a reading?
That impression that you are connecting the dots, that something is materializing
in your brain, is the way our cognition works: eppur si muove!xlvi

Moving Bounds
The second question is more general and includes all decision-making prob-
lems: Do we make decisions depending on the way we are bounded, or do we
make decisions based on a variation of these bounds? Sometimes failure to solve
problems can be described as failure to adequately exploit external resources or
the lack of enough resources to exploit. If the understanding and analysis of a
76 6 Simons Error

problem depend on the three above mentioned activitiesrepresentation, accessi-

bility, and framingthen individuals (and managers) should look more at access to
information than to have more. The same piece of information should mean different
things to different people, and sometimes two pieces of information mean the same
to different individuals. This means that the external resource is what matters most
in this analysis. People could fail in their decision making (i.e., make bad decisions)
precisely because there is a failure in this exploitation process.
The idea is that bounds move with the cognitive process. And if bounds are
not stable and they are not the point in describing rationality, what is the concept
of bounded rationality then? The distributed cognition provides us with an inno-
vative take on decision making and rationality; the next chapter is an attempt to
define how the concept of rationality changes as far as this approach is concerned.
Before that, we still need to unveil Simons error, although it should be apparent at
this point.

What Error?
I do not pretend that my review of the DC framework is exhaustive, nor do I believe
that this chapter presents the distributed cognition approach. I have tried to show
some of the most relevant concepts that many of the DC approaches share, with
an emphasis on those ideas that seem more significant to the discourse of rational-
ity and to their transfer to organizational behavior. What this chapter presents can
hardly be defined as mainstream cognitive science, but ideas described here capture
something of what seem to be an incoming trend in that field, as of today.
I believe that the reader now has an idea of what I mean by Simons error. And
it is the stable divide between internal and external cognitive resources that bounded
rationality shares with neoclassical economics (and expected utility theory). In the
face of distributed cognition, this divide brings the theory to a critical point that has
not been considered by recent BR developments and studies.
Truth to be told, now that you have an overview of DC hypotheses, we can
consider more than one error that can be associated with the idea of bounded ratio-
nality. Some of them have been implicitly or explicitly considered in this chapter,
while others have not. These errorsit would be better to abandon the provocative
word errors and switch to critiques or challengescan be classified into three
categories:

1. Neoclassical flavor. The idea of BR shares some of its assumptions with the
theory Simon wanted to criticize. In particular, it maintains a strict relation-
ship between means and ends (teleology),xlvii it uses a brute-force computational
strategy to mimic the human brain,xlviii and, as we have seen, it does not abandon
the cognitive divide.
2. Computer metaphor of the brain. BR is based on Simons belief that the human
brain works like a computer so that symbols and serial processing of data con-
stitute what needs to be studied.xlix This has also been called Simons paradox:
Notes 77

the extreme high esteem of computer powers and the lack of belief in human
capabilities.l
3. In vacuum. The boundedly rational agent behaves in a social and resource
vacuum.li This means that the decision maker operates in an environment
where social, cultural, political, technological, or demographic variables are not
relevant to the efficiency and effectiveness of his/her decision.

There are many corollaries that stem from these three assumptions. As far as
what is presented in this chapter is concerned, the most significant corollaries are
that (a) information is only quantitatively relevant to the decision maker (i.e., qual-
ity, sources, mediums, content are not relevant), and (b) individuals have stable
cognitive capabilities and preferences over time.
The overwhelming number of studies that map BR takes most of these assump-
tions for granted. It is very unusual to find somebody, among these scholars, that
questions or criticizes some or even one single assumption on which BR is founded.
In the next chapters I suggest how the DC approach urges to start a discussion
on these assumptions that may eventually lead to corrections and to a remodeling
of BR.

Summary
In this chapter I have presented the distributed cognition approach. This theory is
based on the divide removal, i.e., internal and external resources are in constant
interplay, and cognition is defined and shaped by these activities. Boundaries of
the mind are then redefined along with the two representation processes: repro-
jecting and externalizing. The result is an extended mind that challenges several
of the assumptions on which bounded rationality is based. The final part of the
chapter offers the moving bounds hypothesis and presents an overview of Simons
error(s).

Notes
i. This is clear in Simon (1979), where he compares bounded to full rationality; see also
Simon (1997).
ii. This approach to bounded rationality is very diffuse among students of behavioral
economics. It seems that Simon inaugurated a trend that suggests to scholars how to
introduce this idea. Or, it is more likely that these scholars tend to be soft and pay
a tribute to mainstream economics, decision making, organizational behavior, etc. See
for example, Aumann (1997), Camerer (1998, 2007), Conlisk (1996), Foss (2003b),
Gigerenzer et al. (1999), Grne-Yanoff (2007), Hanoch (2002), Kahneman and Tversky
(1979), Klaes and Sent (2005), Knudsen and Levinthal (2007), March and Shapira
(1987), Mumby and Putnam (1992), Patokorpi (2008), Rothschild (2001), Rubinstein
(1998), Selten (1998), Shakun (2001), Todd and Gigerenzer (2003), and Zafirovski
(1999).
iii. Yes, I am referring to what you are thinking: Popper (1935/2002).
78 6 Simons Error

iv. Besides Hutchins (1995), there are several works that offer a framework of what the
distributed cognition approach is; among many others, a starting point could be found
in Clark (2003, 2008), Clark and Chalmers (1998), Thagard (2007), Magnani (2007)
and Magnani et al. (1999).
v. I cover this domain in Bardone and Secchi (2009) and Secchi and Bardone (2009b).
vi. Richardson (2000).
vii. von Neuman and Morgenstern (1944) and Simon (1947/1997, 1955).
viii. This may derive from the interaction between external variables and individual capa-
bilities, so that cognition can be described in terms of some interactive process. This
approach is commonly used in organizational theory; see Beach (1998), Laroche (1995),
Starbuck (1983), and Walsh (1995).
ix. Axelrod (1976) and Cannon-Bowers and Salas (2001).
x. Bardone and Secchi (2009, p. 184).
xi. Clark (2003), Clark and Chalmers (1998), Hutchins (1995), Magnani (2001, 2007),
Norman (1993), Perry (2003) and Wilson (1994, 2004). The distributed cognition (DC)
approach has its enthusiastic fans and it has, of course, its opponents. Among the oth-
ers, I have found arguments by Adams and Aizawa (2008) particularly detailed and I
suggest reading Harris (2004), for a different take on the topic language and DC, also
discussed below in the text. Andy Clark, one of the founding fathers of this approach,
also offers some critical remarks in the last chapter of his book, Natural-Born Cyborg
(2004). A good theorist always tries to find limits in the theory; and Clark undoubtedly
is an excellent scientist.
xii. This approach to human cognition takes a step further to the ecological approach of
the mind as it is presented by Gigerenzer et al. (1999) and Todd and Gigerenzer (2003),
among many other writings of these authors.
xiii. Gazzaniga (2008).
xiv. The social brain hypothesis by Dunbar (1998), explains this point. See also Dunbar and
Shultz (2007).
xv. Better examples on how language could be related to distributed cognition are in Love
(2004), and Sutton (2004).
xvi. This stays within the definition of bounded rationality (see Chapter 3) and it is not
questioned by all of those that have provided maps rationality (e.g., Goldstein and
Gigerenzer, 1996; Gigerenzer and Brighton, 2009; Kahneman, 2003).
xvii. Clark and Chalmers (1998).
xviii. Hutchins (1995, p. 287f).
xix. Clark and Chalmers (1998) and Magnani (2007).
xx. For details on plasticity, see Clark (2003).
xxi. The central chapters of Hutchins (1995), can be read under this assumption.
xxii. This is explained in Gazzaniga (2008), Chapter 3.
xxiii. See Magnani (2007).
xxiv. Mousavi and Garrison (1992).
xxv. There is no explicit discussion of the quality/quantity dichotomy in BR studies. Mousavi
and Garrisons (1992) critique holds.
xxvi. An exemplification of thought doing rationality in the case of morality is provided by
Magnani (2007); see also Magnani et al. (2006), on the same topic.
xxvii. See March (1994), end of chapter 1.
xxviii. I dont know if this strategy is deliberate so that my reading of it is a guess on what is
available through the media. I am using abduction here!
xxix. It is a typical mental leap, as the ones described by Holyoak and Thagard (1995).
xxx. This is also the meaning of Hutchins (1995) in the wild.
xxxi. Wilson (2005), Clark and Chalmers (1998) and Zhang (1997).
xxxii. However, as Gibbs (2005), shows in his work, this remains a tendency, an attempt
only, since all cognitive activities are embodied.
xxxiii. See Magnani (2006).
Notes 79

xxxiv. Or that explain how artifacts such as a statue half-man/half-lion could possibly exist.
That mythological figure has no room inside or outside the brain, it is the interplay
between the two that made it possible. See Magnani (2006), and Mithen (1999).
xxxv. Hutchins Cognition in the Wild is a study of the organizational culture of the ship Palau.
xxxvi. In one of his writings, Magnani (2006) makes this point clear: I maintain that rep-
resentations are external and internal. We can say that (a) external representations are
formed by external materials that express (through reification) concepts and problems
that do not have a natural home in the brain; (b) internalized representations are internal
re-projections, a kind of recapitulations, (learning) of external representations in terms
of neural patterns of activation in the brain. They can sometimes be internally manip-
ulated like external objects and can originate new internal reconstructed representations
through the neural activity of transformation and integration (italics in the original text,
p. 350). See also Magnani (2007).
xxxvii. For example, some of these guidelines are the SA8000, Global Reporting Initiative,
AA1000.
xxxviii. Otherwise, we use yet another external resource: a dictionary.
xxxix. Wilson (2005, p. 230).
xl. Hutchins (1995, p. 288).
xli. Clark (2003, p. 130f).
xlii. Clark (2003, p. 138).
xliii. Wilson (2005, pp. 231232).
xliv. Thagard (2007).
xlv. It is the work by Kahneman et al., (1990; 1991) that I am referring to here. Ariely (2008)
offers an overview of behavioral studies on irrational behavior.
xlvi. Translation: . . .nonetheless, it moves! Statement that Galileo Galilei is supposed to
have pronounced after the trial held by the Inquisition when he was forced to deny that
the Earth moves around the Sun.
xlvii. This aspect has been analyzed by Mousavi and Garrison (1992).
xlviii. Zelen (2001).
xlix. In his own words, thinking is governed by programs that organize myriads of simple
information processesor symbolic manipulating processes if you likeinto orderly,
complex sequences that are responsive to and adaptive to the task environment and
the clues that are extracted from that environment as sequences unfold (Simon, 1960,
p. 81). See also Simon (1993a), and some works that highlight these points, e.g., Sent
(1997) and Patokorpi (2008). It is also interesting to notice that this second error (com-
puter metaphor) has strong relations with the first (neoclassical flavor). As some authors
put it, in so discrediting economic rationality, Simon nonetheless remained true to the
broader but no less conventional notion of what might be labelled cerebral rationality,
that decision making is a cognitive process that can be decomposed into a sequence of
simple, programmed steps (Langley et al., 1995, p. 262).
l. Zelen (2001).
li. Simons psychology of economic rationality not only assumes a socially decontextual-
ized mind whose primary function is to carry out Turing machine calculations, he also
assumes a mind that is decontextualized from the material conditions of the environment
involved in the agents habits and impulses that give substance to economic need and
desire (Mousavi and Garrison, 1992, p. 152). See also Langley et al. (1995).
Chapter 7
Stretching the Bounds (I)

What remains of Simons idea of rationality after the distributed cognition (DC)
approach is considered? The DC approach points out that one of the most important
activities of human cognition is that it is shaped by what is found in the environment,
i.e., external resources or artifacts. What is or is not rational assumes a completely
different meaning within this framework. The idea that the brain works in isolation
and that information is taken from the outside, then computed, and subsequently
expelled as a solution to a given problem largely overlooks what happens during
the simplest cognitive process. One thing is clear: If we accept the DC framework,
we can no longer rely on this input-process-outputi idea of rationality and decision
making.
First, there is no simple input since every external resource is part of the process
and second, the output is a process. Externalizing and reprojecting are aimed at
explaining that there is no divide between these stages in the cognitive process.
The idea of bounded rationality (BR) is based on this input-process-output as it
appears clearly from the first chapters of this book and from the last paragraph of
Chapter 6. While studies and empirical assessments of this concept have emerged in
the last few decades, I agree with Nicolaiii Foss that bounded rationality is much
cited and little used. In fact, there are few studies that reconsider Simons original
idea and try to explore its theoretical fallacies and/or interpretative shortcomingsiii .
As far as I know, the book you are reading remains one of the few works aimed at
getting a different idea of rationality in organizationsiv .
It is apparent from the previous pages that I do not believe the concept of bounded
rationality needs to be abandoned. Quite the contrary: There is enough evidence
leading to a better definition of it, on the basis of a more realistic and, maybe,
counterintuitive approach. However, these last points really dont matter if the DC
approach provides a better understanding of what we study.
The analysis whether the old BR and DC are inconsistent or not needs to be
developed further. I only gave the negative critique without going further on the
positive proposition. This and the next chapter serve exactly this scope. They present
an attempt to define how rationality extends over its bounds and where the analysis
should be focused when a theory of distributed cognition is considered. This is the
core of the book and explains its title. When referring to rationality, I use the term
extendablev to highlight its instability and plasticity, to define how rationality tends

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 81

DOI 10.1007/978-1-4419-7542-3_7,  C Springer Science+Business Media, LLC 2011
82 7 Stretching the Bounds (I)

to modify itself to cope with the challenges of decision making and problem solving.
Why extendable? The idea of extendable rationality is related to the following:

(a) human cognitive processes have the potential to modify themselves according
to the context (external resources) and/or to the problem at hand; this makes
rationality extendable in the sense that sometimes it does extend, sometimes
it does not;
(b) rationality is defined through its fitness to a given situation, it varies depending
on both individual and environmental factors, and how the two fit together in
this respect the modification, dynamic, plasticity is an extension (or, better, a
redefinition) of individual bounds to reach a better fit with the context;
(c) the outcome of a rational process is not suboptimalwhat is the benchmark
for suboptimality?or satisficing (individualistic approach), but it is workable,
i.e., it is appropriate in respect to the situation and/or to the problem at hand.

If rationality is considered in its extendibility, then we need to start analyzing

what makes this extendibility possible. This and the next chapter present two areas
where the old idea of bounded rationality had explanatory problems and the extend-
able rationality comes well into play instead. The first area includes the so-called
through-doing decision processes; these are dynamic contexts where the decision
maker changes and refines the process of decision making together with a first deci-
sion and the change in the context. The interplay with external resources (including
those externalized) is crucial.
The second area (Chapter 8) includes those processes that come into play when
other people are considered, i.e., the social side of rationality (see previous chapter).
There, I analyze how other people affect the way we make decisions, consider-
ing two specific examples (advice giving-taking and bandwagon). This latter point
is largely overlooked in studies of bounded rationality but, as we will see, it is
fundamental for any human decision-making process.vi

Through Doing Decision Making

This category includes decisions that are not well defined when the decision maker
starts the cognitive processvii . These are in fieri decisions, meaning they progress as
the situation evolves and resources become available. Some of them are intention-
ally made to follow up with other decisions (exploratory), while others are thought
to be well-defined and last decisions that reveal the need for further elaboration
while the decision is being made. The typical feedback processviii of a decision is
part of the decision-making process itself, and happens at the same time the deci-
sion is being refined. This defines the through doing or emerging decision-making
process.
There are several cases that should be made to explain how through doing
decision works. In the following pages, I consider two of them as paradigmatic
Through Doing Decision Making 83

examples: emotions and morality. Other examples of through doing decision mak-
ing that are not considered here but seem interesting to explore include maintaining
leadership, making sense of altruism, and personal and professional develop-
ment, understanding corporate culture, adapting to the informal (behavioral) social
structure of an organization, and more.

Emotions and Decision Making

In this book I have considered emotions when mapping bounded rationality. Studies
on emotions show how individuals are boundedly rational since they are together
an aid and a threat to decision makingix . However, we have also seen that neu-
rological studies have found that a rational decision maker cannot come to any
conclusion without emotions. Emotions are needed to make decisions, they help
us when quick decisions are needed, and they trigger heuristics, specifically fast and
frugal heuristics.
What emotions and heuristics in general help us do is that they prevent us from
looking at our bounds and provide us with the solution to a given problem. This
solution may be a certain behavior, a thought, an action, or anything that could
help us with a given problem. The decision comes out of something that remains
related to the way we usually process information, but sometimes it comes out of
the blue, to use an expression everybody is familiar with. It is not crucial here to
define conditions under which emotion can enhance or endanger the outcome of a
decision-making process, although these are very relevant points in recent researchx .
I would like to focus more on the meaning of emotions for cognitive processes: Why
are they so important that we cannot do without them?
One thing is that emotions are embedded in humans so that, at least up to a
certain point (primary emotionsxi ), they are part of our genetic heritage. In human
beings, these primary emotions are associated with complex thoughts or secondary
emotionsxii . When individuals make decisions, the consideration of eventual conse-
quences of that decision may lead to particular emotions when the decision is about
to be made. How many times do you check an important e-mail before sending it
out? And what does your wet hand feel like? Emotions are associated with specific
thoughts, and they can be a way to externalize or to have bodily reactions to these
specific thoughtsxiii . Now, the body is something outside the brain, it is an external
resource according to the definition (everything that is not the brain is an external
resource). We can argue that this external is very close to the brain but still exter-
nal. There is a continuity between body and brain, and this is the interesting part
with emotions. As soon as the emotion comes out, the body reactsxiv . We know
this for sure when we are in love, dissatisfied, experience fear of something, sue a
company, call a friend, or anything else. Our body reacts or represents emotions in
many different ways; this is what I should call the externalization of emotions. It
is in the external resources and body-brain interaction that emotions can be defined
as through-doing activities. This becomes apparent when we witness how emotions
evolve within a specific amount of time.
84 7 Stretching the Bounds (I)

Take the case of change. When change is disruptive, individuals usually react
with emotional arousalxv . When a company introduces significant changes in proce-
dures, or technological innovation is brought into the production process, individual
responses are framed through their emotionsxvi . In the case of mergers and acquisi-
tions, for example, the new board of directors and CEO usually attempt to get the
acquired company closer to the acquiring one. Layoffs, changes in the chain of com-
mand, higher turnover, new standard operating procedures, and more, are likely to
be implemented. Many people that passively witness all of these changes react with
a mix of emotions. Let us take the case of someone being removed from the charge
of leadership and replaced with a new boss who comes from the acquiring com-
pany. The discharged individual would pass through different states, including but
not limited to the (a) shock/surprise (e.g., What is going on?) together with refusal
or denial (e.g., This isnt possible!), (b) defense, such as an attempt to rationalize
(e.g., This was the right thing for the company), (c) discarding or acceptance of
the new situation (e.g., I am okay with this), (d) adaptation, that involves learning
(e.g., How can I do my new job?), and (e) internalization (e.g., This is working
well!; I have a new role in this team.)xvii . Of course, each one of these stages
is associated with emotions. As models of individual emotional reactions to change
show, these need the body to be felt. Once they are out there, individuals can inter-
act with them on a more conscious basis. I am not trying to understand which one
comes first between cognition or emotions (affect)xviii ; my intention is to show that
they usually stay together and that they need a process that is based on (a) external-
izations and (b) continuity. This second condition means that emotions are not stable
over time and they modify together with the exploitation of resources available.
They evolve through these resources.
Less apparent is the reprojecting of our emotions. The fact that they are external-
ized leaves room for taking these as external and to start thinking about them. For
example, the fact that your heart starts beating at a faster rate when you first met
your loved one forces you to consider behaviors associated with that state of mind
when it goes too far. Sometimes it is too late and you have already said something
stupid or that you didnt mean. The point is that there is a process of redefinition of
your cognition that increases when emotional states are more intense. This lets you
realize the emotional state you have run into. A recent idea on intelligence can help
us here.
A recent approach to human intelligence is that it is composed of many differ-
ent areas. There is the traditional (a) mathematical or logical intelligence that
deals with the ability of the individual to solve formal problemsxix , the (b) cultural
intelligencexx associated with the ability to understand and cope with people of dif-
ferent cultures, the (c) moral intelligencexxi that is related to the ability to behave
well in different ethical contexts, (d) social intelligencexxii a concept we will deal
with in the next chapter, and the (e) emotional intelligencexxiii that is about the abil-
ity to deal with ones own and other peoples emotions. I believe that this last area
of intelligence is nothing but the ability to gain self-knowledge of the distributed
process of cognition that deals with emotions. The idea is to connect emotional
intelligence to the DC approach. In particular, reprojecting seems to be very close
Through Doing Decision Making 85

to this idea of intelligence. A research question associated with this could be inquir-
ing if this is a typical trait of people with high emotional intelligence quotients, that
of having a better understanding of their (and others) cognitive processes.
While the self-detection and understanding of emotional intelligence is somehow
integrated to theories of bounded rationalityxxiv , what concerns other individuals is
not. If we externalize emotions, everybody does the same. In a social environment
where bodily reactions (e.g., a forced smile, a twinkle of an eye, crossed arms or
legs, arms down the body) are widely diffuse, we have learned to decode these
expressions and to adjust our behaviors according to perceptions. Or, we use these
as external resources and take them as major bases for our decision making. For
example, it is easier to ask for information when the person at the info-center greets
us with a smile. Otherwise, we opt to pick up a brochure and do whatever we need
to do on our own. This is a typical through-doing process. The decision to ask a
question or not may pass through the (light) emotion that encourages us to adapt
and modify our behavior as we go. Our everyday decision making is affected by
these simple events; we can think of a work environment where the same dynamic
is in place. It is easier to ask advice from a colleague or from your boss when she
greets you with a smile or when you feel from her bodily expressions that it is
a pleasure for her to help you. This external altruistic-sided face of emotionsxxv
is less considered and studied in organizational behavior literature, but it remains
important. How many of us make simple decisionsof potentially high impacts,
especially if repeatedon the basis of these perceptions? How could this affect
your solution to a given problem?
This is a practical note on distributed cognition that needs to be analyzed and
developed further in future works.

Morality
In his book Morality in a Technological World, Magnanixxvi presents an innovative
idea of morality that is consistent with the assumptions of the distributed cognition
approach and with the idea of extendable rationality.
What ethical theory always postulates is that ethics is something fixed, deter-
mined by the rules that guide individual behavior in our societies. De George points
out that
[i]n its most general sense, ethics is a systematic attempt to make sense of our individual
and social moral experience, in such a way as to determine the rules that ought to govern
human conduct, the values worth pursuing, and the character traits deserving development
in life.xxvii

It is apparent from the definition that we should probably be able to distinguish

between ethics and morality, the latter being the behavioral (descriptive) and the
former being the normative sides of the same coin. Even if we define the two sides
this way, it is particularly difficult to divide the two into separate domains of study.
Morality evolves with societies and so does ethics. If we think of racial issues,
86 7 Stretching the Bounds (I)

womens rights, temperance, animal rights, or many other ethical topics, we soon
realize that ethical rules and practices change. What was labeled as outrageous by
the majority of people only a few decades ago is now a moral standard.
The framework that the distributed cognition provides us with is that of a human
being that exploits external resources in a way that redefines her cognition. While
this is easy to figure out when we consider material external resources, it can be less
apparent when we deal with constructs of the mind, such as ethics.xxviii
Consider a work setting. The leader of the team asks one of the members to
perform the following task: Check if the outcome and quality control reveal that
the plant needs to renew machines involved in that production process and submit a
report. One week after a first deadline, the employee has not submitted any report,
but the leader knows that the employees mother had spent the last two days in
the hospital for a heart attack. However, the employee takes this allowance for late
report submission as something that the leader should give him, as something for
granted. The leader doesnt like it and decides to write an e-mail to this employee,
explaining what the policy of the company is and how good-hearted she has been
in overlooking this inefficiency. The e-mail is there and the leader is satisfied with
the lesson she is about to teach this employee. She is ready to send the message.
She clicks on the send button, and a pop-up window associated with spell-check
appears: Are you sure you want to send the e-mail? She never runs a spell check,
so the window is just annoying. Not this time. She has had time to rethink that
e-mail, the problems that the employee just had with his family, the fact that she
wrote the e-mail on the basis of an impression. What if she is wrong? She rereads
the e-mail and decides not to send it.
What happened? The points I would like to stress here are two: (a) What seems
appropriate at first wasnt so appropriate when the second thought came into play;
(b) technology served as a facilitator to this process. A diffuse moral rule (e.g.,
the golden rule of Jesus) states that you must treat everybody as you want to be
treated. This was what was going on in the case of the e-mail. However, the thoughts
of the team leader and the final outcome changed in the making. The decision was
finally made through doing (writing the e-mail). This helped the leader change her
moral interpretation of the case. After all, was it ethical to give the employee a
warning after her mother had that serious health problem?
The second point is also important since it tells us that resources can mediatexxix
morality. In the example, the pop-up window served as a mechanism that helped
the leader change her mind. This mediation is a corollary of the distributed cogni-
tion approach, since every resource that individuals exploit may be interpreted as a
mediator of a particular meaning. This is what morality also is: knowledge that helps
individuals in their decision making. External resourcesthe e-mail software in this
examplesupport cognitive processes and help the individual make decisions.
This simple example offers a basic idea of how morality is distributed. Our ratio-
nality adapts to situations, making the solution more appropriate; i.e., something
that suits our needs.
A final remark on the interpretation of morality is needed here. What does this
cognitive understanding bring to the study of morality? As Magnani points out, the
Through Doing Decision Making 87

distributed cognition approach shows the reasons why most individuals do not stick
with the same rule but adapt its interpretation depending on the situation. This also
has the potential to provide a cognitive explanation of why norms that have been
written several thousand years ago still maintain their impacts on people. In short,
what DC implies for ethical norms is that their understanding is situational because
it adapts to cognitive interactions. This means that ethical norms may be stated once
and for all; it is their framing that changes. A significant role is played, in the pro-
cess, by external resources. In the example that we used above, the boss could have
sent the e-mail. In that case, the result of sending the e-mail and its interpretation
would have been different. Maybe a rationalization process might have prevailed,
with the boss trying to make a rational explanation of the reason why her ethical
principle (e.g., the golden rule) needed to be adapted. The e-mail worked as a medi-
ator of that ethical principle; its absence made the individual replace that mediator
with other mediatorse.g., the e-mail itself, thoughts, other reinforcements such as
the memory of the company policy on internal deadlines or the code of conduct.
There is a need to anchor ones cognition to external resources. And morality is no
exception to this process. For this reason, understanding ethics in practice is better
off when using a through-doing logic.

A Third Logic: The Logic of Adaptiveness

In Chapter 5 I introduced the two logics that help frame rationality and decision
making. One is the logic of consequence and the other is that of appropriateness.
As you may recall, these two logics describe how decisions happen, and they do so
through a set of four and three questions, respectively. The decision maker (or the
analyst) answers these questions and makes sense of the process.
The logic of consequence is based on a strict logical progression of arguments,
and it is particularly useful for interpreting decisions on the basis of the old paradigm
of rationality. Therefore, it analyzes alternatives, expectations, and preferences, and
defines a decision rule. The logic of appropriateness is instead more concerned with
recognition, identities, social roles, and decision rules that apply to a given situation.
This second logic is close to the socialized view of the individual and of rationality.
Does DC fit one of the two logics? Where does it fall?
If one of the two logics has to be picked up, I would go with appropriateness.
That logic invites researchers to make sense of decisions through the cultural, social,
and political environments, where individual identity is questioned and involved;
it also deals with social external resources and internal sense-making activities.
However, how can you include externalization and reprojecting? How do you define
the non-definitive component (i.e., through doing) of every decision? How can you
distinguish among resources? In other words, there is something that the two logics
do not catch and that we can try to include in a third logic.
Consistent with the previous logics, I would like to proceed with questions that
help frame the decision-making processes. What follows here is a description of
88 7 Stretching the Bounds (I)

what I call the logic of adaptivenessxxx that indicates that decisions should be stud-
ied on the basis of resource availability, their actual exploitation, and pertinence to
the problem at hand. There are three questions to this logic:

1. The question of context: What resources am I exploiting?

2. The question of fitness: How good is the use of resources?
3. The question of change: How fast is the situation evolving?

The first question relates to the context: What resources am I exploiting? The analy-
sis of the decision passes through the careful consideration of the type of resources
the decision maker has (implicitly or explicitly) chosen to use in his/her cognitive
process. As we know from the DC general assumptions, what stays inside the brain
depends on what is outside such that resources available are those that are involved
in the cognitive effort. Potential use of resources should be separated by actual use of
resources. The second directly affects the cognitive process, while the first only indi-
rectly shapes it. Although more on external resources is written in the next chapter,
I anticipate here that there is a distinction between social and non-social.
The second question is that of fitness: How good is the use of resources? The
degree of success, efficiency, or effectiveness of a decision may depend on the inter-
play between what is inside and what is outside the brain. This implies that not
every resource is good for everybody. Instead, there should be a good fit between
the decision maker and resources at hand. Consider the following simple example.
The director of a PhD program in Human Resource Management asks one of his
students to write a review of an industrial economics paper. The author of the article
presents and discusses a mathematical model of firm-supplier relations. Our PhD
student feels incompetent and has no idea how to write the review. In this simplis-
tic case, there are two resources available. The first is to carefully read the paper
and try to make sense of the strange language through his own knowledge, Google,
and other library or online resources. The second is to carefully read the paper and
then bother a friend or another PhD student at the economics department. The first
option may be cognitively much more expensive than the second; however, there
are time constraints. The review may not take forever, and the deadline is the next
week. Is the first option worthwhile? How many papers like this will the PhD stu-
dent review in his entire career? It seems that the second option would work better
because its fitness (internal-external interplay) is less likely to produce uncertain
results than the first. Although this case probably needs to be treated more carefully,
the fitness question is of utmost importance and should be asked every time the
decision maker is about to use a cognitive external resource. What is the cognitive
effort needed to make a regression analysis with software you dont know? This may
be another example of low fitness. However, fitness is subjective, and the analysis
should discount personal matches of internal-external resources. In the case of the
PhD student, we may argue that for some the first optionmake sense of the article
on his ownmay result in a greater fitness. This adds complexity to the picture and
forces the analysis to consider the diversity of each decision maker.
The Rationality of Change 89

The third question for the logic of adaptiveness is that of change: How fast is the
situation evolving? This point is crucial for every decision maker. Changing con-
ditions imply a decision that could be (a) subject to fine tuning, (b) definitive but
irrelevant in the long run, or (c) part of a chain of subsequent and distinct deci-
sions. Fast-pace environments usually imply uncertainty and ambiguity. This, in
turn, implies that the decision one comes up with is also uncertain. Moreover, it
may become ambiguous because of the changing context. A way to reduce uncer-
tainty/ambiguity and increase the likelihood of coming out with a rational decision
is that of distributing the resources used in the cognitive process. We can make an
analogy with investment finance. One of the most important criteria to decrease
the risk of a portfolio is that of increasing the number of stocks. Otherwise stated,
traditional finance theory investors should follow the principle of diversification.
Bringing this analogy back to decision making, in order to decrease uncertainty, a
rational decision maker should diversifydistributehis/her cognitive resources
on a wider range of alternatives. It means that, when the environment is undergoing
a significant and fast change, the decision maker should diversify online resources
used; take more advice, suggestions, comments, information from coworkers; hire
consultants; etc. The level of affordance changes depending on how badly change is
related to the situation. Put differently, distribution should not continue ad infinitum,
but should rely on the condition of fitness.
The logic of adaptiveness is not an attempt to substitute the other two logics, quite
the contrary. This is an effort to create a third logic directed toward highlighting
those elements that are not included in the previous two. Therefore, the logic of
adaptiveness integrates the other two.

The Rationality of Change

The logic of adaptiveness points out that change is particularly important when ana-
lyzing the cognitive process of decision making. This aspect of the new logic is
particularly interesting because it reveals that DC may help understand how indi-
viduals cope with the changing resources surrounding them; it is a rationality of
change. Before getting to comments on change, a few words on technology may
help introduce the topic.

Innovative High Technology

In the last 20 years or so, several innovations have changed the way human beings
interact, think, live, work, rationalize, and make decisions. It is probably trivial to
point this out, but anyone who is required to solve any problem, whether it is per-
sonal or job-related, checks on the Internet to find some benchmark or help. Today,
this help is everywhere, even on your phone (I use it a lot!). Or, maybe you are
reading this book in its electronic version via your Nook, Kindle, iPad, or other
90 7 Stretching the Bounds (I)

device. You may have noticed that these devices are not only capable of making a
significant part of your library portable. They also come with a set of tools that allow
you to highlight and take notesand we are in the imitative-substitutive domain
still. They give you the opportunity to browse through a book as if you were looking
for something in your computer, to check the meaning of words using an integrated
dictionary, to buy other books on the same subject, to adapt the text on the basis of
your eyesight needs, to have the device read for you, and more. Why are these tech-
nologies so helpful? Why are humans that quick in making their commercialization
so successful?
I would argue that successful technology is highly compatible with human cogni-
tive processes. In particular, devices such as those mentioned above help individuals
have better externalization processes. Their cognition is distributed on tools that
favor more effective cognitive processes. This is similar to what can be found in
Clarks Natural-born cyborgsxxxi , where the author makes the point that individuals
are so embedded in their distributive processes that these devices are successful
to the extent that they become part of their cognition. Put differently, the con-
tinuous process toward more externalization and distribution is favored by new
technological innovation. This seems to be a constant trend in human cognitive
evolution.xxxii
However, the availability of technology is not the only factor that we should con-
sider in this analysis. In fact, there are costs associated with the use of these devices.
These costs (marginal costs) decrease with the increasing use of technology. This is
to say there are cognitive costs that are costs of learning. New externalizing devices
need periods of time where we understand and learn how to exploit them in order
to fit them into our cognitive processes. The more we use technology, the more our
cognition is affected and shaped by them. One could make the case that it is always
the individual that decides how to use technology, so there is control of that ongoing
cognitive process. Such a statement is very hard to analyze. I mean that it is hard to
have a precise idea of where confines are in cognitive processes involving high-tech
products. The boundaries remain undefined. Take a simple task, that of storing a
phone number on a cell phone. The device works as a memory storage for us. We
store the number so that we can remember it when we need it. It does not make
any difference who (or what) remembers the number; it is ours. If somebody asks
me if I know Toms phone number, my answer is Yes, I do. However, with the
traditional cognitive model of BR my answer should be No, I dont know it but it is
stored in my cell phone. The reason why we dont say that is because we use these
devicese.g., the cell phoneas extensions of our cognition. Is the number in our
brain? No. Is the pathway that leads us to find that phone number on our portable
device in our brain? Hard to tell. There are intuitive designs in the phone that lead
you to find the number. It is based on interaction, not on somebody remembering
every single step that leads you to retrieve the phone number. A user-friendly inter-
face leads you to the number. The process may become automatic (mechanical),
but in that case it is a waste of cognitive capabilities because the device may use
your intuition to find the number. You dont have to use internal memory, you
can externalize it as well as the process. Who controls this process? Is it you that
The Rationality of Change 91

pushes buttons on your phone to retrieve the number, or the software engineers that
designed the process that way? I think the answer lies somewhere in between these
two alternatives. What matters here is that the cognitive process comes out through
interactions. To support the idea that asking the question of control may be mis-
leading, I invite you to think on the fact that everybody seems to have a slightly
different way to use even simple devices such as a cellular phone. We have just dis-
covered that DC makes it easier to consider every cognitive process different and,
for this reason, to value the diversity of each human being.
Today, technology stays at the core of any business. We can no longer imagine
businesseseven farmswithout computers, for example. If we follow what we
have been arguing above, we may come to the conclusion that managers have the
opportunity to get the best from people. If we extend this paradigm to the next level,
we should probably lean on companies that auto-regulate, where control is a tool of
the past. However, I am going too far here. Chapter 9 helps to get a better idea of
some of the implications of DC for management and organizations.

The Process of Change

It is common sense that people do not like change. This is what is usually found in
management textbooksxxxiii , and what we tend to believe due to our own personal
experience. During the 2008 presidential campaign something strange happened,
though. The two candidates, Mr. John Mc Cain (Republican) and Mr. Barack Obama
(Democrat) had very different, if not opposite, campaigns. It was Mr. Obama that
campaigned under the slogan change we need. That and many other decisions
proved to be quite successful, as he became the 44th President of the United States
about America. Americans voted for Mr. Obama for several reasonsand I do not
know enough about American politics to comment on thisalthough some of them
explicitly liked his campaign, and wanted change, they were looking forward to
new management. The question is why do people seem to be willing to address
and support change in a situation like the elections, and are more reluctant when that
change is brought into the company where they work. What triggers repulsion in this
latter case and approval in the former? DC may help understand this phenomenon
better.
Change has been addressed earlier in this chapter. We connected change to emo-
tions and to how they follow a through-doing logic (that we may now extend to the
logic of adaptiveness). As many scholars point out, change processes are related
to learning mechanisms.xxxiv Following Lewins modelxxxv , there are three stages
through which change is analyzed: (1) unfreezing, (2) changing, and (3) refreezing.
The whole theory is based on the idea of a previous freezing state, where stabil-
ity is prevalent. That seems to be the state of nature for Lewin. This natural
state may change and be set on a different level depending on individual think-
ing and behavior. However, this means that a break in the continuity of the state
of nature needs to be made (i.e., the crisis). The problem becomes that of defining
stability and what perturbs it so to make it unstable and then stable again. Recent
92 7 Stretching the Bounds (I)

interpretations of this modelxxxvi suggest that unfreezing happens when the status
quo is disconfirmed. And this is also a fundamental process that occurs whenever
learning mechanisms are elicited. Many consider Lewins as a paradigmatic model
of change; in the following, I try to elaborate how it relates to DC and extendable
rationality.
DC provides a significant way to reframe change. Is change something that
occurs in isolation? Is it detached from the environment where it is happening?
Is it in the head of the one that is experiencing it? The answer to these questions is
negative. Whether change is natural, planned, or unplannedxxxvii , it occurs in a con-
text where external resources mediate it. As in the case of the manager that is about
to send an e-mail as a reprimand to one of her employees, we can argue that indi-
vidual as well as organizational change do not happen in isolation. There are always
resources that mediate what change could mean. In order to frame change from the
DC perspective, understanding the role of these mediators becomes crucial. This is
very similar to what Schein considers cognitive redefinition:

(1) semantic redefinitionwe learn that words can mean something different from what
we had assumed; (2) cognitive broadeningwe learn that a given concept can be much
more broadly interpreted than what we had assumed; and (3) new standards of judgment
or evaluationwe learn that the anchors we used for judgment and comparison are not
absolute, and if we use a different anchor, our scale of judgment shifts.xxxviii

By rephrasing what Schein wrote under the umbrella of DC, we are able to argue
that change is not a mere process of adaptation to external circumstances. It is some-
thing that needs to be cognitively processed. It is not a cycle that functions out of
input-transformation-output process, but needs a strict interplay with outside and
inside resources. These external or outside resources may represent changing con-
ditions, depending on an individuals ability to frame them. Without this matching,
no change could ever happen. The function of recognition or cognitive broadening
may happen only if individuals extend their minds, learning how to distribute their
cognitive resources in more effective ways. This means that bounds of rationality
are moving. Moreover, the unfreezing/changing/refreezing (UCR) process mimics
what is postulated by the distributed cognition approach, although (1) it is now clear
how individuals could stop performing UCR as a typical cognitive mechanism, and
(2) there is no such a thing as a stable state or absolute freezing. Therefore, UCR
processes never cease to happen; what the DC approach brings to the change dis-
course is that UCR is a matter of intensity. When intense states of UCR happen
then, change may be recognized by external observers, Otherwise, in a case of low
UCR intensity, we are witnessing the regular change process, typical of a distributed
cognition. For this reason, the DC is a rationality of change.
The next step of the study of how distributed cognition affects organizational
and individual changes may be that of analyzing what resources lead to perceptions
of change or stability and why. The intriguing part of such a study lies in the fact
that a need to analyze shared meaning of resources becomes fundamental to under-
stand decision-making dynamics in both individuals and groups. The next chapter
provides more clues on this approach to organizational studies.
Notes 93

The most important factor of change and learning mechanisms is that they are
processes that lead individuals to set themselves on different and new levels of
conceptualization. Put differently, they are mechanisms that we use to go beyond
particular bounds that limit our cognition. With learning and change, bounds of
rationality move and adapt to the new internal-external conditions.

Summary
The idea of extendable rationality has been introduced. It is based on the hypoth-
esis that rationality (1) has the potential to change according to the context, (2) is
defined through its fitness to a given situation, and (3) has workable outcomes, i.e.,
they are appropriate in respect to the situation and/or to the problem at hand. In this
chapter we have analyzed two particular aspects that relate to the bounds of ratio-
nality. Among the ways individuals have to stretch their rational bounds, there are
through doing decision processes and change processes. Although through doing
processes are pervasive, we have considered two examples: emotions and morality.
That analysis brought us to define a third logic for decision making that integrates
those of consequence and appropriateness: It is the logic of adaptiveness. Three
questions define this logic: the questions of context, fitness, and that of change. The
final pages of the chapter have been dedicated to (a) innovation technology and
its fitness to distributed cognitive processes, and (b) a further elaboration of how
change could not be relegated to outside conditions that affect individual thinking
but needs to be considered as part of the interplay occurring between internal and
external cognitive resources.

Notes
i. This is typical of neoclassical and mainstream economics (Von Neuman and
Morgenstern, 1944) but it is also consistent with bounded rationality, as Patokorpi
(2008) evidences.
ii. His point is that Simons Grand Theme of bounded rationality, first, has been rather
incompletely absorbed in the economics of organization; second, does not constitute
a necessary part of theorizing on economic organization and mostly serves a rhetori-
cal function; and, third, that part of the reason for this is that Simon and those who
have followed his lead did not develop a distinct, affirmative program for incorporating
bounded rationality in the economics of organization that would be satisfactory to most
economists (Foss, 2003b, p. 246).
iii. Most of the criticism comes from students that try to switch back to the traditional
paradigm; this is in part the reason why those who introduce and use bounded rational-
ity need to point out that it is a substitute for the full-rationality model (see above and
specifically Camerer, 2007, on how the contrast is still alive). However, many schol-
ars dont use the concept at all (making assumptions such as flawless and uniform
access to information or uniform cognitive abilities of players; e.g., Gilboa, 2010) or
mention it as a rhetorical construct (e.g., Nelson and Winter, 1982). Examples of the
latter are in Foss, 2003b, and any microeconomics textbook offers a picture of what I
mean (see Mas-Colell et al. 1995). One of the seminars organized every month by the
94 7 Stretching the Bounds (I)

Economics Department at the University of Wisconsin, La Crosse, featured a young and

freshly graduated PhD student that presented findings from his dissertation. Presenting
his complex and elegant model he explained that learning was not considered as a ratio-
nal characteristic of the players and that the model was a better fit for the data if learning
was excluded from individuals cognitive capabilities. Of course, we should be better
off without learning! This is not only against any BR, it is also against any observation
of real-life situation, including the one that brought that brilliant scholar to create his
model.
iv. Among those who criticize BR from a non-neoclassical perspective are Langley et al.
(1995), Zelen (2001), Patokorpi (2008), Sent 2005, Mousavi and Garrison (1992).
v. I am not sure of this word. I believe that a better word can be found to define the dynam-
ics and instability of human rationality. However, words such as dynamic, unstable,
plastic, debounded, moving when describing rationality all seem to emphasize one sin-
gle aspect of what I mean while extendable gives a potential modification at will and it
is related to the idea that rationality is an attempt to fit a given context.
vi. An exception of this overlooked domaini.e., the social side of rationalitymay be
represented by studies on the anchor bias. This bias needs external resources to come
into play and sometimes it cannot exist if a social actor doesnt point out the data, sym-
bol, word, or else, that function as the anchor for the decision maker. See chapter 4 for
references and further details.
vii. One may argue that all decisions pass through a phase of through doing, at least at the
beginning of the process when the decision maker has no clear idea how to frame it.
viii. March (1994), explains how a feedback mechanism works when the decision maker
uses a satisficing method for search.
ix. Hanoch (2002); Hanoch et al. (2007).
x. The role of emotions in decision making is considered under many circumstances;
there are studies in transient emotions, see Andrade and Ariely (2009), emotional labor,
Bono and Vey (2005), justice and emotions, Weiss et al. (1999), Barclay et al. (2005),
leadership, Emmerling and Goleman (2005), and many more (e.g., Hrtel et al., 2005).
xi. The seven attributes for defining a basic emotion are described in Panksepp (1982,
1992); in the discussion of disgust in relation to basic or primary emotions, Panskepp
writes that [p]rimary emotional systems, as far as we know, are intrinsic within
brain tools for allowing animals to generate complex, dynamically flexible instinc-
tual action patterns to cope with specific environmental enticements and threats. [. . .]
Their arousals are not restricted to narrow stimulus-driven survival issues, but ones
that can be related to fairly large-scale organismic survival concerns arising from many
environmental opportunities and exigencies (2007, p. 1821).
xii. See Damasio (1995), on this point.
xiii. On the primacy of emotions or cognition an important debate took place 25 years ago
between Zajonc (1984), and Lazarus (1984).
xiv. See Damasio (1995), and Gibbs (2005).
xv. Lazarus (1991); Liu and Perrew (2005).
xvi. For example, Liu and Perrew (2005) offer a model to study individual emotional reac-
tions to change. They argue that, in a planned organizational change, individuals go
through a cognitiveemotional process, in which they try to make sense of the change,
struggle with their emotional tensions, and choose their ways of coping. Moreover,
a change process typically involves an emotional episode that has four sequential but
distinguishable stages. Following Lazarus (1991), the first three stages are termed the
primary appraisal, secondary appraisal, and coping stage, respectively. The last stage
is termed the outcome stage of the planned change (p. 265; italics added).
xvii. The coping cycle of change is taken from Carnall (1990), chapter 7.
xviii. See the note above. This is the debate between Zajonc (1984), and Lazarus (1984).
xix. This is the traditional idea of intelligence; see Eysenck and Kamin (1981).
Notes 95

xx. This is an emerging concept; see Earley (2002), Earley and Mosakowski (2004), Thomas
and Inkson (2003).
xxi. An example of this can be that of Werhane (1999).
xxii. Albrecht (2006) offers an idea of social intelligence that is only in part similar to what
we intend here. The next chapter offers an explanation of how this concept relates to
rationality.
xxiii. Emotional intelligence became popular after Goleman (1995). It has been recently
discussed in its potential by Ashkanasy and Daus (2005), and limitations by Locke
(2005).
xxiv. See, for example, Hanoch (2002).
xxv. There is something in Damasio (1994), on this particular relation between altruism and
emotions.
xxvi. The book was published in English in 2007; the idea has been presented in many other
writings by the same author and in one working paper that we wrote together with
Bardone (2006).
xxvii. De George (1999), p. 19.
xxviii. Thoughts on thoughts are sometimes called meta-cognitive processes.
xxix. Magnani (2007).
xxx. Payne et al., (1993) are entirely dedicated to the adaptive decision maker. This logic
shares with them the intentions to capture the dynamic and flexible adaptability of
human decision making; and their work is a landmark contribution in the field. However,
I believe there are also significant differences between our approaches. Besides what is
in the text, Payne et al., share with Simon all assumptions typical of BR. They write that
their view of decision making as the application of a series of operators to knowledge
states is not unique to them. [. . .] More generally, their view of decision strategies is
closely related to views of problems solving as the application of a sequence of men-
tal operators (see, e.g. Newell and Simon, 1972; [. . .]) (Payne, Bettman, and Johnson,
1993, p. 11, italics added). Later in the book, they specify that [t]he idea that strategy
selection is the result of a compromise between the desire to make a correct decision
and the desire to minimize effort [. . .] fits directly into the concept of decision making
as bounded rationality (Payne et al., 1993, p. 73). They build on BR although their
analysis leads somewhere farther than Simons original concept.
xxxi. Clark (2003).
xxxii. See, for example, how Magnani (2006) interprets findings by Mithen (1999).
xxxiii. For example, Jones and George (2009).
xxxiv. See, for example, Schein (1999), or Carnall (1990).
xxxv. Lewin (1951).
xxxvi. Schein (1999).
xxxvii. Schein (2002).
xxxviii. Schein (1999, p. 61).
Chapter 8
Stretching the Bounds (II)

The previous chapter set the logic for analyzing cognitive processes in the making,
and defined change and high technology within the framework of distributed cog-
nition. This chapter looks at a broader set of variables in an attempt to broaden the
picture. It is about the unavoidability of the social context.
After presenting a classification of external resources, a model of information
sharing among individuals is considered. Then, a proposal is advanced on how
to define richness in information mediums. The chapter ends with a few notes on
bandwagon.

The Others
One of the most important, or probably the most important improvement that a
theory of distributed cognition brings to the idea of rationality is that of others. In
many of the examples used in this book I havent omitted the typical environment
that characterizes our life in and outside organizations. This environment includes
many external resources of which the most widely diffused, in terms of availability
and importance, are other individuals. It is almost impossible to think of a decision
that happens without the occurrence of another human being. Even when we think
of a decision in isolation, such as those of experiments, we need to discount the
experimenters biasi or the fact that individuals have been asked to be part of the
experiment by another human being and behave as if they are being observed. Or, we
can consider a decision in isolation similar to that of passing through a supermarket
lane to buy milk and sugar, and ending up grabbing bread, caramel, and apples for
instance. These look like decisions in isolation but they are, in fact, affected by other
people and many other stimuli:ii people that we observe buying that bread, or people
that set those ads with colors so bright that they are impossible to overlook! Most
of what we are affected by is social in that it has been set up by another human
being with the explicit objective of finding us receptive to it. This leads us to a first
classification of external resources.
According to what is stated above, external resources can be classified on the
basis of the type of relation that they establish with other people. Therefore, they
can be

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 97

DOI 10.1007/978-1-4419-7542-3_8,  C Springer Science+Business Media, LLC 2011
98 8 Stretching the Bounds (II)

(a) social, when there is an explicit willingness from other human beings to
communicate with other people;
(b) non-social, when the message is not intended to be public or when activities of
human beings are not crucial for interpreting the information embedded in the
medium.

We can, for example, tell that a newspaper article, a label on a bottle, drug directions
or warnings, a book, signs, etc., are all external social resources, while a river, a
certain behavior that is not intended to be communicative, your private notestaken
when you are not in publicfrom a conference/lecture/meeting, and the like are
non-social resources.
There is one caveat that needs to be made explicit. In the social world where
we live we cannot really state that there is something non-socially tied. The river
of the example is not completely non-social, and a tree is not really non-social.iii
The words that we use to define a river or a tree or to understand that what we
see is a river or a tree come as social constructs to our mind. Even if that process of
learning is so remote that we cannot remember anything (i.e., when we learned those
words), we know that somebody our parents, close relatives or somebody else
taught us that the woody perennial plant typically having a single stem or trunk
growing to a considerable height and bearing lateral branches at some distance from
the groundiv is a tree. From here we can see that every single piece of knowledge
is social because we can always define it through another human being. Even when
we create a new piece of knowledge, we try to set it up in a way that it can be
socially transmissible. The limit of this approach to external resources is that it
does not help to explain differences or to classify them. And, if it does not help
to classify resources so as to clarify our concepts, it is probably not worth using.
The difference in the proposed classification stays in the words willingness to
communicate something.
In the following I try to describe two cases where the DC approach may be con-
sidered and applied to decision making and rationality. The first is an attempt to
consider active and mindfulv exchange of advice as tools for decision making; the
second relates to passive or mindless imitation of other peoples behaviors.

Advice Taking
People take advice. This activity is widely diffuse in daily decision-making pro-
cesses as well as in one-time decisions. It happens with simple as well as with
complex decision processes. The answer to the question why people take advice
can be easily connected to the fact that these are simple, often free, and available
resources for people. This particular kind of external resource becomes available
when there is somebody willing to give advice that meets a request made by another
person. It goes without saying that in order for this social exchange of giving-taking
advice to take place, there must be a minimum of two people involved.
Advice Taking 99

The Judge-Advisor System

The two-person system is called the judge-advisor system, or most commonly JAS.vi
The advisor is the one that provides the other with the advice, while the judge is the
decision maker, i.e., the one who is supposed to make a decision on the basis of the
received suggestion. The system is defined in a way that the judge is about to make
a decision, and is offered with the opportunity to ask and/or take another persons
advice. Now, how many situations do you know that can be defined through a JAS
framework? I believe there are more than we can think of. And the reason for this is
very simple.
Since our childhood we have been taught to lean on advice from our parents and
other important family members and friends. This system continues to work every
day of our lives. The system is so diffuse that advice is offered even when it is not
solicited. When a window pops up and suggests updating your software now (this
happens especially with antivirus, for those who are still fighting against viruses),
that is typical unsolicited advice. Also, this is offered independent of the importance
of the situation you are facing and the decision you are about to make. Whether you
are choosing your dessert or have decided to marry your girlfriend or boyfriend, you
will find that there is always somebody providing advice. Moreover, advice comes
to you even if you dont want to listen to it. When you decide to quit your job and
look for a new one, people tend to suggest what to do. Sometimes you are not in the
appropriate mood to take or even listen to the advice.
I have isolated three factors of advice, then, since they (1) dont require explicit
requests from the judge, (2) are not related to the importance of decisions, and (3)
are not always wanted.
Of course, when advice is given, the judge faces two alternatives. Suppose the
CEO of an important firm that is facing a product recall has the option to go to the
press and deliver a message or to not go and opt for a standby tactic (e.g., she is ready
to deliver the speech as soon as the supplier responsible for the defective product
delivers its official version to the press). The CEO has been advised by the CFO
not to deliver the speech. In this case, she has two options: (a) follow the advice;
(b) dont follow the advice. The situation can be easily depicted by the following
Table 8.1:

Table 8.1 Judges options

Advisor Judge Decision

Option (a) Not Not Consistent

Option (b) Not Delivery Inconsistent

The decision of the judge (the CEO) is consistent with that of the advisor (our CFO)
when they both agree that the standby tactic is the best. The decision is inconsistent
when advisor and judge have different ideas about what to do and the judge ends up
delivering the speech.
100 8 Stretching the Bounds (II)

This simple model can be modified to fit a wide set of different events. For
starters, I consider n advisors situation and the judges opinion at time zero.
This latter situation is the standard starting point for judges. Except when people
face extremely uncertain or new situations, they often have an opinion on what the
preferred decision is. This opinion is going to affect the outcome of the final decision
depending on how strong the advice (and the advisor) is perceived by the judge. If
we consider the above mentioned example, new and different combinations become
available.
Table 8.2 A judge-advisor system (one type of advice)

jt0 A jt1 Decision

A Not Not Not Consistent

B Delivery Not Delivery Inconsistent
C Not Not Delivery Inconsistent
D Delivery Not Not Consistent

Note: jt0 judges opinion at time zero; a advice; jt1 judges opinion at
time one.

The judge may have two ideas on what to do, and the final decision may be
consistent with the advice or not (Table 8.2). Here we have two cases: case A in
which the judge doesnt change her mind and this is also consistent with the advisor,
and case B, where the CEO stays consistent with her prior opinion and disregards
the advice. The third option C is less likely to happen: It considers a situation where
the advisor is not consistent with the advice and a change of mind has occurred
(maybe because of the advice!). The final option D represents the case when the
judge changes her mind and conforms to the advisor.
Consider the case where we dont know what the advice of the CFO is. In this
case, the Table 8.3 below includes four more options, E, F, G, and H. This picture is
more complete and includes all the possible outcomes of the decision.
Now, what is the probability that the judge changes his/her mind? What is the
probability that he/she does follow the advice? Research on advice giving and

Table 8.3 A judge-advisor system (two types of advice)

jt0 a jt1 Decision

A Not Not Not Consistent

B Delivery Not Delivery Inconsistent
C Not Not Delivery Inconsistent
D Delivery Not Not Consistent
E Not Delivery Delivery Consistent
F Delivery Delivery Not Inconsistent
G Not Delivery Not Inconsistent
H Delivery Delivery Delivery Consistent

Note: jt0 judges opinion at time zero; a advice; jt1 judges opinion at
timeone.
Advice Taking 101

taking suggests that decision makers personal opinions strongly affect the final
decisions.vii However, a better answer to these questions requires an additional step.

Variables Affecting Advice Taking

It may happen that a judge has many advisors.viii In the example above, the CEO
may have advice coming from the CFO, somebody from the supplier company, a
union representative, and an external consultant. Now, the question is when advice
is too much. Is more advice always better than one single piece of advice? And, how
can we discriminate between them? Which one counts the most? Which is the most
important? Which can be disregarded and which cannot?
Research has indicated a set of variables that affect advice taking. I consider the
three that have the highest explanatory potential when we analyze JAS in organiza-
tions. The first is the extent to which the advisor is expert on the subject matter. It is
intuitive that the advice has more chance to be taken when the person providing it is
knowledgeable and/or has experience in that situation.ix Suppose that, in our case,
the CFO happens to have more experience than the other advisors because he is not
facing the situation for the first time, having worked for 25 years in that company,
knowing the supplier well, or similar factors that make him an expert. If this is the
case, then the CEO is more likely to trust him and end up with a consistent deci-
sion. However, the experience of the advisor needs to be discounted with what the
judge thinks his/her own expertise is and with whether the advice is required, by the
judge or not. When the advice is required, the probability of a consistent outcome
is higher.
The second variable is the social relation between the two (or more) players of
the JAS.x With this I mean the fact that the advisor is close to (or is considered close
by) the judge plays a crucial role in the final decision. If the CFO has never given
advice to the CEO and this is the first time this happens, while the relation between
a top manager at the supplier company is stabler, the advice with the highest weight
will be that of the latter, not the one coming from the CFO. Now, when is a social
relation stable? This is a point that is not easy to analyze, but I believe that we can
bring a bit of sociology into this.xi The social relation must be important; it doesnt
matter if it is based on the fact that the two share the same country club membership,
meet at Sunday masses, play tennis together, graduated from the same university, or
have established a sound professional relation. As far as advice taking is considered,
what matters is that this judge-advisor relation is stronger than other relations, and
it is so strong that the advisor (e.g., the supplier) affects the judges (our CEOs)
decision.
The last variable that I would like to consider in this short analysis of advice
taking and giving is the medium that is used to let the judge get the advice.
Unfortunately, research in this particular domain is not developed enough to present
its results.xii However, there are studies that support the idea that the medium
through which information is transmitted makes the difference. I try to present a
few hypotheses on how the medium affects decision making in the following pages.
102 8 Stretching the Bounds (II)

Information Richness
First of all, we need to find a criterion to define the sources of information based on
how well they transfer the data they carry. The degree of information that a medium
can carry is called information richness (Fig. 8.1). To be more precise, a medium
is high or low in richness based on its capacity to facilitate shared meaning.xiii
Scholars define a scale of information richness that allows us to distinguish between
mediums so that at the lower part of the scale there is an unaddressed document and
the highest part is where face-to-face is located.xiv

Unaddressed Written,
Telephone Face-to-Face
documents addressed comm.

LOW HIGH

Fig. 8.1 Information richness scale Source: Daft et al., 1987, p. 358.

Very little specific research has been conducted on media information richness
and advice taking.xv Also, most of the experiments on advice giving and taking
do not use direct human advice but computer-based. This is a very interesting fact
because, instead of seeing it as a limitationas those who run experiments do in
their writingsI believe it is a significant and valuable aspect. The argument is sim-
ple and related to the question: How do people communicate the most? What is the
medium that most of us use when at work? The answer is simple, as I anticipated. It
is the computer that shapes the way we stay in touch with people, we communicate,
we feel about our job, we ask for advice, etc.xvi There is an entire universe of free
and proprietary software out there that shapes our way of communication. Most of
our information passes through these media.
By the way, the information richness scale was defined in 1987 by Daft, Lengel,
and Trevino (Fig. 8.1). This means that the authors had the first thoughts about it
in the years before the so-called IT revolution, and I guess it may have been 1985
when we consider publishing time. In those years the computer was not as diffuse
as it is right now, and it is understandable that their scale does not include e-mail,
chat software, video conferencing, web sites, blogs, social networking websites, and
the like. Advice giving and taking is something that heavily depends on the medium
used to share the information. The definition of an up-to-date information richness
scale (also called the Media Richness Theory, or MRT) is particularly needed and
useful, although there are studies that suggest it should be abandoned when related
to IT.xvii
A few caveats apply here. According to the definition, information richness is
defined by and associated with a given medium. This is not entirely correct. How
rich a medium is depends on the information it could potentially contain and on the
two (e.g., human or computer) parts involved in the exchange. Therefore, we should
Advice Taking 103

write that the scale is about the potential information richness that a medium could
carry, given the fact that how rich that information is can be defined by the recipi-
ent only. Emotional states, knowledge on how to handle and use a given medium,
frequency and past history of usage (confidence), and other variables could affect
the scale. Consider the unaddressed document, an advertisement that you find in
your mailbox. Most of the time they are completely useless, but not when you come
home really hungry and your eye cannot do anything but stare at the pizza delivery
ad you just found in your mailbox. In that case the medium provides you with poor
but extremely useful information. Or, in that case, the information contained in that
medium is rich in the sense that you didnt need more than that information; that
one is appropriate. This is to say that richness depends on the fitness between the
individual and the external resource available. Can we associate an absolute value
to any given medium? Is information richness all things stated once and for all,
independent of the individual that uses that medium?
It is worth noting that, according to our definition of social resources, all of the
mediums in the scale are social resources. Therefore, it is apparent that the points
on information richness are nothing but a corollary of DC and the relative idea of
rationality and decision making. The ability to facilitate shared meaning lies in
the interaction between individuals and the medium, not in the individual, not in the
medium. The ability to facilitate shared meaning is thus related to how individuals
are willing to exploit that medium so that they can share their knowledge.
How can we redefine this scale in order for it to fit the JAS model? Three fac-
tors affect the scale: (a) advisor, (b) judge, and (c) medium. We are interested in
the fact that richness depends on two individuals willing to share their knowledge.
This brings us to the theory of distributed cognition.xviii Individuals make sense
of everything through their own introspectiveness; however, part of the meaning
of any given problem is common to more than one individual, i.e., it is shared.
Introspection is the key that individuals use to get acquainted and informed of any
given problem, resource, person. It is their personal way to access the problem. In
the following Fig. 8.2, the green circle represents the common understanding of the
problem while the yellow corners are the different perspectives (or framing access)
that every individual (A through D) has on it.
The four arrows have double directions because we assume that the problem at
hand shapes the cognition of these four individuals. This is a very approximative
and (hopefully) effective representation of what happens when individuals share the
meaning of a specific piece of information. It is helpful to point out that if a medium

Individual A
Individual B

Individual C
Individual D
Fig. 8.2 The shared meaning
of communication
104 8 Stretching the Bounds (II)

is something that helps individuals share any part of their cognition then it is not
the medium that appears to be rich, but this richness depends on the interactions (or
interplay) between resources.
Moreover, mediums are not rich in vacuum (i.e., in isolation), and this is the rea-
son why, as already stated, we should call the original scale potential richness
information scale. The context and/or the situation provide the actual level of rich-
ness for a given medium. However, any medium has the potential to be extremely
rich, depending on who is using that and why. This leads us back to the interplay
between external and internal resources, once again. Consider e-mail writing. You
are not very good in English and you need help. A close friend of yours is there
and helps you; you also use a dictionary and the Web. How many resources help
the transfer of meaning? What is the upper limit where one more resource decreases
information richness? There are no easy answers to these questions, and studies on
advice taking have barely one. It seems that we are coming out with a way to classify
media and advice taking, though.

Information Richness and Advice Taking: A Proposal

The analysis of the medium should be carried out on the side of the advisor also,
not only on that of the judge, as it has been with the previous discussions. Choosing
the right medium could prevent the fact that the advice is taken or even considered
by the judge. For the medium to be effective in the JAS system, it needs to meet
the judges needs and the advisors expectations. The judge needs information that
has some degree of usefulness, while the advisor has expectations that the judge
finds the shared information useful. These two, i.e., needs and expectations, can
be considered as the common motivation for advice giving and taking. Also, this
means that the medium must meet the condition of appropriateness in order to be
effective. And that piece of information is appropriate when both judge and advisor
find a common ground where it is possible to share it. This does not mean that
the judge will agree with the advisor, but it highlights the fact that the medium
is transferring information well. Put differently, that medium becomes exploitable
from a cognitive standpoint, i.e., it becomes an available external resource. Within
this system, information richness can be measured by the successful use that the
judge makes of the advice. If the information is used then the tool (medium) selected
has been rich enough to help; if the information has not been correctly categorized
in the medium and hence transferred, then the result will be poor. The concept of
shared meaning that Daft et al., use is absolute while it must be relative. Therefore,
I propose to explain the richness scale according to Fig. 8.3.
The two variables in the figure are distributed cognition (dc) and motivation or
predisposition toward the advice (mo). The latter may be negativea negative moti-
vation to take advicewhile the former is only positive, indicating that individuals
may distribute or not, but it is hard to think of a negative exploitation of resources.
Different levels of dc may be associated with mixed values of mo. What we should
Advice Taking 105

Fig. 8.3 Information richness fit

consider as the probability that the judge has to take the advice is the area included
by the y axis and the curve. Depending on motivation, the curve may move up or
down. An overall tendency not to take advice may have the curve move down such
that the negative area increases (i.e., taking the advice becomes not very useful,
therefore it is perceived as not needed by the judge). The parable may also become
wider or narrower, depending on the dc potential attitudes of the decision maker.
For any given level of dc there is a variability of mo so that the use of the external
resource (advice) through the medium may be probable or improbable (depending
on the area). The more the medium and the advice fit the dc system of the judge, the
more likely it is that the external resource will be exploited.
The four curves that you see in Fig. 8.3 are examples of different mediums and
of their fitness (or appropriateness) to the judges distributed cognition. The more a
medium employs dcmeaning that it is easier to get for the decision makerthe
more likely it will be used. Motivation is also higher in that case.
In a world where we can choose, the advisor selects the medium depending on
how comfortable she is with that medium and on how she thinks the medium is
good for the advice to be accepted (e.g., writing e-mails, if she believes this is the
appropriate medium for that kind of advice). She expects that the advice (informa-
tion) is easily transferred to the judge, i.e., enters in the cognitive process and helps
decision making. Put differently, it gets shared. In our hypothesis, there are four
alternatives (in the Fig. 8.3 they are m1, m2, m3, and m4). However, the advisor will
not choose a medium that takes too much of a cognitive effort; that is to say, that
is not likely to become part of how the judge distributes cognitive resources. For
example, instant messaging on the cell phone usually takes more effort if you are
older while it is easy, usually fast, effective, and widespread for the young. The use
of text messaging depends on who you are (in most of the cases). If what I argue
holds true, then the advice is more likely to be taken when cognitive efforts are on
106 8 Stretching the Bounds (II)

their average and expectations that the judge will make good use of it are reason-
ably high (it is m4 in Fig. 8.3). When cognitive efforts are too high, then according
to this hypothesis, it is more unlikely that the advice hits the target. The cognitive
effort has to be commensurate with the motivation (i.e., expectation) that leads the
advisor to give advice. The example of text messaging with the cell phone shows
what the proposed model means. If texting is too much of an effort for you, it means
that you are not confident enough with the medium you are using. It is likely that
your perceptions on the level of uncertainty surrounding this medium are high, and
you might prefer to use another medium.
Although the perspective of the advisor is very important, Fig. 8.3 shows what
happens to the judge. The judges needs are described as motivation in the graph.
This is because there may be motivations different from the need or utility of
the medium, such as expectations or goal attainment. In the example, motivation
becomes more and more positive as the compatibility with the cognitive (distributed)
decision-making process increases. The negative/positive domain may indicate the
likelihood of the judge to take the advice. Let us try to explain this dynamic with a
simple example. The judge is not comfortable with short text messages taken by the
phone. It takes a while to read them and they are, according to the judge, not reliable
because they are too short and impersonal. Therefore, the likelihood for this judge
to take the advice is given by what shown in Fig. 8.4.

Fig. 8.4 A low richness medium

The grey area represents the probability that our judge may take the advice. The
more the medium selected gets closer to what the judge thinks appropriate, the more
a positive motivation toward the advice arises. As I keep repeating, this also means
that the medium is fully integrated in the cognitive processes of the judge. It is rich
because it fits the judges needs; it is more appropriate. We can hypothesize that,
for the judge, a small chat may be the optimal way to communicate. In that case,
the medium is m1 in Fig. 8.4. It is worth noting that, if we use this approach, the
Advice Taking 107

same medium (a) maintains an area of substantial inappropriateness (the negative

domain of mo), and (b) its degree of richness may be different when we consider
other individuals.
The first point (a) takes into consideration the intrinsic ambiguity related to
any sort of communication. The medium, even that considered significantly rich,
may not be chosen by the judge (the negative domain in the Fig. 8.5). The second
point may be explained by considering how a text message could be accepted by
another judge, one that is more familiar with that kind of communicationcognitive
fitnessand highly motivated to accept the message (Fig. 8.5). The grey area (m1),
in this case, is significantly greater than that of the previous Fig. 8.4, and extends
the richness of the medium and the likelihood that the advice may be taken.

Fig. 8.5 A high richness medium

To sum up, the medium used conveys information and is rich when it is (a) needed
by the judge, (b) cognitive efficient for the advisor, and (c) the advisor retains decent
expectations that the advice will be useful.

Perspectives on Advice Taking

There are several limits on the studies of advice giving and taking. One of the typi-
cal criticisms is that almost all of the findings derive from experimental settings and
field studies are missing. This limitation applies to all recent model/theory devel-
opments based on experiments. Field studies, i.e., observation and interviews on
site, usually in organizations, are time consuming, difficult to set up because of
the increasing skepticism of managers toward research, and have variables that are
harder to control. However, they are richer than experiments and give the observer
a clearer idea of mechanisms that are in play in the working environment.xix
108 8 Stretching the Bounds (II)

A more stringent limitation is that of measuring the quality of the advice and the
extent to which the advice has been taken. This is not a secondary topic as far as
advice is concerned. Sometimes advice content is only partially considered, and the
quality of it is not taken into consideration. The richness scale that I proposed in the
previous pages can be a useful tool for defining this last point. Studies have focused
on the successful solution of a given problem when advice has been used by the
judge, and they have taken this as a measure of the quality of the advice. This works
for task-oriented advice only but not with life, behavior, and other similar kind of
advice. If my personal advice to you is to continue reading at least the next chapter
of this book because you will have food for thought, you may follow only half of
that advice and read a few pages at the end of that chapter. However, I will never
know how good this food has been for your thoughts unless I ask you a specific
question on this point. And even if you answer that question, I expect your opinion
to be fairly well-motivated and grounded in your knowledge (both newly acquired
and previous knowledge). How can we measure this? How can we consider this
complex aspect of information sharing and byproduct of advice taking?
This is not clear yet from research, and one of the reasons why it is not so relates
to one major weakness of the above-mentioned models. A general theory of advice
giving and taking is missing. What scholars do in this domain is collect data on how
people react under specific circumstances. The implication of this is that we have
several ad hoc explanations of how people behave when advice is provided, but we
never get the whole picture. In the next chapter I will attempt to provide a theory of
individual behaviors in organizations that includes advice taking as one of its major
points.

Passive Advice Takingxx

As defined above, bandwagon is a fallacy, something that is bad for logic but
(sometimes) good for life, that results in a bias. In particular, this ad populum fal-
lacy (see Chapter 5) describes the situation that people face when, for example, they
have to make decisions under uncertainty, are short on time, seek legitimacy in the
group, dont see other alternatives, and the like. Broadly speaking, whether they use
a rational calculation or feel under social pressure,xxi biases, prejudices, and rational
boundsxxii work to make the choice of bandwagon more likely. These shortages of
cognitive capabilities can be, and sometimes are, used by companies, organizations
at large, and governments to suggest (or nudge, as Thaler and Sunstein put it)xxiii a
preferred alternative.
Bandwagon is a social imitative behavior that becomes widespread or popular.
The idea is that it works as a threshold. Granovetterxxix explains that bandwagon
could emerge at different times and on different base for different people. What
triggers mine can be different from what triggers your imitative behavior. How many
people do you need to see wearing orange before you start thinking that they look
nice and that you want to dress in the same color? And how many people do you
Passive Advice Taking 109

need to see with ice cream in their hands before deciding Yes, I would like to get
one? How many people must discount information coming from that turnover ratio
before you decide that it is not affecting the low performance of the company?
It is the case of the employee stealing pens or copy-paper from work because
everyone is doing it. Another example might be the idea that you cannot ask that
colleague to retrieve data for your report, because nobody does and there must be a
good reason for that. These are very simple and quite meaningless cases, but think
of employees disregarding one, two, three, and then n symptoms (i.e., data) of an
upcoming crisis (reasons for sales increase/decrease, shareholders claims, environ-
mentalist protests, etc.,) because everybody thinks they have no meaning and It
makes you feel like a fool if you think they are. And the crisis arrives. Think of the
number of cases where this could be a simple but accurate explanation for what hap-
pened. Are you thinking of subprimes, the crisis companies are facing right now?
Well, I do not have the data and do not pretend bandwagon can be the sole cause
for the crisis, but it is fair to consider that it could have contributed. You decide the
extent to which it is important in such cases; however, the point is that bandwagons
are an important part of organizational behavior and life.xxv
What stays at the basis of bandwagon? Let us study the simple example: stealing
pens for personal use from the company stockroom. The employee tells himself or
herself that these pens are for his or her child as a justificationwhich might happen
to be true; however, that is not the point here. The idea of thresholdxxvi is particu-
larly useful. The employee decides to steal pens only when he or she observes that
a certain number of other people behave that way. This number of people is his
or her threshold. Then, the employee imitates what other people do. Bandwagon
is about imitation. As biologist Kevin Laland suggests,xxvii imitation is related to
social learning and it is very common in primates and other animals. However, when
we refer to human beings, what kind of learning is this apparently mindless behav-
ior? If you steal a pen from the stockroom because you imitate other people, are you
learning anything besides how to steal?
Bandwagon is one of those phenomena that managers would like to minimize
in their company. Especially in conditions like those of the western countries where
economic services are key, companies tend to ask people to think, to use their brains,
to try not to follow the mainstream. There are plenty of companies that are in des-
perate need of people that give their unique contribution, that develop attitudes like
creativity and continuous learning.
Bandwagon facilitates integration of a new hire, because it is based on the
same mechanism cognition works: it leverages external (social) resources. However,
these cognitive processes are cheap, they do not involve active choices but lean
on mechanical processes instead. This fact, when not recognized, becomes dif-
fuse in too many areas of ones behavior and could undermine performance, social
relations, and individual contributions to the company. Bandwagon is a passive phe-
nomenon that equals advice taking since the decision maker takes other peoples
behaviors as (sometimes) involuntary advice giving. You see people stealing pens
from the stockroom, therefore it might be okay to do that. It is a surrogate for asking
Is it okay to steal pens from the companys stockroom? The behavior provides the
110 8 Stretching the Bounds (II)

answer to that question, with no need for you to ask it. As many cognitively passive
phenomena, it has drawbacks that need to be fully addressed.
This topic is not new for management scholars; however, most studies focus on
bandwagons between companies or as an individual phenomenon.xxviii Imitation has
been studied in strategic management to understand how innovation passes from
company to company. The integration of bandwagon in organizational behavior has
yet to come. It should attach to the study of culture, routines, corporate norms, and
other typical elements that shape human behavior in organizations. It is relevant here
because it is close to cognitive processes described in this chapter.
The idea of extendable rationality includes the fact that a seemingly sound behav-
ior is effective and appropriate to a particular situation. Nevertheless, an idea of
rationality based on the distributed cognition approach highlights that this extension
lacks the crucial mechanisms that make human decision making so complex. In the
next chapter I will be back on bandwagon and compare it to a similar but more
distributed phenomenon: docility.

Summary
In this chapter we classified external resources into (a) social, when there is an
explicit willingness from other human beings to communicate with other people,
and (b) non-social, when the message is not intended to be public or when activ-
ities of human beings are not crucial to interpret the information embedded in the
medium. This distinction led us to focus more on a human typical communication
process, that of advice giving and taking. We have defined the judge-advisor system
as a useful model to analyze whether advice is consistent between two individu-
als or not. Moreover, we have learned that the effectiveness of advice depends on
actual or perceived expertise, social relations between judge and advisor, and the
medium. A discussion of information richness and a proposal for a new model to
frame it followed that analysis. According to the model, information is rich when it
is appropriate to the circumstances that affect the judge-advisor relation, i.e., moti-
vation, cognitive processes, and their fitness. The chapter ends with the proposal to
consider bandwagon as passive advice taking.

Notes
i. One of the first times it has been found in management was during the famous Hawthorne
studies conducted by Elton Mayo. This is also known as the interviewer bias (Groves
et al., 2009, p. 292f).
ii. The behavioral economics literature on cues and cueing supports this non-isolated
hypothesis (e.g., Politser, 2008, p. 64f). Thaler and Sunstein, 2008 in their book Nudge,
offer several examples of how cues affect behavior.
iii. A similar point of view is expressed in Maturana and Varela (1987).
iv. New Oxford American Dictionary.
Notes 111

v. The two words mindful and mindfulness are related to the work of Langer (1989).
See also: Fiol and OConnor (2003) and Levinthal and Rerup (2006).
vi. See Bonaccio and Dalal (2006).
vii. See, for example, Gardner and Berry (1995), Harvey and Fischer (1997) and Yaniv and
Kleinberger (2000).
viii. This and the following points are treated in the extensive literature review by Bonaccio
and Dalal (2006).
ix. There are studies that have isolated this phenomenon of the impact of expert/novice
advice to the judge; e.g., Harvey and Fischer (1997) and Sniezek et al. (2004).
x. This is highlighted by Bonaccio and Dalal (2006, pp. 145146), and in Slaugher and
Highhouse (2003).
xi. See for example Granovetter (1973), and Chiang (2007) on a particular kind of social
relation known as bandwagon.
xii. I was able to find Hedlund et al. (1998), on the difference between face-to-face and
computer interactions in advice giving and taking.
xiii. Daft et al. (1987, p. 358).
xiv. Daft et al. (1987).
xv. There is some research on negotiation and media richness. See for example Purdy et al.
(2000), Drolet and Morris (1995) and Moore et al. (1999).
xvi. I take a slightly different perspective from that of Kock (2005) here since he suggests that
face-to-face is still what people prefer in business communications.
xvii. El-Shinnawy and Markus (1997) and Dennis and Kinney (1998).
xviii. This particular assessment of distributed cognition overlaps with what Cannon-Bowers
and Salas (2001) call shared cognition.
xix. Social network analysis is particularly useful and helpful when it comes to analyzing
advice taking. See Knoke and Yang (2008) (at page 13 they explicitly refer to advice
taking).
xx. I owe the idea to treat bandwagon here to Emanuele Bardone, friend and coauthor of
many articles. His PhD dissertation (2008) is a clear example of how the multidisciplinary
approach to cognitive science is the past and especially the future of this field of research.
This section of Chapter 8 is a summary of the working paper A Model of Organizational
Bandwagon.
xxi. Abrahamson and Rosenkopf (1993).
xxii. Kahneman (2003).
xxiii. Thaler and Sunstein (2008).
xxiv. Granovetter (1978).
xxv. Chiang (2007) and Granovetter (1978).
xxvi. Granovetter (1978).
xxvii. Laland (2001).
xxviii. See Fiol and OConnor (2003).
Chapter 9
The Docile Organization

When I visited the Acropolis at Athens, I remember that beside the wonderful
Parthenon there is a smaller temple called Erechtheum, which is dedicated to
Athena, goddess of knowledge, war, arts, and justice. This small temple has a world-
famous characteristic: Its columns are feminine statues that carry the weight of the
roof. They are the caryatids. They serve a very useful purpose, that of preventing
the temple from falling down and, in doing that, allow everybody to get into it.
It is not in the power of these beautiful statues to decide who can and who can-
not enter the temple. The previous chapters are like the caryatids since they bring
the theoretical infrastructure up but cannot help state what can and cannot be a
future theoretical outcome. They are premises and allow multiple options to become
available.
This chapter presents a theory of decision making in organizations that uses
extendable rationality and the distributed cognition approach in a way that brings
us to a better understanding of organizational dynamics. It is the theory of
docility.

The Docile Individual

The idea of extendable rationality represents a connection between the theory of
bounded rationality and the distributed cognition approach. The link between the
two explains what happens with the through doing and the social aspects of deci-
sion making. In the following pages I try to generalize these assumptions. Is there
a trait that defines human beings on the basis of distributed cognition? What makes
us behave as rational beings?
Ethics, bandwagon, emotions, and advice taking all have a social basis. The dis-
tributed cognition approach seems to support hypotheses on the social functions
of our intellecti , as well as those on the social brain hypotheses,ii and on ultra-
sociality.iii This gave the input to the late Herbert Simon to explain the concept of
docility. This idea was not new since he mentioned it in one of his first writings,
left it undercover for almost 40 years, and then published two important articles
on that in 1990 and 1993. The first paper was published in the prestigious Science,

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 113

DOI 10.1007/978-1-4419-7542-3_9,  C Springer Science+Business Media, LLC 2011
114 9 The Docile Organization

while the second was published in the American Economic Review. Notwithstanding
these important outlets and the Nobel, the ideas exposed there and the implications
were never the subject of any subsequent scientific papers until 2003, when Knudsen
published the first paper dedicated solely to docility after Simon. Considering what
I am about to write, this is quite impressive. Why did this happen? Why did the
scientific community overlook this idea? I have two distinct explanations.
Before getting into these explanations we need a definition. Docility is
the tendency to depend on suggestions, recommendations, persuasion, and information
obtained through social channels as a major basis for choice.iv

It is a typical human trait that emerges when we make decisions on the basis of
information coming from other human beings. The fact that Simon defines it through
the terms suggestions, recommendations, persuasion, and information allows us to
relate docility to advice giving-taking and to emotions in a fairly direct way. Docility
is a trait that unveils the way our cognition works; it is a behavior consistent with the
way our cognition is distributed. As we state elsewhere in this book, social resources
are major sourced of cognitive activity, and there is no surprise that we can find a
specific human trait that defines how we handle this information. From this specific
trait, behavior comes out. What is typical of human beings in social settings? They
exchange opinions, tend to learn from each other, and make decisions on the basis
of information they gather from other members of that team, group, organization, or
social environment. Now that we have a first idea of what docility is, we can go on
with the following points on why the scientific community overlooked it.
The first is that the idea of docility was not consistent enough with that of
bounded rationality. Even if Simon brings docility into consideration because
of bounded rationalityhe states that we are docile because we are boundedly
rationalv the passage is not clear enough. The fact that our rationality is bounded
doesnt lead directly to docility. It is probably a sufficient condition, but it is not
necessary. It is sufficient because without limits to rationality we should never rely
on external resources since we are omniscient in that case. If rationality is limited,
then we get information from the outside where social channels supply part of it.
However, this condition is not necessary since it is not clear why humans should
lean on social channels more than other resources. It is not a need that we have to
rely on external social resources; broadly speaking, the fact that we are bounded
doesnt necessarily lead to a particular kind of external resource to overcome our
limits. Therefore, I believe this makes the connection sufficient only. Simon never
made this point clear, and I believe this didnt encourage scholars.
The second argument is that the scientific community has had difficulty using the
concept. Simon presented a simple model where he applied a Darwinian approach
to a community of human beings showing that non-docile individuals do not sur-
vive and instead go into extinction. This strict interpretation of social Darwinism
came late in the scientific debate, at a time when fallacies of these applications had
already been identified.vi Therefore it was too hard for an idea like that of social
Darwinism to enter the debate again. What students of decision making, manage-
ment, and economics overlooked was the deepest meaning of being docile. As I
The Docile Individual 115

explained in one of my articles,vii I agree with most of the critics that any form of
social Darwinism is too deterministic to work. I also believe that we have softer
theories that explain how two or more species could survive in the same nicheviii
(or environment), and we can argue that social dynamics are complex to a point
where it is not clear how a certain attitude (e.g., thought or behavior) has really
gone. This is only one of the possible explanations of why students overlooked the
idea of docility. Another could be related to the fact that distributed cognition theory
had yet to come in the early 1990s. It is only a few years later, but we now have the
emerging trend of studies pointing out what external resources are for cognition.
And I believe this makes the difference when going backward to interpret an idea
that wasnt successful in the past. We didnt know how to deal with it. Now, we
know.
A third explanation can be that the concept is not worth studying because it has
a limited explanatory power. Well, even if we had one reviewer making exactly
that argument one time, on the basis that there is insufficient literature to support
the argument, Bardone and I disregarded this last option since it is a typical ad
populum fallacy.ix It is too early to dismiss this concept given its matching with the
DC approach and extendable rationality. To get to the point, we need to discuss a
few more assumptions underlying the model and the theory. In the following pages,
I discuss fitness, altruism, and selfishness, docility as a theoretical background for
advice taking, and conclude with decision making.

Fitness
I would like you to run a thought experiment to understand docility. Imagine a world
based on individuals that are 50% docile and 50% non-docile.x This means that half
of them lean on external resources (suggestions, recommendations, etc.,) to make
their decisions, and half of them do not. It may be difficult to understand how a
human being could exercise even the basic function of thinking without this sort
of interaction. But this is a thought experiment, and it is called that way because
you can hypothesize the craziest things. So, imagine these two kinds of persons
exist. Who will survive in a social world? Who will be the fittest? This is a typical
question that evolutionary biologists ask, and while it has no simple answer in most
of cases it does in our experiment. To help you get a better idea of what we are about
to do, I suggest you think of the docile individual as an altruist, and to imagine the
non-docile as selfish.
The docile individual is open to the social side of human relations while the non-
docile is not; in other words, he/she is capable of experiencing emotions and makes
decisions to please other people. The docile individual is willing to support the cost
of information sharing while the non-docile is not. When individuals decide to put
something in common there always is a cognitive effort underlying that attempt.
Again, the docile agrees that there are rules that regulate social life while the non-
docile has a hard time understanding them and, in most of the cases, these social
116 9 The Docile Organization

rules are secondary to their utility function. To make a long story short, what defines
the difference between non-docile and docile is that the latter understands the social
dimension better. Thus, these individuals fit the social environment while the non-
docile has a hard time doing that. The result is apparent and obvious if we look at it
through the glasses of real life. Although economic models attempt to show you that
there is no room for social beings, the field is moving forwardxi and these models
look like applications of an old nineteenth century ideology. Our experiment is not
concluded yet.
Does the non-docile individual disappear? Stop reading. Look around (I hope
not in the mirror) and think. Have a look at the cover page of the newspaper again;
I am sure you will find something on the top executive, war master, politician, or
man/woman of the day. Do you think that selfish individuals do not exist? Do you
think there is no room for them? Although I dont intend to make hasty generaliza-
tions, there is a possibility you think that one out of the many individuals you find
cited in the newspaper is selfish. In fact, if we want the theory to be consistent with
our experience, we cannot allow them to disappear in our experiment.
The assumptions that we have made lead to a model where non-dociles do
not disappear, show a low fitness, and are part of our hypothetical world. This is
where this thought experiment differs from Simons model, and it is far from social
Darwinism.xii Here comes another critical thought, since sometimes we meet peo-
ple that constantly and consistently behave selfishly (i.e., are non-docile). There
are organizations where the number of non-docile people is particularly high and
persistent over time. How can we explain this phenomenon?

From Evolution to Social Relations

What if Simon presented his idea of docility for purposes other than those that I have
in mind here? What he wanted to do with his later work was to show that docility
is a human trait that allowed evolution to go the way it did. One of the byproducts
of this mindset is altruism, as I anticipated. This way he showed that neoclassi-
cal economics portray an idea of human beings that should be rejected because
it does not match the reality of human evolution. The objective of those papers
was, once again, to show how descriptively poor the economic models have been.
His simple model makes this point very clear. However, and this is my take here,
docility may probably be more useful than that if we use it to understand decision-
making trends in social environments, like organizations for example. Here is the
idea.
Docility is the tendency to lean on other peoples dataxiii to make decisions. This
happens all the time during our lives. When we are children, we lean on our par-
ents and strict relatives knowledge; when we grow up, the circle extends to include
teachers, friends, teammates, etc. The adult individual has a lot of people that pro-
vide data that he/she uses to make decisions. Since this is what happens in every
domain where the individual lives, we need to make some restrictions if we want
to understand and analyze docility: (a) consider a limited social environment (e.g.,
The Docile Individual 117

a company, an organization), (b) focus on social relations. How does our idea on
docility change if the world of our thought experiment becomes an organization?
How can we analyze docility in that case? I believe that the choice of a restricted
environment, such as an organization (vs. society as a whole), and the idea to study
it in strict relation to decision making (i.e., closer to its definition) make it easier to
study and, at the end, enhances its practical use.
Let us focus on what docility is about. We know that it involves a socially based
retrieval and use of data that allows the making of a decision. How does this happen?
It happens through social external resources which constitute the basis for social
relations to emerge. Remember that a social channel (Chapters 7 and 8) is some-
thing that transfers information that somebody wanted you to have. What makes a
resource (or channel, here the two are interchangeable) social is this willingness to
provide data. Mediums may be different, but as far as a social transfer is involved,
we are dealing with a social relation. This happens all the time. You can have con-
tact with another human being or may be using a computer to get information. What
counts from this perspective is that you are leaning on other peoples data, and are
using that data to make decisions. You are being docile. If you do this in an orga-
nizational setting and make decisions that affect your working life, you are being
docile in that environment. This discourse leads to three implications:

(a) we can measure the extent and the degree to which a person is being docile;
(b) docility attitudes may vary depending on the context/environment;
(c) docility attitudes may vary depending on time.

Degrees of docility. The extent to which one person depends on social resources
defines the individual degree of docility. There are many ways to measure this degree
of docility. First, there is a personal belief, or what one person thinks of himself or
herself in terms of how many times he or she takes data from other people, on
average. I tend to believe that people, especially when at work, think of themselves
as open to comments, suggestions, and information, and thus think they are highly
docile. This is a positive attitude to have about oneself, isnt it? However, this idea
may be different from what other people think of the same person. The manager
may think (of herself) she listens to comments, suggestions, etc., coming from her
subordinates, although these people may have completely different ideas.xiv This
contrast leads us to the second measure of docility: what others think of that person.
The environment. The previous argument leads us to define docility in relative
terms. What we experience is that peoples behaviors and ways of thinking vary
greatly depending on their environment. If you will, this derives also from the DC
approach and can be explained through the idea of rationality presented in this book.
Individuals develop different cognitive processes depending on the interactions they
are exposed to. It may well be, for example, that an individual shows docile trends
when volunteering for the Cancer Society, while the same individual is completely
non-docile at work. Moreover, this attitude may change within the same organiza-
tion, depending on the team, project, group of people, leadership, relation with the
118 9 The Docile Organization

boss, subordinates, coworkers, rules and regulations, financial and economic situa-
tion the company is facing, etc. We cannot expect a constant level of docility from
the same individual in different contexts. When I refer to the level of docility of an
individual then, I can see it as average or as related to a given situation.
Another interesting point emerges here. Different organizations usually present
diverse levels of docility. This is what Bardone and I call the docility effect.xv
Organizations, in other words, support docile behaviors when they enhance organi-
zational performance, well-being, working lifestyle, and when they help to smooth
relations, provide support to useful practices, help build up ones self-esteem, lead
to creativity, and the like.
Time. Docility is not only space-based, it depends on time. The fact that we all
use our cognition everyday incessantly gives us a high degree of variation in the
extent to which we would like to use it. Although we cannot stop using external
social resourcesI do not provide explanations here but it is worth consideringwe
can vary the intensity with which they affect our cognition. Moreover, we can vary
the type of social resources used. For example, we can start to gather suggestions
from a new hire instead of continuing to use our usual advisor. Or, we can opt
for computer-based resources instead of hard social interaction. These changes
happen at a given time and support our behaviors. When you are not motivated to
stay in your position at work and you decide that you need to start looking for a new
job, your behavior changes dramatically (even if you tend to hide it). In particular
situations you diminish the quantity, and especially the quality, of social interactions
in the organization you want to quit. Your docile attitudes tend to slow down to a
point where you are no longer docile to those around you. You start thinking of your
new job, and maybe you implicitly think of your new sources of docility.

Active and Passive

What is very strange in Simons definition is that if you stick with it you can-
not describe what happens within an organization or in any social environment.
Consider the last example of an individual that wants to quit the job and look for a
new one. Docility is defined only on the basis of information, suggestion, comments
that the individual receives from other people. Is this sufficient? Do we only receive
data from external sources? In other words, do we only reproject? Or, do we exter-
nalize also? This is how the original concept of docility can be modified through the
distributed cognition approach.
For the concept to be useful and to gain more explanatory power, it needs to
include an active aspect. There is a Latin origin of this word (late 15th Century).
Docility means apt or willing to learn, from Latin docilis, deriving from
docere, i.e., to teach. This is very important for defining what we mean by the
word active docility, since what we are talking about here is the attitude that peo-
ple have to share information with each other. And, as anthropologists explain, this
is a core attitude for human beings.xvi The definition of docility becomes then
Levels of Docility 119

Docility is the tendency to depend on suggestions, perceptions, comments, and to gather

information from other individuals on the one hand, and to provide information on the
other.xvii

It is this active-passive process that defines what docility is about. It is rare or

uncommon that people lean on social channels without exchanging with them. Of
course, it depends on many variables related to the medium, the individual, and the
basis of knowledge transferred. It is apparent to me that this attitude needs to be
considered in its entire process, not simply limited to a one-sided relation.
The argument that could emerge here is whether we stop considering docility
and talk about learning. I believe there is a chance that we end up confused if we do
this. Docility is not learning; it defines a cognitive trait that is apparent in specific
behaviors. While we may say that docility can lead to learning, it is not clear whether
the latter is always related to the former. Moreover, docility is strictly connected to
the DC approach while, as far as I know, there is no theory of learning that is related
to that same approach beside what Hutchins considers in one of the last chapters
of his landmark work.xviii Docility has the advantage of bringing together cognitive
processes (decision making) and human behavior when they are based on social
resources.

Levels of Docility
Using the active and passive framework, we can define different levels of docility
depending on the intensity they have in the organization. The more individuals show
both active and passive docility, the more the organization supports these behaviors,
and the more intense docility becomes within that organization. It is a loop. The
idea that I would like to present here is an analysis of what docility means within
an organization. What are the byproducts of docility? What behaviors stem from
individuals being docile?
This section is dedicated to the analysis of docility in respect to decision mak-
ing in organizations, bandwagon effects, socially responsible behavior, and advice
taking and giving. I believe that the definition of these aspects could help to bet-
ter understand what the contribution of this concept is to rationality and decision
making.

The Prerequisites of Docility

Docility is defined as a decision-making activity where external social resources
are involved in the process. What happens to the organization and to the relation
between individuals when the social environment is analyzed through these lenses?
There are three basic conditionsxix that support the way docility emerges in
any given environment between individuals. These conditions are fundamental to
describing docility in organizations since the stress on the first, the second, or
120 9 The Docile Organization

the third element can change the quality and quantity of socially based decisions.
Docility depends on (a) the community, (b) the standards that are used to encrypt
information, and (c) the public dimension of information.
A community is necessary because docility emerges only if and when individ-
uals share something, this being the place where they live, a goal, a thought, an
ideology, or more. The sense of being part of a community enhances sentiments of
trust and facilitates the sharing of information. Sometimes an organized commu-
nity presents a structure for power where people are supposed to share information
depending on specific rules, loyalty, group dynamics, etc. The underlying idea is
that it is very unlikely people are docile in a community of strangers, i.e., outside of
their world. In other terms, and all other conditions being equal, docility emerges
in social environments where there is something to share: communities. The exam-
ple is a travel abroad. Suppose you are in Chile and people do not speak English.
Besides the problem of communication that relates to the language, you do not feel
part of that community and tend to make decisions based on information sharing
that takes place with people that travel with you, for example. This is because that is
your temporary community and you are being docile with them, not with Chileans.
Let me state it once again: Docility is community-based.
The second condition for docility to emerge is the existence of a standard. This
seems an obvious condition that is the basis of any communication whether it is
social or not. However, by standard I do not mean only the code that we use to
share information; it is not only the language (as in the case of Chilean people).
It is about a specific pattern, behavior, or way of thinking that people of a given
community use to share their information. For example, the community of math-
ematicians is one of the most docile because they work together on the basis of a
shared formal code, which is mathematics. It is not only the language that they
use in their articles, and it is not only the symbols that they operate with. There are
rules that they must follow to demonstrate theorems and to advance in their disci-
pline. They strictly follow a standard, defined as the rules of mathematics. If you
think about this, you realize that there are many standards in our society. Culture is
one of these standards that has many substandards. Docility emerges when people
use the appropriate media, methods, behaviors, and follow the (formal and informal)
rules that facilitate a decision based on information coming from social channels.
In other words, social information conveys and drives decision only when a stan-
dard has been fulfilled. For example, there are many different ways you can use to
give a suggestion to a friend that isin your opinionabout to make a mistake.
In the US culture, it is all right to be straightforward and tell the friend I believe
you are wrong, dont do that! The same sentence pronounced, for example, in a
Mediterranean and/or Arab country has a completely different meaning and cannot
be delivered as I wrote it. Even US culture has different approaches to this simple
point, depending on who the two friends are and where they are. Standards for com-
munication in a bar are extremely different from those you have at home, in a work
environment, or at a funeral. A failure to stay within those standards transforms
into a (potential) lack of docility which is, in turn, a threat to individual decision
making.
Levels of Docility 121

The public availability of information you share is the third condition for docility
to emerge. When people exchange information in a community using the appro-
priate standards, different people can access that information. Of course, it is the
individual that decides to whom the information is made available, but still the
public dimension is concerned. This condition is particularly binding when we
consider an organization. There are many social channels that are used to share
information with a significant amount of people (e.g., annual reports, press releases,
invoices). Some other interpersonal exchanges are less public and may involve a
limited number of people (e.g., confidential communications, board of directors
meetings, communication to employees). The meaning of this public dimension is
that people externalize their thoughts, and this is what makes docility possible. It
is apparent that the public dimension is partly unavoidable, especially if we con-
sider that behavior is something publicly available. However, what counts here is
a public dimension that involves a conscious willingness to externalize itthis is
consistent with our definition of external social resourceand to let others use it as
part of their decision-making processes.
If only one of these three conditions is missing, docility is less likely to
emerge. Docility in organizations depends on the fact that one, two, or all three
conditions/dimensions prevail. The following explains it better.

Docility in Organizations
As I mentioned in the previous pages, we can define individuals on the basis of
the extent to which they show docile behavior (and cognitive attitudes). With this
simple idea in mind, we can isolate people who show significant levels of docility
from people who dont. People that fall into the latter category (non-docile) do not
use significant data from other people when they make decisions, while people that
fall into the former category do. Among docile individuals there is a divide between
people who are only ordinarily docile from those who are exceptionally docile.xx
I believe that we can find these three categoriesnon-docile, ordinarily docile, and
exceptionally docile individualsin every organization; what is interesting here is
fully understanding the implications of considering docility a human trait. The first
step is to better define who these people are.
Ordinarily docile. By definition these people make decisions considering social
channels (specifically other individuals) as a major basis for their choices. They like
to exchange information, to give and receive comments, suggestions, and advice in
general. If you pick a company, any company, these people are the ones that lean on
standardized social practices, rules, routines, and bylaws. Ordinarily dociles (oD)
do not make a distinction between social channels such that any social information
is useful for them to make a decision. The identification with their jobs and with the
company is solid for these individuals such that they tend, on average, to trust stan-
dardized practices and procedures as they are available in the organization. Routines
are what they are very good at. It is apparent that I am describing somebody who
122 9 The Docile Organization

develops docility on the basis of the prevalence of one of the preconditions. All these
three are in play when docility exists, but individuals may stress one more than the
other. This is what happens with ordinarily dociles. These people use docility in its
passive side most of the time; they take information more than they provide it. What
I mean is that oD expresses also active attitudes, but this is a secondary strength
for them. Making decisions on the basis of the information they gather from social
channels is the core of their cognitive and behavioral abilities.
Exceptionally docile. If every community has individuals that show close-to-
average levels of docility, we can also have outliers. This is the case for upper-level
outliers. These people are the ones that show the highest levels of docility in a given
community. Active and passive sides are both present, and they operate at their
best. Exceptionally docile (eD) individuals are the ones who make the exchange
of information, suggestions, comments, and advice the true basis for most of their
decisions. One of the most significant differences is that data are clustered around
them in a way that makes them very knowledgeable and proficient in their jobs. The
information they provide to other people in the organization is of high quality and
particularly appropriate. This is related to the fact that they (implicitly or explicitly,
consciously or unconsciously) understand very well all of the three conditions that
allow docility to emerge and expand in an organization. An exceptionally docile per-
son should provide coworkers with useful comments and insights on what they need
to know, and should listen to their coworkers very carefully. It is worth noting that
I am not writing about accuracy of decisions done by eD people, but of particularly
rich and appropriate decisions. For example, suppose a bank is in the process of
developing and changing the software they use to analyze corporate information.xxi
They need a very limited pool of people to run the transition. These people should
be able to meet the employees (at least the managers, if the bank is large), explain
to them the motives for this change, how the new software works, how to read
the outcomes, how their customers will benefit from this analytical tool, and other
related issues. Most important, these people should listen to critics and comments
coming from the employees so that the software can be changed according to their
users needs and annoying bugs can be eliminated (or reduced). The objective of
these people is to have a better company, an organization that could offer superior
services to customers, and let the employees be confident that the tools they use
fit their needs. This is an example of how exceptionally docile individuals oper-
ate; they make decisions for the sake of the company and lean on other peoples
information.
Non-docile. I already described how these people (nD) behave in organizations in
the previous pages. I can recall here that these individuals are not satisfied with the
companys goals, with their teams goals, and probably with their personal objec-
tives fitting the organization. It is likely that these persons are looking for a position
outside the organization. nDs are not cooperative and do not see any value in shar-
ing information with the rest of the people in the organization. They behave like the
tourist that travels alone and arrives in a foreign country where he doesnt know the
language and doesnt want to know, nor does he want to be there. As a consequence,
he doesnt trust anybody and tends to avoid social contacts.
Understanding Docility 123

Table 9.1 Characteristics of individual docility

Prevalent behavior Prevalent precondition Info quality

Exceptionally docile (eD) Active and passive All three High

Ordinarily docile (oD) Passive Standards, or community, Medium
or public availability
Non-docile (nD) None None Low-null

Table 9.1 offers a summary of what I presented for each type of individual. If
we follow these assumptions, we will find that individuals in organizations make
decisions on the basis of exchanges that exploit social channels. A limited num-
ber of them are exceptionally docile, and a very limited number are non-docile
individuals.xxii
DC supports the idea that individuals are social beings and that social resources
are among the most exploited. This idea of docility also helps define the potential
for the expansion of rationality. Social channels help people make decisions that
better fit the organizational needs and define the extendibility of rationality.

Understanding Docility

A significant contribution to the understanding of docility is given by the description

of what docility is not. This is not an easy task since when you find an interesting
tool (such as docility) that explains some basics of organizational behavior, you tend
to focus on these positive aspects instead of seeing its limits. However, a theory
grows exactly because of the deep understanding of its limitations, as well as of its
potentials.

Bandwagon Versus Docility

Sociologist Mark Granovetterxxiii studied a phenomenon that falls under the name
of bandwagon effect. It is the concept of conformityxxiv to a crowd, to a group of
people, or to any other apparent behavioral or thinking pattern. I already introduced
this concept in the previous chapters; however, here I need to focus on a different
aspect of the same phenomenon. Imagine students sitting in a class and the professor
asks a question. What is the students attitude? If nobody raises a hand or speaks,
it is less likely that anybody will do that. If you are in the class, would you rather
say something or wait until somebody else does it? If you dont care about being
the ice breaker then we would say that your threshold is very low; if you wait until
everybody else in the class has had their say on the point raised by the professor,
then your threshold is particularly high. How many students do you need to listen
to before deciding to speak? This is a measure of bandwagon, i.e., the tendency to
repeat what others do without a specific reason or motive. Bandwagon is very close
124 9 The Docile Organization

to imitation and it does not require a consciousxxv understanding of the fact that
those who jump on it realize that they are doing so.
Is this behavior related to docility? How? I believe that docility and bandwagon
are two different aspects of human behavior, and they are related in that the latter
can be associated with levels of docility. However, I dont see a pattern of causality
between the two: (a) docility may facilitate the emergence of bandwagons but it
is not their cause; (b) bandwagon does not support docile behaviors. In order to
specify these two statements, I need to get into the bandwagon phenomenon more
closely.xxvi
What explains and fosters bandwagons? There are two major variables explain-
ing bandwagon. The first is organizational culture.xxvii A stable and well-
established culture is a prerequisite for imitation to emerge and grow. When change
is not significant or it is perceived as not significant in a given social environment
so that peoples behavior is based on shared and common values and beliefs, then
individuals tend to reproduce the same behavior the majority of the other people are
showing. This is the church effect. There are certain rhythms and regularities that
shape a mass, for example. If you are a non-believer or if you believe in a different
god(s) or goddess(es) and you happen to be in a church during a religious function,
youll do whatever other people are doing: you stand up when they do, sit down
when they do, even sing (if you know the song) and pray if you can or want. You
follow the bandwagon. A mass is made of regularities; regular churchgoers share
the same beliefs, the same values, and have built a common way to live the reli-
gious experience. They have what we should call here a culture. Of course, we are
not giving to this word the full array of meanings that anthropologists do. However,
we are characterizing a specific way of doing things or thought processes that are
shared and define any given community. This church effect is very powerful, and
shall happen to anybody who becomes part of that community, even if it is for a few
hours.
This could also happen in a company or in any organization. When there is a
common and well-established way of thinking or behaving, that is how bandwagon
is more likely to emerge. It doesnt matter if it is a trivial behavior, like having
lunch in front of your computer, always smiling at customers (does anything like
that exist? Oh yes, McDonalds), never talking about personal problems with col-
leagues, or if it defines how people approach a problem, such as the right software
to use, what outcomes to use, from what source comes the most relevant informa-
tion, and the like. Bandwagons can be very pervasive, and this can be related to
the stickiness of a culture. A strong culture may also support change as a major
variable; nevertheless, if this culture of change is strong enough people shall still
jump on bandwagons. A strong organizational culture can be defined as a system of
shared values, beliefs, and rules that affect peoples behaviors, ways of thinking, and
expectations such that it becomes pervasive and somehow predictable. An example
could help explain this better. How hierarchy is perceived and practiced has to do
with organizational (and national) culture. Relations could be based on a friendly
approach to power and authority, where subordinates are allowed to speak frankly
and openly to their bosses and to anybody who is in charge of something. The
Understanding Docility 125

presence of this aspect is related to the way people within the organization think
about this egalitarianism in organizational relations. Sometimes it does not mat-
ter what the corporate bylaws (if any) prescribe on these top-down or bottom-up
relations; it is the openness and actual behavior that helps establish a positive and
supportive climate. A bandwagon is easy to follow with respect to power and author-
ity relations. Especially for new hires, the rule is to do whatever other people are
doing, i.e., follow what seems to be a common and shared pattern of behavior. They
will conform to what they see. When the culture is structured, formalized enough, it
is possible that people do not even think they can behave differently.xxviii
It is worth noting that few studiesxxix have addressed bandwagon at the orga-
nizational behavior level of analysis. There are clues to state what I have in these
pages; however, an empirical evaluation of what I am hypothesizing here has yet
to come. So far, what was written above can be summarized by the following:
The stabler and stronger the culture, the more likely people are to jump on a
bandwagon.
The second determinant of bandwagons is the nature of social relations. This
passive and mindless imitation usually emerges when the imitator has a limited and
superficial relation with people around him. The whole point is that when individu-
als know each other well, they talk, exchange opinions, comments, advice, exchange
information, etc. Especially when at work, and depending on organizational and
national culture, people may be reluctant to share their ideas on a specific behavior
or to ask if something is appropriate. In this case, they tend to adjust their behav-
ior to what seems prevalent.xxx The significant implication of this is that between
friends or close coworkers it is more likely that behavior derives from high levels of
trust, shared ideas, commonality of intentions, and sympathy.
The basis for bandwagon may look similar to what is needed for docility to
emerge. I have hypothesized that loose social relations and strong organizational
cultures foster bandwagons; how does this relate to docility? The two phenomena
lean on similar elements; however, docility facilitates close social relations due to
the fact that it needs some level of trust to emerge and become prominent in any
given social environment. Moreover, it favors the creation of organizational culture,
up to a point. The point being that too much docility may create instability and make
a culture lose its strength. It seems to me that docility is based on distributed mindful
interactionsxxxi between organizational members while bandwagon is not.
In the previous pages I presented what I called the (pre)conditions of docility. Let
me recall them in brief here; they are (a) community, (b) standards, and (c) public
availability of information. At a first superficial look, these three conditions may
seem to work well for bandwagon too. However, the three elements here are always
consciously and deliberately chosen with the individuals docile tendency. The first
element, the community, is not needed when we analyze bandwagon, for example.
The tendency to imitate other buyers choices is very high, but that does not mean
that the two buyers are part of the same community. Imagine that you bought a Jimi
Hendrixs CD because, while at the store, you saw two people buying the latest
remastered edition of Experience. The first person was about your age, the second
was a young kid. Are you part of the same community? Maybe the only thing you
126 9 The Docile Organization

share is Hendrixs music. Does this suffice to call this a community? This same
pattern could be found in organizational behaviors too. Especially in large compa-
nies, the fact that you share the same place where work does not mean a community
is out there. However, to the extent to which there are common interests, culture,
and goals, the organization is a community. It just doesnt seem to be an element
that defines bandwagon since this could emerge even without it. The second ele-
ment, standard, is not needed either. The only significant pattern that imitators are
willing to follow is something that is recognizable. The standard here is not in the
way people communicate or in the appropriateness of that communication process;
it relates to something that is ordinary behavior or way of thinking. It is what peo-
ple do on average; it really doesnt matter if they are following any communication
standard. Imitation can emerge with or without standards. The third element, public
availability of information, is relevant for both phenomena. Bandwagon is based on
the fact that information (i.e., a behavior or thought) is observable, i.e., available.
However, what makes this different in the case of bandwagon is that there is no
willingness to share information. We do not buy a CD because we want to share
our choice with other customers in the store; the willingness is lacking. Docility
is based on the active sharing of data, opinions, suggestions, etc. You listen to a
comment because you want to, and you give an opinion because you want to. The
perspective is completely different.
One last interesting point is that of cognition. While bandwagon has been defined
as a logical fallacy (see Chapter 5), docility is more a strategy to avoid fallacies or
use them in a more meaningful way. The use of social channels does not neces-
sarily imply fallacy avoidance but extends the possibilities and opportunities for
our cognition to operate in a more consistent and successful way or, to state it more
clearly, more rationally. Mindfulness becomes socially distributed and hence subject
to some sort of organizational check when docility is widespread.
At the end of this paragraph, I believe it is apparent that docility is not band-
wagon and is not even close to it! However, the relation we should consider at this
point is that docility can help organizations limit the emergence of bandwagons
(and maybe of other similar fallacies) through mindfulness that becomes distributed
among individuals.

Individual Social Responsibilityxxxii

Social responsibility is one of the most widely discussed, analyzed, published, and
practiced ideas that has recently hit the ground of management. To be precise, the
idea is not recent at all, but it received a full consideration in recent years. This
doesnt seem a fad because it is backed by deep social and economic concerns on our
uncertain future. The attention has always been over corporations.xxxiii In fact, the
acronym that is used more frequently is CSR, or corporate social responsibility.
Focusing on individual social responsibility can be interesting and unveil impor-
tant circumstances under which a responsible behavior finds an explanation.
Understanding Docility 127

Managers, not corporations, sit in the company boards, speak at meetings, go to

work, take responsibilities, have duties, and ultimately make decisions. Therefore,
emphasis on individuals seems to be well-placed, at least. I am not arguing that this
focus is new or unusual, but it is not prevalent. While socially responsible behaviors
are growing, a cognitive approach to social responsibility is lacking. This is what
this section is about: Is there a way to find evidence that an individuals cognition
supports social responsibility?
As the reader may guess, docility is my answer. How? Social responsibility
could also be defined as the tendency of individuals to establish or maintain
cognitive advantages from the social resources (channels) that they exploit more
frequently.xxxiv Otherwise stated, social responsibility helps individuals keep and
reinforce connections to their most exploited (or relevant) social channels. What is
a cognitive advantage? It can be defined in a relative way, through a comparison
between ones cognition and that of the others. A cognitive advantage may be also
defined in terms of better performance and results. It is a procedural and substantive
enhancement that can be achieved if we exploit external resources in the most appro-
priate way. This means that external resources need to be tied up to ones cognition:
A resource is not good for everybody, and different resources may lead different
individuals to similar outcomes. Therefore, a resource is useful when it provides the
individual with an advantage; otherwise, people tend not to use the same resource
again (unless they are forced to do so). When comparing types of resources, we have
to admit that some social resources need a special care if we want to use them more
than once.xxxv To explain this better, a brief inquiry on what responsibility is might
be helpful.
As already stated above, external social resources have been defined as they relate
to other individuals willingness to share information, while non-social resources do
not show this relation. We used a tree before to exemplify the second, and we can
use a newspaper article for a resource of the first type. Of course, the idea of a
tree or of a river that is held in our mind is significantly influenced by its socially
construed representations. However, it is not (to some extent, at least) a product of
human intervention, production, shaping, etc. It is not a work of art, it is Mother
Nature. In turn, a social resource/channel is something that a tree or a river is not. It
is a mediator of socially-based information, in which the sender actively gives that
communication, and the receiver actively takes it.xxxvi The example of a newspaper
article serves the purpose of showing the difference. An external resource becomes
social only if and when a reader makes use of it. Advice giving and taking offer a
better example (see below).
The short recap on social and non-social resources helps us answer the question:
How could responsibility lead to a cognitive advantage? Responsibility can be seen
as a reinforcement mechanism employed in the exploitation of social channels. A
short example may define this better. Consider the journal article and think of its
author. Imagine that the journalist decides to write false information for the only
purpose of gaining visibility. Due to the professional code, this person faces the risk
of being fired right away if someone finds out. However, this is not the worst case
scenario. In fact, he may face a dramatic exclusion: Readers may label the journalist
128 9 The Docile Organization

as unreliable and stop reading his articles. Otherwise put, the social channel ceases
to be exploited because of the journalists irresponsible behavior. When irresponsi-
ble behavior is perceived as repeatable, it makes the use of the social channel less
likely. This is how responsibility may be intended as a reinforcement of the use of
any given social channels; this, in turn, means that distributed cognitive processes
are reinforced, or more stable.
Now, and from a more detailed perspective, what is responsibility? To take
responsibility for something is equivalent to the expression to have a duty in
regard to something. This definition may be sufficient for general purposes; how-
ever, responsibility may be internal or external. The common way to think of
responsibility is internal. This is the case of somebody taking responsibility for
an action, a thought, or something else. External responsibility is a subtler con-
cept as it can be associated with self-deception or bad faith. This happens when
self-deception, or bad faith, creates a situation in which human beings relinquish
freedom and externalize responsibility.xxxvii A person in a condition of bad faith
deceives himself by constructing a limited reality that does not take into account
the full range of choices available to him, and this, alas, is a condition in which
many people live all their lives. It is from himself that he is hiding the truth; the
deceiver and the deceived coalesce into a single consciousness in a way that must
be distinguished from true mental illness or malfunction of consciousness.xxxviii
Therefore, the idea of social responsibility can be redefined as internal only or, on
a negative angle, as bad faith avoidance. Moreover, this approach suggests that social
responsibility enhances and extends the range of choices that individuals have. On
the contrary, those who lean on external responsibility are more limited than they
think. In supporting the exploitation of social channels, social responsibility seems
to be connected to those individual attitudes that we have called docile behaviors.
It happens that this appears to be a reinforcement or, better, a byproduct of docility
attitudes.

A Theoretical Framework for Advice Giving and Taking

The definition of docility includes, but is not limited to, advice taking and giving.
This is a trivial point since it is part of the way we defined docility. However, I
believe that docility does not explain the why of advice taking and giving; it explains
the how. Individuals take advice because this is the way their cognition works: they
lean on external resources.
First of all, advice is defined in the literature (and in this text in particular) in
a very narrow and strict way. It is a recommendation that an individual offers to
another individual, and it contains a specific suggestion on a decision. Advice is a
suggestion on how to make a certain decision. This makes it a very specific kind of
social interaction. Suppose the decision maker (judge) is about to send an e-mail to
the boss with a recommendation on quitting a product line that is not selling well.
An advice from a coworker is of the type do it or dont do it, while another
What Is a Docile Organization? 129

source may be more vague. For example, it may happen that the judge receives an
e-mail with data on people losing their jobs because of the cuts in the companys
product lines; or, the e-mail may be from the boss and suggests the judge look at
new data on the quality of those products. It is apparent that docility explains advice
taking-giving, as well as other information that helps people make decisions. From
this angle, the study of docility is a general theory of decision making in social
environments.
There is no general theory of advice. I suggest that docility could offer a frame-
work where we can put advice giving and taking together with all of the other
byproducts of docility.xxxix However, these processes dont seem to be byproducts
over part of what docility is, at least at first sight. The question that needs to be
addressed is whether the decision maker has to be docile in order to take and use the
advice. Or, is a person that simply exploits others expertises, knowledge, or posi-
tions for personal and selfish purposes a docile individual? Or, when the purpose of
taking the advise is not that of using it but that of pleasing the advisor, does this still
fall under a typical docile behavior?
The reasons why people take advice can be very complex. For example, when
an individual is obsessed with a specific and single goal, lets say career advance-
ment, he or she could do anything to achieve that objective. Among the strategies
that one can exploit, there is the one that includes a fake reaction to advice coming
from people in power. To please these people that are in charge of making decisions
on the judges career, the judge can take the advice even when he or she thinks it
is not good. Once again, is this person docile? Docility is about using information
from social channels to make decisions; however, there is trace of purpose, moti-
vation, or of other psychological and cognitive variables that distinguish non-docile
individuals from dociles.

What Is a Docile Organization?

Scholars scarcely employ Simons concept of docility. This, as discussed already at
the beginning of this chapter, is based on (a) its vague link to bounded rationality,
and (b) the fact that it needs a strong cognitive approach to support it. In this chapter
I have presented a more accurate and specific definition of docility and linked it to
the distributed cognition approach and rationality.
One of the most important implications of what we have examined in the pre-
vious pages is to consider whether organizations take docility as part of their
structure. What I mean here is that if the large majority of people are docile and
show docile attitudes, then the organization may include facilitating mechanisms
that support and foster the emergence of these behaviors. This is what I mentioned
above and falls under the name of docility effect. Organizations are complex social
environments that function on the basis of, among many other things, individu-
als exchanging relevant information and making decisions on the basis of these
pieces of data. It is likely that organizations provide individuals with appropriate
130 9 The Docile Organization

social channels where they can be docile. These structured mechanisms may fall
under rules and/or under informal or behavioral norms. A few examples could help
understand what I mean by this.
The airway companies, supported by the International Civil Aviation
Organization (ICAO), have specific rules for their pilot-copilot communicationsxl
during emergency operations. The general rule is that the pilot must listen to what
the copilot has to say, and act accordingly. Pilots get trained on this. Of course, expe-
rienced pilots with the largest number of flights could come out to also be those that
are right most of the time. However, studiesxli on the outcome of situations where
this explicitly docile behavior was not taking place have been a disaster.
I am not an expert on risky surgeries, but I imagine there must be a procedure that
helps surgeons get support and advice from other individuals in the room. This is to
limit the risk of doing the wrong thing. The team is involved in making a decision
that could save a life. The final decision is at the expense of the one who makes it,
though opinions have to be expressed.
Information and communication technologies (ICT) help companies develop
docility. Many companies have a sort of in-house chat line where employees can
exchange messages on their jobs or on particular projects they are working on. This
facilitates the number of interactions between people and induces the exchange of
information, some of them useful to decision making. The corporate dedicated
chat line is an example of a modern social channel where individuals easily get their
chances to take other points of view under consideration. Although it is not very
easy to get sophisticated advice with the chat line, it is possible to exchange rel-
evant and sensitive information that could help the decision-making processes. It
could also foster a team or group decision making when the final decision is shared
by the majority or all of the members of a team.
On a different take, but close to ICT, are corporate cell phones and advanced
walkie-talkies. Here the interchange varies depending on the intensity with which
people need to communicate with each other. However, these are channels through
which individuals get information, comments, and suggestions that help them to
make decisions.
What I mention here are communication tools that become docility enhancers
or facilitators to the extent that the individual use them to make decisions.
Communication does not necessarily lead to a decision-making activity. Therefore,
the point here is not that the company using these communication tools is docile,
but that these tools favor and push the probability that docility emerges among
organizational members.
Docility can also be related to a specific behavior that is supported by organiza-
tional culture. It is the case of cooperation. Depending on national cultural variables
and on the way these are integrated in the organization, individuals may have the
tendency to work in isolation, to cooperate with all employees, or to cooperate with
coworkers only. This tendency may also depend on industry standards and individ-
ual personality traits. However, all other variables being equal, the organization may
push people toward more or less cooperation. The last case, that of cooperation that
is not affected by the power structure of the organization, is very important because
Notes 131

it depends on behavior. You cannot force people to cooperate if they are not willing
to do so. The result is a cooperation that results in limited outcomes that do not
exploit its full potential. In other words, those with the higher managerial positions
have to show themselves willing to cooperate with everybody else in their team. If
the perception of their attitudes is not cooperative, then people shall not cooperate
in any meaningful way. They face the probability to be let down by their boss, for
example. When individuals show their attitude to cooperate, it means that they are
being docile and inviting other people to do the same. And this is another aspect of
how a docility effect can be achieved.
The organization can integrate docile attitudes in its structure and enhance
peoples docility tendencies. This, in turn, makes the whole organization docile.

Summary
This chapter has introduced and explored the concept of docility. First we have
defined the conditions that allow docility to emerge: (a) community, (b) communi-
cation standards, and (c) public availability of information. Second, we have seen
that docility can be used to define different roles and tendencies of individuals in
organizations. Third, docility has been related to social responsibility, bandwagon,
and advice giving and taking. Fourth, I examined how the docility effect may define
a docile organization. We are now ready for some concluding remarks.

Notes
i. See Humphrey (1976).
ii. Dunbar (1998) and Dunbar and Shultz (2007).
iii. Richerson and Boyd (1998, 2005).
iv. Simon (1993b, p. 156).
v. Simon (1993b, p. 156).
vi. See Maturana and Varela (1987), for an example of this. Recently, the theory of niche-
construction builds on some inconsistencies of a strict Darwinism (Odling-Smee et al.,
2003).
vii. It is A Theory of Docile Society, Secchi (2007).
viii. See for example Odling-Smee et al. (2003).
xi. This was a first version of the paper on super-docility; Secchi and Bardone (2009b).
x. The original model by Simon (1993b) presents a set of three equations that define fit-
ness. In particular, they define the fitness of the selfish (fS), that of the unintelligent
altruist (fU)individuals that do not discriminate with whom to be altruistic, and of
the intelligent altruist (fI). The equations are
fS = fn + faI x qI x cI + faUx qU x cU
fI = fn + fd x dI + fal xqI x cI + faU x qU x cUc x cI
fU = fn + fd x dU + faI x qI x cI + faU x qU x qU x cU x cU c x cU

Parameters are defined as follows: fn is normal fitness; fd x dI and fd x dU are incre-

ments of fitness for docility; faI and faU are increments in fitness from others altruism;
cI and cU represent the extent to which I and U are altruistic; dI and dU denote the
132 9 The Docile Organization

abilities of I and U to benefit from docility; qI and qU are percentages of I and U in

the population. The cost of altruism is c (Simon 1993b, pp. 157158). If we attribute
the following values to parameters, fn = 1.01; fd = 0.02; c = 0.005; faI = 0.01;
faU = 0.005; qI = 1/3; qS = 1/3; pU = 1; dI = 2; cI = 0.8; cU = 1; dU = 1. (Simon,
1993b, p. 158), in 30 generations, altruists reach 72% of the population while selfish
decrease to 18%. In 721 generations, society has only intelligent altruists (or highly
docile individuals; Secchi, 2007).
xi. The most interesting steps are those of behavioral economics (see notes on previous
chapters). However, there are other parts of the field that move in equally interesting
directions; see Frank (2004).
xii. Details of the model summarized here are in Secchi (2007).
xiii. Data here stays for comments, suggestions, recommendations, and information.
xiv. I believe that techniques developed by Social Network Analysis (e.g., Knoke and Yang,
2008) are particularly helpful to structure a study on docility in organizations.
xv. Bardone and Secchi (2006, 2009), Secchi (2007), and Secchi and Bardone (2009b).
xvi. See Humphrey (1976), Dunbar (1998), and Dunbar and Shultz (2007).
xvii. Secchi and Bardone (2009b, pp. 347348).
xviii. In Hutchins (1995), there is a chapter on learning that explicitly connects the DC
approach to this cognitive activity.
xix. What I present here is an elaboration of findings published in Secchi and Bardone
(2009b). See also Bardone and Secchi (2009).
xx. This is what is formalized in Secchi and Bardone (2009b).
xxi. This example comes from a real case and I witnessed it happening while working as
consultant for a bank.
xxii. The result comes from the model analyzed in Secchi and Bardone (2009a).
xxiii. Granovetter (1978).
xxiv. A bandwagon effect emerges when the demand for a commodity is increased due to the
fact that others are also consuming the same commodity (Leibenstein, 1950, p. 189).
This happens when consumers purchase a commodity in order to get into the swim of
things; in order to conform with the people they wish to be associated with; in order
to be fashionable or stylish; or, in order to appear to be one of the boys (Leibenstein,
1950, p. 189).
xxv. Fiol and OConnor (2003).
xxvi. In a working paper with Bardone and Secchi (2009), we elaborate on bandwagon and
docility and present a formal model and a simulation to study the conditions that support
these phenomena in organizations. A second paper is forthcoming and takes one step
further in mathematical modeling.
xxvii. See for example, Schein (1990, 1996).
xxviii. The stability of a corporate/organizational culture is also supported by a positive social
and economic performance. When everything goes extremely well or even fine it is
difficult to think out of the box and to challenge established and common behavior
patterns. On the contrary, when a crisis emerges and the organization is challenged by
financial shortcomings or the management is threatened by external stakeholders, it is
easier that criticism and deviant (from the ordinary) behavior emerges.
xxix. Organizational behavior studies focus on the so-called macro aspect of the phenomenon
so that interest is in the processes of imitation between companies (e.g., Abrahamson
and Rosenkopf, 1993, 1997). A mix of micro and macro analyses can be found in Fiol
and OConnor (2003). Sociology offer insights on micro-organizational behavior (see,
for example, Chiang 2007).
xxx. I am trying to describe tendencies, in any case I believe that this is what always happens.
xxxi. The word distributed is very important. Docility is an individual trait but also a commu-
nity property (that comes from its preconditions). Therefore it is the network of relations
Notes 133

that supports or rejects docile behaviors. There is no mindfulnessi.e., conscious ratio-

nal behaviorwhen the organization (other human beings, the community) does not
support this cognitive activity. It is intuitive that one individual cannot be continuously
mindful for a long period of time. That individual needs social support to be as mindful
as possible. This is when organizational docility comes into play (see below in the text)
and this is also the difference between what is in this book and what Fiol and OConnor
(2003) write in their work.
xxxii. This paragraph summarizes my paper The Cognitive Side of Social Responsibility,
Journal of Business Ethics (2009).
xxxiii. This is apparent from literature reviews: Garriga and Mel (2004) and Secchi (2007).
xxxiv. Secchi (2009, p. 575).
xxxv. This makes also the difference between non-docile and docile. Non-docile individuals
dont care about exploiting the same social channel more than once since they dont have
a socially oriented (or pro-social) mind. Their attitude can be thought of as a response
to immediate needs, to a focus on the short term. Preservation of social relations is not
their objective. This provides a micro-explanation of the reason, according to orthodox
neoclassical economists, social responsibility finds no place in their theoretical system.
xxxvi. Secchi (2007, p. 575).
xxxvii. Magnani (2007, p. 129).
xxxviii. Magnani (2007, p. 131).
xxxix. It is not in the economy of this book to analyze the byproducts of docility. However, a
short list may include cooperation, altruism, social responsibility, advice taking-giving
(Secchi, 2009).
xl. I took this from Gladwell (2002).
xli. It is, again, Gladwell (2002) that offers a vivid and effective reconstruction of this point.
Chapter 10
Conclusions

Here we go! This is the final section of this book on rationality and decision making.
After this tour of my thoughts on some of the most important aspects of individual
and organizational life, the final question that every author has to address at the end
of his/her work is, What have we accomplished?
This is not an easy question, especially for a book that attempts to merge a sig-
nificant amount of otherwise scarcely connected models, approaches, and theories.
Since what I presented here is a theory of human behavior in organizations, the
result is based on implications of considering the distributed cognition approach
and docility as key components of decision-making processes in organizations.
There are different sets of arguments that we can address in this concluding sec-
tion. Although these remarks provide insight on the material in this work, none of
them is a true conclusion strictu sensu. They serve as guidance for future research
to me and to anybody who is willing to join this exciting thought adventure. The
first part of this chapter is dedicated to the exploration of past studies in the light
of the distributed cognition approach and docility. The second section questions
the appropriateness of the old distinction between group and individual decision
making. Third comes a methodological discourse, and finally an overview of the
meanings of the proposed redefinition of rationality.

The Point on Rationality

Among the achievements of this book is the idea of extendable rationality. This
concept is based on the fact that the two limits defined by Herbert Simon do not seem
to hold if we switch from a separatist or isolationist to a distributed paradigm.
Rationality is bounded because of limited access to information and limited compu-
tational capabilities. The former is the external, and the latter is the internal limit.
If this divide falls apart we can think of rationality as based on the communication
(interplay) between internal and external limits. From here, I have argued that the
results we get present a completely different picture of individual rationality. What
emerges from the break of this old inside-outside barrier is a new rational being,
capable of modifying the limits of rationality depending on how these two bounds

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 135

DOI 10.1007/978-1-4419-7542-3_10,  C Springer Science+Business Media, LLC 2011
136 10 Conclusions

are coupled. The external is not external anymore, it is part of cognition and defines
how rational human beings can be in their decision-making processes. Rationality is
therefore not stable, it is not defined once and for all for each one of us. It depends
on how cognition changes itself depending on this internal-external interplay.
Rationality changes as time passes by, as availability of new artifacts or external
resources emerges, as fitness to the social environment grows.
Our rationality is time sensitive. This is easy to understand if we think of
ourselves or if we observe how people change during their lifespans. Cognitive
processes of older people are not the same as those of a child; we can argue that
rationality changes as well when it is confronted with age. However, if we com-
pare two individuals, one can always argue that their limits are different because
of their internal build-up rationality. In fact, for this thought experiment to be of
some value, we can look at ourselves. How has your cognition changed since you
were a child? Consider an individual who is a PhD student right now: How do you
think his cognition changed since he was a child? What we do all our lives is learn.
Some learning mechanisms are more accurate and work better than others; how-
ever, it is this process that modifies the way we develop our rationality. Moreover,
it changes the way we make decisions. This is not only true when we consider
the evolution of our cognition over time, it is particularly significant for specific
decision-making processes. Individuals dont make decisions at a definite moment
in time. A decision may take hours, days or even months (sometimes years) to be
made. During this period new resources may become available, new insights on
the problem may appear, variables can acquire different meanings, etc. We become
acquainted with external resources, how to use them, how they affect our cognition,
and it is this experience that lets us change the way we make decisions. Sometimes
everything happens in a limited period of time where the decision process changes
through manipulation of external resources, or through doing. And this makes our
rationality extendable over time. Extendable means that it has potential to move,
that bounds are not fixed or stable; whether it does so at its higher potential or not
depends on the integration between internal and external resources. This leads to the
second variable.
The process through which resources become available is worth analyzing as
far as cognition is concerned. The process through which individuals learn how to
exploit these resources is ongoing. A decision process made in isolation then means
that the individual has no contact with anything (e.g., visual, touch, hearing) which
makes it perfect from a neoclassical economist perspective, although impossible.
Isolated decision making is something that communicates directly and internally
with the brain only. I am not aware of any common practice that allows anything
like that. In a real decision-making setting, rationality could change in relation to
resources that become available at different times. It may be a piece of information,
an advice, a new thought on a previous comment (re-projecting), an activity that you
perform (e.g., write, read, speak, behave in a specific way). These activities move
the bounds of individual rationality so that a solution becomes more apparent.
The third variable that helps explain the extendibility of rationality is the fitness.
A decision and its process are rational if they fit a specific social environment. This
What Are We Mapping? 137

is to say that individuals need to use their docility to be rational. Implications of this
are socially responsible behavior, cooperation, altruism, advice taking and giving,
and the tendency to avoid bandwagons (whenever possible). All of these processes
help individuals make decisions that work in a given social environment.
It must be clear by now that I am not presenting any theory of unbounded or full
rationality. The whole point can be reduced to the fact that rationality is not limited
in the way Simon (and many others after him) thought it was. It is not limited by
internal and external variables; it is limited by a composition of these two sets of
variables, where what is internal and what is external is not exactly definite ex ante.
Decision makers overcome limitations when they exploit external tools so that their
cognition adapts to them, and they can make a decision. This process highlights the
fact that what makes human beings so peculiar is not the fact that they are bounded.
Quite the opposite. We are exceptionalcompared to other animalsbecause we
can modify these bounds through their manipulation; we can extend the potential
of our rationality and cognition depending on this interplay. Our rationality is not
stable, our limits are relative to specific conditions and situations. And even when
we can define these limits, they may change very fast due to interaction with exter-
nal resources. This instability is what makes our rationality extendable, adaptable,
plastic, and modifiable.

What Are We Mapping?

When we consider the experiments that I presented in the first part of this book
(Chapters 4 and 5) then, what do they look like? If bounds are not the most important
variable that define human beings, then what do these experiments really measure?
Is it accurate to say that they present maps of our bounded rationality? Or, are they
mapping something else?
A typical outcome of studies that map bounded rationality is to show how old
paradigms dont work. These are very important results when we think of what it
takes to abandon well-established scientific paradigms. Studies of bounded ratio-
nality have shown us that the idea of a fully rational individual has very limited
explanatory (descriptive) and normative (prescriptive) power. However, these stud-
ies are oriented towards the past; anomalies of the past paradigm are what they
are concerned with. Moreover, students of bounded rationality are not interested
in defining rationality. It is rare to find inquiries on what rationality means, on
the meaning of bounds, or on the assumptions that hold when embracing BR.i
When I presented the maps of bounded rationality, I have intentionally hidden these
points since scholars are mapping something they forgot to ask about. The con-
cept remained was that of Simon as he defined it more than 50 years ago. Anyway,
what if rationality is not bounded the way students of BR think it is? What are they
mapping then?
I believe that it is possible to have a new reading of these experiments and of
their findings. Of course, for some this process is easier than for others since they
138 10 Conclusions

were not set up to measure change in rationality bounds. Take the San Antonio-
San Diego experiment.ii In order to show how heuristics work, researchers asked
European and American students which city is larger between San Diego and San
Antonio. Answers vary depending on personal knowledge, experience, and the fact
that you have heard about one city at least. The point is that if you dont know the
answer, you go with the guess that is close to the only clue you may have (e.g., heard
of San Antonio/San Diego through movies, friends). This simple experiment shows
that individuals are not stuck in their position, but they adjust their perspectives
depending on the level of interaction they have with the external resource. Results
show how people move their bounds when they have no or limited clues on what
the right answer is. Of course, in a nonexperimental setting this aspect falls down
in a second as Web search engines can be exploited. When these are not available
then advice can easily come up as a resource.iii When nothing like that is avail-
able, then we are defining a situation that is very difficult to become real. What is
that study really analyzing? What rationality is that of not asking for further infor-
mation when needed? It is a rationality that shows its adaptability to a particular
(extreme) situation; researchers were showing how rational individuals cope with
uncertainty remodeling their rationality to fit the needs of that moment with avail-
able information. Bounds were moving. The proponents of this experiment discuss
about bounded rationality as an adaptive toolbox, and this is probably a link we
may find between their findings, distributed cognition, and the idea of extendable
rationality.
We can easily reinterpret many other experiments illustrated in the previous
pages of this book through an extendable rationality perspective. And I believe that
this rereading makes more sense than the original view of BR simply because (a)
it includes it and (b) it forces focus on the plasticity and malleability of human
rationality. Researchers were mapping exactly this aspect of human beings: the
adaptability to a different set of constraints in the use of external resources. They
just didnt know it.

The Individual and the Group

One of the ideas that remained latent in this work is that of groups. I explicitly
assumed that individuals live in social environments, where organizations are the
most diffused and effective places where their lives are spent. However, many of us
work in environments that are not composed of thousand people that participate in
large organizations. Every one of us has his or her own social niche composed by a
limited number of other individuals that help to make decisions.
A significant amount of the literature on decision making is thus dedicated
to team and group decision making. These studies usually stress how different
the individual is from the group-related decision process.iv They also highlight
how organizational decision making can be analyzed more properly if groups and
teams are considered. Members of organizations, whether they are workers, middle
The Individual and the Group 139

managers, shop managers, top managers, or members of the board of directors, do

not make decisions in isolation. The more complex the decision, the less these peo-
ple tend to act as if they are alone. As I have specified in the previous pages, there
are mechanisms that prevent people from making decisions without involving other
people.
I think these studies are particularly relevant for the study of decision making,
and I would like to be provocative in pushing the argument forward. Is the group-
individual distinction still valid? How individualistic can an individual decision
maker be in organizational settings? The use of external social resources is but a
marginal part of the way individuals make decisions. Our cognition is social, our
rationality takes advantage of this intertwined relation between social channels and
decisions. Decision making in organizations is a social process. There are two major
reasons.
First, even though we have stated (more than once at this point) that individuals
lean on external social resources and that they exploit social channels to make deci-
sions, we havent clearly specified what makes the difference between an individual
doing that as an organizational member or as an isolated individual. Is rational think-
ing within organizations different? Why? One of the most significant variables that
makes the organizational setting different is the pertinence of the social element that
affects each of the following:

Social roles. Every member of the organization plays a role that depends on posi-
tion, authority, and power.v This role is usually defined by social norms and values
that are specific to the organization. There are two aspects that help to define indi-
vidual social roles: leadership and hierarchy.vi The latter is the structure through
which power is formally established in the organization, while the former relates
to individual capabilities to be respected, authoritative, find support, affect other
peoples thinking and behavior, share values, and lead. Sometimes the two stay
together, sometimes they do not.
Culture. Organizations develop specific values and rules that can be partly writ-
ten in codes and lean partly on behavior.vii This culture is related to the national
culture, but it is also original and dependent on its members. Depending on orga-
nization size, subculturesviii may develop and define behavior and thinking for a
limited group of people that, for example, work together or meet with some sort
of continuity, or that establish specific and long-term social relations.
Goals. Organizations have goals. The usual explanation for the existence of orga-
nizations is that they exist to accomplish goals that cannot be achieved by a single
individuals effort. Another explanation is that organizations are more efficient
than markets when it comes to acquiring specific resources (e.g., labor) and let-
ting them work. Goals vary depending on their importance so that we may have
a hierarchy (primary, secondary, tertiary, etc., or goals, sub-goals, sub-sub-goals,
etc.,). They define everyday activities in organizations, and also help to under-
stand the organization of labor and responsibilities in organizations. The same
goal can be achieved through a multiple array of different strategies; how formal
140 10 Conclusions

these strategies are vary as well depending on organizational culture, social roles,
and docility.

Students of organizations add more than these factors. When they define orga-
nizations, they also add the environment, production process, and participants.ix
However, my purpose is not to define an organization but to understand what makes
it different for making decisions in that setting. Therefore, I am not disregarding
the importance of organizational elements as they emerge from decades of stud-
ies, I am only trying to state their relative importance in relation to my arguments.
This probably results in a clear-cut distinction or dismissal of some of these orga-
nizational elements. A short explanation of why I disregard these three factors
may be helpful to get the point. I dismiss the first just because having an envi-
ronment is not typical of an organization, since everything has an environment.
Everything we can think of exists in a given environment. Moreover, according to
the approach we have introduced here, a clear-cut distinction between the organiza-
tion and its environment is not that relevant: The organization is what it is because
of its environment, or because of processes of interaction with the environment.x
The production process may well be something that happens at the individual level
also, e.g., writing a report is a typical individual production process (of course, I
should include resourcese.g., computerthat the individual uses), and therefore
it does not define organizations only. The third element, participants, is somehow
implicit in my arguments and cannot be regarded as typical of an organization:
Unorganized groups of people have their participants too. I believe that the dif-
ference is in the social role (formal and informal) that each member has in the
organization.
If all of these conditions hold and if we integrate them with docility then, social
relations become the basis for individual decision making in organizations. A group
is defined by individuals that establish social relations on the basis of a formal and/or
informal structure. Organizational groups include the three factors listed above and
specific channels to facilitate the emergence of docility. Otherwise stated, decision
making is a social phenomenon.
The second reason why groups are fundamental for the study of decision making
in organizations is that individuals think of consequences of their actions. A decision
may be successful in terms of goals, but sometimes the decision maker prefers not to
pursue it to its end simply because it can be harmful for the group. In other words,
a decision may be particularly good for the single individual but harmful for the
relations established in the group. This also includes communication issues, i.e.,
when a group member thinks she has to communicate to the upper levels but prefers
not to do that in the interest of the group. I am not arguing that this is what always
happens, I am just emphasizing that these are social phenomena that constitute part
of usual organizational life. There are multiple loyalties in an organization, and these
play a significant role in how individuals make decisions. Is it the R&D department
that counts the most or the people you work with? Is it the organization as a whole
or the department where you work? The closer the social relations, the less likely is
an individual dissonance in the decision-making process.
A Methodological Note 141

As it may be apparent by now, this is another aspect of docility that we have

called social responsibility in one of the previous chapters. More than on a logic
of consequences, the individual behaves on the basis of his or her responsibility to
the social environment (group) he or she partakes. As we know, this is a logic of
appropriateness, together with one of adaptiveness.
This is a short explanation of the reasons why I suggest that the study of indi-
vidual decision making in organization should be intertwined with that of group
decision making. We cannot understand the former without the latter, and vice
versa.

A Methodological Note

The study of organizational behavior has made dramatic improvements in the

last few decades. One of the points that always troubled me was the almost
null advancement on the idea of bounded rationality. The recent cognitive
revolutionxi shed a new light on the field and added insights that otherwise
would have remained only marginal (e.g., shared cognition processes, sense-making
approaches).xii When this happened, I started to think that the attribute behavior
was close to its end in this field, and that it was about to be replaced by the term
cognition.
This didnt happen. What happened was that the old OB division of the
American Academy of Management had a spin-off and generated the Managerial
and Organizational Cognition (or MOC) division. The old paradigm didnt give
up and generated a new (and smaller) group of people interested in cognition
that probably believes this is the future of organizational studies. I am not inter-
ested in recriminations or in making a proposal for name changes (in fact, I am a
member of both divisions!) but it is apparent that mainstream management failed
to embrace the new that was coming. Why? Thomas Kuhn explains this better
than me.
With the advent of cognitive science in the organizational behavior field, we can
ask if behaviorism is definitely dead or if it is still alive. While I believe it is in
very good shape, I also hope that the field starts moving forward. The question is
one of motivations that bring people to behave the way they do. It is not only a psy-
chological approach that we need, it is a cognitive one: Do we still need/want to base
our analysis on behavior and overlook the inner core of human mental processes?
Isnt thinking coming back of age? What is behavior without an explanation of how
our cognitive processes work?
More is needed. In the recent years a new and very promising field was born:
neuroeconomics. Students in this field explore individual choices mapping their
brain activities; it is an extension of what is the field of behavioral economics. It is,
once again, opposed to mainstream economics and its should approach to science;
that really looks like a product of the nineteenth century now. The term behav-
ioral in economics stays for new and evidence-based as opposed to old and
142 10 Conclusions

theory-based. I suggest that this may be the same that should happen in the OB
field with cognition and behavior. However, both approaches are related to manage-
ment and organizational practices, and they complete and integrate each other more
than they offer contrasting views of the individual (as the old and new economics).
I hope this work helps in this shift toward the neurobehavioral approach to the field
of organizational behavior and management.
There is a second methodological point that I would like to mention, although
very shortly. Scientific fields, as many of human activities, are not exempted from
fads. I do not write this because I hope that the use of cognition is not such a thing
but because I believe that the use of systems theory in organizational studies has
been a fad. Systems theory emerged as a field soon after World War II and left
two branches of studies to the scientific community: (a) systems theory and (b)
cybernetics. The first is the study of interrelations between components that make
every system work and of relations between the system and other systems (known
as system-environment relations). The second is the study of control systems in
animals and machines. While the second did not abandon the common scientific
approach, systems theory presented a completely different and, at that time, new
approach to science.
The scientific approach and its methodology is reductionist. It consists in decom-
position of a given phenomenon in its parts and analyzes these in isolation. The
whole is a simple sum of component parts. For example, the study of a molecular
structure is defined through the atoms that compose it, as well as a study of the
emergence of crime in society can be the result of a study on individual behaviors,
attitudes, education, and so on. Systems theory thinkers had another approach to
science. They believed that the whole is not a result of the simple sum of its com-
ponent; there is something that cannot be explained by this simple sum function.
Individual parts are different from when they are together with the other parts of
the system. This is apparent in organizations where, for example, culture emerges
only in relation to its members or, for example, when the head of the finance
department is defined through her team, coworkers, leadership, authority, and every-
thing that is organization related. This is the reason why organizational scholars
adopted a systems theory approach. To have an idea of how significant this approach
was (maybe still is) for organizational studies, one can read Richard Scotts
classification.xiii
My question, then, is if we agree that systems approach is relevant when it
comes to studies of organizations, why does reductionism prevail? With no excep-
tions, studies on heuristics, biases, emotions, ethics, decision making, and cognition
in organizations follow a perfect reductionist schema. This is a major weak-
nesses of experiments and surveys that provide the basis on which our theories of
organizations are based.
I believe that the distributed cognition approach is a contemporary attempt to
let the system effect re-emerge in our studies. Docility is also a perspective that
starts and ends in a system of cognitive resources; the docility effect is a sim-
ple way of taking systematic effects into consideration when studying decision
making.
Extendable Rationality 143

Extendable Rationality
What is really new with extendable rationality? Why am I suggesting moving from
the bounded rationality approach? Hasnt it been successful?
This work followed a simple model to present extendable rationality. The first
part of the book has been dedicated to the definition of bounded rationality and
to the attempts to map it. The second part introduced the distributed cognition
approach showing that bounds are not stable; humans can stretch (or extend)
them to fit needs for rational decisions. This stretching has been defined through
the logic of adaptiveness that explains through doing, change processes, and the
social dimension of decisions.
Love is a wonderful sentiment, although it is not appropriate for scientists. I
am not in love with the idea of bounded rationality nor with that of extendable
rationality. What I suggest is that the latter is more useful for a number of reasons.
This entire book is about these reasons. Before getting into them once again, and
for the last time in this book, I would like to offer a definition of this extendable
rationality:
Rationality is extendable when it adapts to the changing conditions that define available
cognitive resources, providing workable solutions to a given problem.

The idea of rationality as extendable suits organization scholars needs because

it extends and integrates the original idea of bounded rationality more than being
opposed to it. I like to think of this as an extension of that original idea, a study on
rational bounds and on how they work. At the very end, this theory of an extendable
rationality continues the legacy of Herbert Simon and all other students of bounded
rationality simply because it answers the same question: How do people actually
make decisions?
The answer that this book provides to the question is particularly related to the
distributed cognition approach and to the docility hypothesis:

1. Instead of focusing on limitations, this theory puts emphasis on potential. It

is not particularly important how bounded our rationality is than it is how we
overcome these limitations to make decisions. Isnt this what we do all the time
when uncertainty grows high? Rationality is exactly about how we are capable of
exploiting our limitations at their best, moving the boundaries of what was a limit
for us.
2. The continuity with the bounded rationality theory stays in that we do have lim-
its, but they are not stable and they cannot be defined as internal or external.
Limits are both internal and external; a problem-solving activity is defined also
on the basis of what resources the decision maker can exploit. I have pushed
these bounds to a higher and systemic level, I should say: It is the system of
internal and external resources that defines our rationality. And it is the interplay
that moves these bounds. This is a theory that takes into consideration the fact
that resources available together with individual efforts change the perspective of
rationality and decision making. This aspect was missing in theories of bounded
rationality.
144 10 Conclusions

3. Another significant difference from the theory of bounded rationality is that the
decision maker does not get suboptimal results. The decision maker gets viable,
feasible, or workable solutions to a given problem. This changes the perspective
of the analysis. I am not interested in making a difference with the neoclassical
theory; Simon had that problem. Neoclassical theory is not the benchmark for
the extendable rationality theory because it is not for real decision makers! My
purpose is to get an idea of rationality that is more related to a real decision-
making process; there individuals get solutions (alternatives, options) that work.
The questions I always had were, suboptimal to what? What is the optimum? If
we have constraints, we can only get sub-something results. However, having a
benchmark that we cannot define is like not having a benchmark at all. Why leave
this benchmark then? I suggest that rational decision makers have the problem to
make decisions that fit a particular situation, and that they dont care about what
is an optimum or the best. If they care, these are always relative concepts that
are defined depending on the decision maker, the problem, the organization, and
available resources.
4. This theory extends rationality to external resources and in particular to social
resources. Docility emerges not because individuals are boundedly rational but
because of its opposite: Individuals move their bounds and can be rational
through the exploitation of social channels.
All in all, this theory of extendable rationality considers the problem of bounds from
the opposite side from what a theory of bounded rationality does. I do not state that
human beings have no bounds, what I argue is that this is not the most important
characteristics of their rationality. Individuals are rational because they can move
their cognitive bounds and adapt through external resources from their environment.
This enhances their fitness, especially when levels of docility grow at the individual
and at the organizational levels.
This is an approach that looks at the bright side of rationality. We do have
limitations, but we also overcome these limitations. This distinct aspect makes us
social human beings, not its opposite.

Notes
i. What I mean here is that there are no inquiries on the nature of bounds. These studies show
how these bounds affect individual behavior but the focus is on the outcome, rarely on the
cognitive mechanisms underlying this process. There is a generalized lack of philosophical
and epistemological background in studies of BR. At least, few are those that discuss or take
into consideration what is the starting point of the theory. And, since scholars do not take
a stand, one may assume that they share whatever was originally in the theory of BR. This
means that everybody that writes on or maps bounded rationality is a logical positivist, as
Simon clearly states in his book (1947, chapter 3, p. 55). Do they know?
ii. Gigerenzer, Todd, and ABC Research Group (1999). The authors here present their idea of
an ecological rationality. However, they are too shy and continue to use the old paradigm of
bounded rationality that expands to the environment and becomes situational.
Notes 145

iii. Of course, one may argue that this was not the goal of the experiment as it was conducted.
However, it is not the point of the book to falsify important results of previous experiment
but to suggest that there might be implications overlooked by experimenters. It is quite the
opposite then, that of highlighting that these experiments are far richer than one may expect.
iv. For a review, see Kerr and Tindale (2004).
v. Pfeffer (1992).
vi. In this respect, see Martin (2007) and Clawson (2006).
vii. This is explained in the first chapter of Scott (2003).
viii. Schein (1990, 1996).
ix. This is, once again, from Scott (2003).
x. An application of recent developments on niche-construction (Odling-Smee et al., 2003) to
organization theory may be helpful to support this argument. For an exploratory study of its
application, see Hench and Secchi (2009).
xi. Ilgen et al. (1994) and Hodgkinson and Healey (2008).
xii. Weick and Roberts (1993), Weick (1995), Langan-Fox et al. (2001), Langfield-Smith and
Wirth (1992), Laroche (1995), and Cannon-Bowers and Salas (2001).
xiii. This is found in Scott (2003), Organizations. Rational, Natural, and Open Systems.
Afterword

The journey always arrives at destination. In this case, the destination is a starting
point for a new journey. I hope youll enjoy it.
It is unusual for an author, but I believe it is fundamental in this case. While I
was writing this book I was thinking all the time of how alone I am in this travel.
Besides Emanuele, I do not know of anybody else that is interested in studying a
new approach to rationality. This is probably because we have spent our PhD
years in European countries, or probably because there are better theories out there.
In any case, write me (extendable.rationality@gmail.com) if you want to do research
with me, if you want just to exchange opinions, or if you simply found something
interesting or something that you hated in this book.
Thank you!
Davide Secchi
P.S. Am I being docile?

D. Secchi, Extendable Rationality, Organizational Change and Innovation, 147

DOI 10.1007/978-1-4419-7542-3,  C Springer Science+Business Media, LLC 2011
References

Abelson, R.P. 1963. Computer simulation of hot cognitions. In Computer simulation and per-
sonality: Frontier of psychological theory, eds. S. Tomkins, and S. Messick. New York:
Wiley.
Abrahamson, E., and L. Rosenkopf. 1990. When do bandwagon diffusions roll? How far do they
go? And when do they roll backwards? A computer simulation. Academy of Management Best
Paper Proceedings, 50:155159.
Abrahamson, E., and L. Rosenkopf. 1993. Institutional and competitive bandwagons: Using math-
ematical modeling as a tool to explore innovation diffusion. Academy of Management Review
18:487517.
Abrahamson, E., and L. Rosenkopf. 1997. Social network effects on the extent of innovation
diffusion: A computer simulation. Organization Science 8(3):289304.
Adams, F., and K. Aizawa. 2008. The bounds of cognition. Malden, MA: Blackwell.
Albrecht, K. 2006. Social intelligence. The new science of success. San Francisco, CA: Jossey-
Bass.
Allen, I.E., and J. Seaman. 2008. Staying the Course. Online Education in the U.S. The Sloan
Consortium. http://sloanconsortium.org/sites/default/files/staying_the_course_2.pdf, Retrieved
3 October, 2010 10:48 AM.
Anderson, M.H. 2006. How can we know what we think until we see what we said? A citation and
citation context analysis of Karl Weicks the social psychology of organizing. Organization
Studies 27(11):16751692.
Andrade, E.B., and D. Ariely. 2009. The enduring impact of transient emotions on decision making.
Organizational Behavior and Human Decision Processes 109(1):18.
Argyris, C. 1986. Reinforcing organizational defensive routines: An unintended human resources
activity. Human Resource Management 25(4):541555.
Ariely, D. 2008. Predictably irrational. The hidden forces that shape our decisions. New York,
NY: Harper Collins.
Ariely, D., and G. Loewenstein. 2006. The heat of the moment: The effect of sexual arousal on
sexual decision making. Journal of Behavioral Decision Making 19(2):8798.
Ashkanasy, N.M., W.J. Zerbe, and C.E. Hrtel. E. 2005. A bounded emotionality perspective
on interpersonal behavior in organizations. In Emotions in organizational behavior, eds.
C.E. Hrtel, W.J. Zerbe, and N.M. Ashkanasy, 113115. Mahwah, NJ: Lawrence Erlbaum
Associates.
Ashkanasy, N.M., and C.S. Daus. 2005. Rumors of the death of emotional intelligence in organi-
zational behavior are vastly exaggerated. Journal of Organizational Behavior 26:441452.
Augier, M. 2000. Models of Herbert A. Simon. Perspectives on Science 8(4):407443.
Augier, M. 2004. Marching towards a behavioral theory of the firm James G. March and the
early evolution of behavioral organization theory. Management Decision 42(10):12571268.
Aumann, R.J. 1997. Rationality and bounded rationality. Games and Economic Behavior 21:
214.

149
150 References

Axelrod, R. 1976. The structure of decision: The cognitive maps of political elites. Princeton, NJ:
Princeton University Press.
Barclay, L.J., D.P. Skarlicki, and S.D. Pugh. 2005. Exploring the role of emotions in injustice
perceptions and retaliation. Journal of Applied Psychology 90(4):629643.
Bardone, E. 2008. Human cognition as chance-seeking system. PhD dissertation submitted to the
Department of Philosophy, University of Pavia. Unpublished.
Bardone, E., and D. Secchi. 2006. Challenging bounded rationality. The cognitive role of external
resources. Presented at the Academy of Management, Atlanta (USA), August 1116.
Bardone, E., and D. Secchi. 2009. Distributed cognition: A research agenda for management. In
Current topics in management. Organizational behavior, performance, and effectiveness ed.
M.A. Rahim, vol. 14, 183207. New Brunswick, NJ: Transaction.
Bazerman, M.H. 1994. Judgement in managerial decision making (3rd edition). New York, NY:
Wiley.
Beach, L.R. (ed.) 1998. Image theory. Theoretical and empirical foundations. Mahwah, NJ:
Lawrence Erlbaum Associates.
Bechara, A., and A. Damasio. 2005. The somatic marker hypothesis: A neural theory of economic
decision. Games and Economic Behavior 52:336372.
Benartzi, S., and R.H. Thaler. 1999. Risk aversion or Myopia? Choices in repeated gambles and
retirement investments. Management Science 45(3):364381.
Betsch, T. and S. Haberstroh. (eds.) 2005. The routines of decision making. Mahwah, NJ and
London: Lawrence Erlbawm.
Bilboa, I. 2010. Rational choice. Cambridge, MA: MIT Press.
Block, W., N. Snow, and E. Stringham. 2008. Banks, insurance companies and discrimination.
Business and Society Review 113(3):403419.
Bonaccio, S., and R.S. Dalal. 2006. Advice taking and decision-making: An integrative literature
review, and implications for the organizational sciences. Organizational Behavior & Human
Decision Processes 101(2):127151.
Bono, J.E., and M.A. Vey. 2005. Toward Understanding Emotional Management at Work: A
Quantitative Review of Emotional Labor Research. In Emotions in organizational behavior,
eds. C.E. Hrtel, W.J. Zerbe, and N.M. Ashkanasy., 213234. Mahwah, NJ: Lawrence Erlbaum
Associates.
Bornmann, L., and H.-D. Daniel. 2006. What do citation counts measure? Counts measure? A
review of studies on citing behavior. Journal of Documentation 64(1):4580.
Camerer, C. 1998. Bounded rationality in individual decision making. Experimental Economics
1:163183.
Camerer, C. 2007. Neuroeconomics: Using neuroscience to make economic predictions. Economic
Journal 117(March):C26C42.
Camerer, C., G. Loewenstein, and M. Weber. 1989. The curse of knowledge in economic settings:
An experimental analysis. Journal of Political Economy 97:12321254.
Cannon-Bowers, J.A., and E. Salas. 2001. Reflections on shared cognition. Journal of
Organizational Behavior 22:195202.
Carnall, C.A. 1990. Managing change in organizations. New York, NY: Prentice-Hall.
Case, D.O., and G.M. Higgins. 2000. How can we investigate citation behavior? A study of reasons
for citing literature in communication. Journal of the American Society for Information Science
51(7):635645.
Chao, G.T., and H.W. Willaby. 2007. International employment discrimination and implicit social
cognition: New directions for theory and research. Applied Psychology: An International
Review 56(4):678688.
Chiang, Y.-S. 2007. Birds of moderately different feathers: Bandwagon dynamics and the threshold
heterogeneity of network neighbors. Journal of Mathematical Sociology 31:4769.
Clark, A. 2003. Natural-born cyborgs. Minds, technologies, and the future of human intelligence.
New York, NY: Oxford University Press.
Clark, A. 2008. Supersizing the mind. Embodiment, action, and cognitive extension. New York,
NY: Oxford University Press.
References 151

Clark, A., and D.J. Chalmers. 1998. The extended mind. Analysis 58(1):719.
Clawson, J.G. 2006. Level three leadership. Getting below the surface (3rd edition). Upple Saddle
River, NJ: Prentice Hall.
Conlisk, J. 1996. Why bounded rationality? Journal of Economic Literature 34(2):669700.
Daft, R.L., R.H. Lengel, and L.K Trevino. 1987. Message equivocality, media selection, and man-
ager performance: Implications for information systems. MIS Quarterly September:355366.
Damasio, A. 1994. Descartes error: Emotion, reason, and the human brain. New York, NY: Avon.
Deephouse, D.L. 1996. Does isomorphism legitimate? Academy of Management Journal 39:
10241039.
De George, R.T. 1999. Business ethics (5th edition). Upper Saddle River, NJ: Prentice Hall.
Dennis, A.R., and S.T. Kinney. 1998. Testing media richness theory in the new media: The effects
of cues, feedback, and task equivocality. Information Systems Research 9(3):256274.
Descartes, R. 1984/1641. The philosophical writings of Descartes (trans: Cottingham, J., Stoothoff,
R., Murdoch, D.). Cambridge: Cambridge University Press.
Dickson, P.R., and A.G. Sawyer. 1990. The price knowledge and search of supermarket shoppers.
Journal of Marketing 54:4253.
Dreyer, T.P. 1993. Modeling with ordinary differential equations. Boca Raton, FL: CRC Press.
Drolet, A.L., and M.W. Morris. 1995. Communication media and interpersonal trust in conflicts:
The role of rapport and synchrony of nonverbal behavior. Paper presented and the Academy of
Management meeting, Canada: Vancouver.
Dunbar, R.I.M. 1998. The Social Brain Hypothesis. Evolutionary Anthropology 6(5):153190.
Dunbar, R.I.M., and S. Shultz. 2007. Evolution in the Social Brain. Science 317(5843):13441347.
Earley, P.C. 2002. Redefining interactions across cultures and organizations: Moving forward with
cultural intelligence. Research in Organizational Behavior 24:271299.
Earley, P.C., and E. Mosakowski. 2004. Toward culture intelligence: Turning cultural differences
into a workplace advantage. Academy of Management Executive 18(3):151157.
El-Shinnawy, M., and M.L. Markus. 1997. The poverty of media richness theory: explaining
peoples choice of electronic mail vs. voice mail. International Journal of Human-Computer
Studies 46:443467.
Emmerling, R.J., and D. Goleman. 2005. Leading with Emotions. Leadership Excellence 22(7):
910.
Eysenk, H.J., and L. Kamin. 1981. Intelligence: The battle for the mind. London: MacMillan.
Festinger, L. 1957. A theory of cognitive dissonance. Stanford, CA: Stanford University Press.
Fiol, C.M., and E.J. OConnor. 2003. Waking up! Mindfulness in the face of bandwagon. Academy
of Management Review 28(1):5470.
Foss, N.J. 2003a. Bounded rationality and tacit knowledge in the organizational capabilities
approach: an assessment and a re-evaluation. Industrial and Corporate Change 12:185201.
Foss, N.J. 2003b. Bounded rationality in the economics of organizations: much cited and little
used. Journal of Economic Psychology 24:245264.
Frank, R.H. 2004. What price the moral high ground? Ethical dilemmas in competitive environ-
ments. Princeton and Oxford: Princeton University Press
Freeman, E.R., A.C. Wicks, and B. Parmar. 2004. Stakeholder theory and the corporate objective
revisited. Organization Science 15(3):364369.
Friedman, M. 1953. Essays in positive economics. Chicago, IL: University of Chicago Press.
Gabbay, D.M., and J. Woods. 2007. Seductions and Shortcuts. Fallacies in the cogni-
tive economy. Book in progress, http://www.dcs.kcl.ac.uk/staff/dg/SeductionsandShortcuts-
complete.pdf, Retrieved January 12, 2009 9:36 AM.
Gardner, P., and D. Berry. 1995. The effect of different forms of advice on the control of a simulated
complex system. Applied Cognitive Psychology 9:S55S79.
Garriga, E., and D. Mel. 2004. Corporate social responsibility theories: Mapping the territory.
Journal of Business Ethics 53:5171.
Gazzaniga, M.S. 2008. Human. The science behind what makes us unique. New York, NY: Harper
Collins.
152 References

Gibbs, R.W. Jr. 2005. Embodiment and cognitive science. Cambridge: Cambridge University Press.
Gigerenzer, G., H. Brighton. 2009. Homo heuristicus: Why biased minds make better inferences.
Topics in Cognitive Science 1:107143.
Gigerenzer, G., and R. Selten. (eds.) 2001. Bounded rationality. The adaptive toolbox. Cambridge,
MA: MIT Press.
Gigerenzer, G., P.M Todd, and ABC Research Group 1999. Simple heuristics that make us smart.
New York, NY: Oxford University Press.
Gilboa, I. (2010). Rational choice. Cambridge, MA: MIT Press.
Gladwell, M. 2002. The tipping point: How little things can make a big difference. New York, NY:
Back Bay Books.
Gladwell, M. 2008. Outliers. The story of success. New York: Little, Brown, and Co.
Goff, P.A., J.L. Eberhardt, M.J. Williams, and M.C. Jackson. 2008. Not yet human: Implicit knowl-
edge, historical dehumanization, and contemporary consequences. Journal of Personality and
Social Psychology 94(2):292306.
Goldman, A., and K. Mason. 2007. Simulation. In Philosophy of psychology and cognitive science,
ed. P. Thagard, 267293. Amsterdam: Elsevier.
Goldstein, D.G., G. Gigerenzer. 1996. Reasoning the fast and frugal way: Models of bounded
rationality. Psychological Review 103(4):650669.
Goleman, D. 1995. Emotional intelligence. New York, NY: Bantam Books.
Golinelli, G.M. 2005. Lapproccio sistemico al governo dellimpresa, vol. I Limpresa sistema
vitale. Padova: CEDAM.
Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78(6):
13601380.
Granovetter, M. 1978. Threshold models of collective behavior. American Journal of Sociology
83(6):14201443.
Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness.
American Journal of Sociology 91:481510.
Grove, A. 1999. Only the paranoid survive: How to exploit the crisis points that challenge every
company. New York, NY: Broadway Books.
Groves, R.M., J.F. Fowler, M.P. Couper, J.M. Lepkowski, E. Singer, and R. Tourangeau. 2009.
Survey methodology (2nd edition). Hoboken, NJ: Wiley.
Grne-Yanoff, T. 2007. Bounded Rationality. Philosophy Compass 2/3:534563.
Guerrera, F. 2009. Welch denounces corporate obsessions. Financial Times, March 13.
http://www.ft.com/cms/s/0/3ca8ec2e-0f70-11de-ba10-0000779fd2ac.html?nclick_check=1,
Retrieved 8 May 2009 6:56 PM.
Hanoch, Y. 2002. Neither an angel nor an ant: Emotion as an aid to bounded rationality. Journal
of Economic Psychology 23:125.
Hanoch, Y., S. Wood, and T. Rice. 2007. Bounded rationality, emotions and older adult decision
making: Not so fast and yet so frugal. Human Development 50:333358.
Hansen, H.H. 2002. The straw thing of fallacy theory: the standard definition of fallacy.
Argumentation 16(2):133155.
Harris, R. 2004. Integrationism, language, mind and world. Language Sciences 26(6):
727739.
Hrtel, C.E., W.J. Zerbe, and N.M. Ashkanasy. 2005. Emotions in organizational behavior.
Mahwah, NJ: Lawrence Erlbaum Associates.
Harvey, N., and I. Fisher. 1997. Taking advice: Accepting help, improving judgment, and sharing
responsibility. Organizational Behavior and Human Decision Processes 70:117133.
Hedlund, J., D.R. Ilgen, and J.R. Hollenbeck. 1998. Decision accuracy in computer-mediated
versus face-to-face decision-making teams. Organizational Behavior and Human Decision
Processes 76:3047.
Heine, S.J., and D.R. Lehman. 1997. Culture, dissonance, and self-affirmation. Personality and
Social Psychology Bulletin 23:389400.
Hench, T., and D. Secchi. 2009. Organizational niche-construction and stakeholder analysis:
Concepts and implications. Philosophy of Management 8(3):4764.
References 153

Hodgkinson, G.P., and M.P. Healey. 2008. Cognition in organizations. Annual Review of Sociology
59:387417.
Holyoak, K.J., and P. Thagard. 1995. Mental leaps. Analogy in creative thought. Cambridge, MA:
MIT Press.
Hoshino-Browne, E., A.S. Zanna, S.J. Spencer, M.P. Zanna, S. Kitayama, and S. Lackenbauer,
2005. On the cultural guises of cognitive dissonance: The case of Easterners and Westerners.
Journal of Personality and Social Psychology 89(3):294310.
Humphrey, N.K. 1976. The social function of intellect. In Growing points in ethology, eds. P.P.G.
Bateson, and R.A. Hinde, 303317. Cambridge, MA: Cambridge University Press.
Hutchins, E. 2000. Distributed cognition. Working paper. Available online at:
http://eclectic.ss.uci.edu/drwhite/Anthro179a/DistributedCognition.pdf. Accessed 12
Jan 2009 11:00 AM.
Hutchins, E. 1995. Cognition in the wild. Cambridge, MA: MIT Press.
Ilgen, D.R., D.A. Major, and L.T Spencer. (1994). The cognitive revolution in organizational
behavior. In Organizational behavior: The state of the science, ed. J. Greenberg, 122.
Mahwah, NJ: Lawrence Erlbaum Associates.
Jones, G.R., and J.M. George. 2009. Contemporary management (5th edition). New York, NY:
McGraw-Hill.
Kahneman, D. 2003. A perspective of judgement and choice. Mapping bounded rationality.
American Psychologist 58(9):697721.
Kahneman, D., and A. Tversky. 1979. Prospect theory: An analysis of decision under risk.
Econometrica 47(2):263292.
Kahneman, D., J.L. Knetsch, and R.H. Thaler. 1990. Experimental tests of the endowment effect
and the Coase theorem. Journal of Political Economy 98:13251348.
Kahneman, D., J.L. Knetsch, and R.H. Thaler. 1991. The endowment effect, loss aversion, and
status quo bias. Journal of Economic Perspectives 5(1):193206.
Katz, M.L., and C. Shapiro. 1985. Network externalities, competition, and compatibility. American
Economic Review 75:424440.
Kavanaugh, A., D. Reese, J. Carroll, and M. Rosson. 2005. Weak ties in networked communities.
Information Society 21(2):119131.
Kerr, N.L., and R.S. Tindale. 2004. Group performance and decision making. Annual Review of
Psychology 55:623655.
Khalil, E.L. 2004. What is altruism? Journal of Economic Psychology 25:97123.
Klaes, M., and E.-M. Sent. 2005. A conceptual history of the emergence of bounded rationality.
History of Political Economy 37(1):2759.
Knoke, D., and S. Yang. 2008. Social network analysis (2nd edition). Thousand Oaks, CA: Sage.
Knudsen, T. 2003. Simons selection theory: Why docility evolves to breed successful altruism.
Journal of Economic Psychology 24:229244.
Knudsen, T., and D.A. Levinthal. 2007. Two faces of search: Alternative generation and alternative
evaluation. Organization Science 18(1):3954.
Kock, N. 2005. Media richness or media naturalness? The evolution of our biological commu-
nication apparatus and its influence on our behavior toward E-communication Tools. IEEE
Transactions on Professional Communication 48(2):117130.
Krackhardt, D., and M. Kilduff. 2002. Structure, culture and Simmelian ties in entrepreneurial
firms. Social Networks 24(3):279290.
Kunda, Z. 1999. Social cognition. Making sense of people. Cambridge, MA: MIT Press.
Laland, K.N. 2001. Imitation, social learning and preparedness as mechanisms of bounded rational-
ity. In Bounded rationality. The adaptive toolbox, eds. G. Gigerenzer, and R. Selten, 233247.
Cambridge, MA: MIT Press.
Langan-Fox, J., A. Wirth, S. Codea, K. Langfield-Smith, and A. Wirth. 2001. Analyzing shared
mental models. International Journal of Industrial Ergonomics 28:99112.
Langer, E.J. 1989. Minding matters: The consequences of mindlessness-mindfulness. In Advances
in experimental social psychology, ed. L. Berkowitz, vol. 22: 137173. San Diego, CA:
Academic Press.
154 References

Langfield-Smith, K., and A. Wirth. (1992). Measuring differences between cognitive maps. Journal
of the Operational Research Society 43(12):11351150.
Langley, A., H. Mintzberg, P. Pitcher, E. Posada, and J. Saint-Macary. 1995. Opening up decision
making: The view from the black stool. Organization Science 6(3):260279.
Laroche, H. 1995. From decision to action in organizations: Decision-making as a social
representation. Organization Science 6(1):6275.
Lazarus, R.S. 1984. On the primacy of cognition. American Psychologist 39(2):124129.
Lazarus, R.S. 1991. Emotion and adaptation. New York, NY: Oxford University Press.
Leibenstein, H. 1950. Bandwagon, snob, and Veblen effects in the theory of consumers demand.
Quarterly Journal of Economics 64(2):183207.
Levinthal, D., and C. Rerup. 2006. Crossing an apparent chasm: Bridging mindful and less-mindful
perspectives on organizational learning. Organization Science 17(4):502513.
Lewin, K. 1951. Field theory in social science. New York, NY: Harper & Row.
Liu, Y., and P.L. Perrew. 2005. Another look at the role of emotion in the organizational change:
A process model. Human Resource Management Review 15(4):263280.
Locke, E.A. 2005. Why emotional intelligence is an invalid concept. Journal of Organizational
Behavior 26: 425431.
Love, N. 2004. Cognition and the language myth. Language Sciences 26(6):525544.
Luce, R.D., and H. Raiffa. 1989. Games and decisions. Introduction and critical survey. Mineola,
NY: Dover.
Magnani, L. 2001. Abduction, reason, and science. Processes of discovery and explanation. New
York, NY: Kluwer.
Magnani, L. (ed.) 2006. Model-based reasoning in science and engineering. Cognitive science,
epistemology, logic. London: College.
Magnani, L. 2006. Mimetic Minds. Meaning Formation Through Epistemic Mediators and
External Representations. In Artificial cognition systems, eds. A. Loula, R. Gudwin, and J.
Queiroz. 327357. Hershey, PA: Idea Group.
Magnani, L. 2007. Morality in a technological world. Knowledge as a duty. Cambridge:
Cambridge University Press.
Magnani, L., and Nersessian, N.J. (eds.). 2002. Model-based reasoning. Scientific discovery,
technological innovation, values. New York, NY: Kluwer/Plenum.
Magnani, L., N.J. Nersessian, and P. Thagard. (eds.). 1999. Model-based reasoning in scientific
discovery. New York, NY: Kluwer/Plenum.
Magnani, L., D. Secchi, and E. Bardone. 2006. Docility and through doing morality: An alter-
native approach to ethics. Working paper, 8, Department of Economics. Varese: University of
Insubria.
Maramatsu, R., and Y. Hanoch. 2005. Emotions as a mechanism for boundedly rational agents.
The fast and frugal way. Journal of Economic Psychology 26:201221.
March, J.G. 1994. A primer on decision making. New York, NY: Free Press.
March, J.G., and Z. Shapira. 1987. Managerial Perspectives on Risk and Risk Taking. Management
Science 33(11):14041418.
March, J.G., and H.A. Simon. 1958. Organizations. London: Wiley.
Martin, R.L. 2007. The opposable mind: How successful leaders win through integrative thinking.
Cambridge, MA: Harvard Business School Press.
Mas-Colell, A., M.D. Whinston, and J.R. Green. 1995. Microeconomic theory. New York, NY:
Oxford University Press.
Mathis, R.L., and J.H. Jackson. 2008. Human resource management (12th edition). Mason, OH:
Thomson South-Western.
Maturana, H., and F.J. Varela. 1987. The three of knowledge: The biological roots of human
understanding. Boston, MA: New Science Library.
Meerschaert, M.M. 1999. Mathematical modeling (2nd edition). London: Academic.
Merton, R.K. 1996. The Matthew effect, ii. In On social structure and science, ed. P. Sztompka.
318336. Chicago, IL: Chicago University Press.
References 155

Meyer, J., and W.R. Scott (eds.). 1983. Organizational environments: Ritual and rationality.
Beverly Hills, CA: Sage.
Michel, A.A. 2007. A distributed cognition perspective on newcomers change processes:
The management of cognitive uncertainty in two investment banks. Administrative Science
Quarterly 52(4):507557.
Milkovich, G.T., and J.M. Newman. 2008. Compensation (9th edition). New York, NY: McGraw-
Hill.
Mithen, S. 1999. Handaxes and ice age carvings: hard evidence for the evolution of consciousness.
In Toward a science of consciousness iii. The third Tucson discussions and debates, eds. A.R.
Hameroff, A.W. Kaszniak, and D.J. Chalmers. 281296. Cambridge, MA: MIT Press
Moore, D., T.R. Kurtzberg, L.L. Thompson, and M.W. Morris. 1999. Long and short routes
to success in electronically mediated negotiations: Group affiliations and good vibrations.
Organizational Behavior & Human Decision Processes 77(1):2243.
Mousavi, S., and J. Garrison. 1992. Toward a transactional theory of decision making: creative
rationality as functional coordination in context. Journal of Economic Methodology 10(2):
131156.
Mumby, D.K., and L.L. Putnam. 1992. The politics of emotion: A feminist reading of bounded
rationality. Academy of Management Review 17(3):465486.
Munier, B., R. Selten, D. Bouyssou, P. Bourgine, R. Day, N. Harvey, D. Hilton, M.J. Machina,
P. Parker, J. Sterman, E. Weber, B. Wernerfelt, and R. Wensley. 1999. Bounded rationality
modelind. Marketing Letters 10(3):233248.
Nagel, E., and J. Newman. 1958. Gdels proof. New York, NY: NYU Press.
Neale, M.A., and M.H. Bazerman. 1991. Cognition and rationality in negotiation. New York, NY:
Free Press.
Nelson, R.R., and S.G. Winter. 1973. Toward an evolutionary theory of economic capabilities.
American Economic Review 63:440449.
Nelson, R. and S. Winter. 1982. An evolutionary theory of economic change. Cambridge, MA:
Belknap Press.
Newell, A. and H.A. Simon. 1972. Human problem solving. Englewood Cliffs, NJ: Prentice-Hall.
Newton, K. 1997. Social capital and democracy. American Behavioral Scientist 40(5):
575586.
Nickerson, R.S. 1999. How we knowand sometimes misjudgewhat others know: Imputing
ones own knowledge to others. Psychological Bulletin 125:737759.
Norman, D. 1993. Things that make us smart. New York, NY: Addison Wesley.
Odling-Smee, F.J., K.N. Laland, and M.W. Feldman. 2003. Niche construction: The neglected
process in evolution. Cambridge, MA: Princeton University Press.
Oh, H., G. Labianca, and M.-H. Chung. 2006. A multilevel model of group social capital. Academy
of Management Review 31(3):569582.
Olive, D. 2001. The man who taught an elephant to dance: General Electrics Jack Welch was one
of the first celebrity CEOs. National Post. Don Mills, Ont.: Sep 7, p. C16.
Pahor, M., M. kerlavaj, V. Dimovski. 2008. Evidence for the network perspective on organi-
zational learning. Journal of the American Society for Information Science & Technology
59(12):19851994.
Panksepp, J. 1982. Toward a general psychobiological theory of emotions. The Behavioral and
Brain Sciences 5:407467.
Panksepp, J. 1992. A critical role for affective neuroscience in resolving what is basic about basic
emotions. Psychological Review 99:554560.
Panskepp, J. 2007. Criteria for basic emotions: Is DISGUST a primary emotion? Cognition and
Emotion 21(8):18191828.
Patokorpi, E. 2008. Simons paradox: Bounded rationality and the computer metaphor of the mind.
Human Systems Management 27:285294.
Payne, J.W., J.R. Bettman, and E.J. Johnson. 1993. The adaptive decision maker. New York, NY:
Cambridge University Press.
156 References

Peng, W.-S. 1992. A critique on H.A. Simons administrative behavior theory. Public
Administration Quarterly 16(2):254264.
Pennington, D.C. 2000. Social cognition. London: Routledge.
Perry, M. 2003. Distributed cognition. In Hci, models, theories, and frameworks: Toward a
Multidisciplinary Science ed. J.M. Carroll. 193224. London: Morgan Kaufman.
Pierce, C.S. 1955. Abduction and induction. In Philosophical writings of pierce, eds. C.S. Peirce
J. Buchler. 150156. New York, NY: Dover.
Pfeffer, J. 1992. Managing with power. Politics and influence in organizations. Cambridge, MA:
Harvard Business School Press.
Plato, 2005. Phaedrus (trans: Rowe, C.). London: Penguin Books.
Polanyi, M. (1966). The tacit dimension. London: Routledge.
Politser, P. 2008. Neuroeconomics. A guide to the new science of making choices. New York, NY:
Oxford University Press.
Popper, K. 1935/2002. The logic of scientific discovery. New York, NY: Routledge.
Post, T., M.J. van den Assem, G. Baltussen, and R.H. Thaler. 2008. Deal or No Deal? Decision
Making under Risk in a Large-Payoff Game Show. American Economic Review 98(1):
3871.
Prigogine, I., and I. Stengers. 1986. La nouvelle alliance. Paris: Gallimard.
Purdy, J. M., P. Nye, and P.V. Balakrishnan. 2000. The impact of communication media of
negotiation outcomes. International Journal of Conflict Management 11(2):162187.
Rabin, M., and R.H. Thaler. 2001. Risk Aversion. Journal of Economic Perspectives 15(1):
219232.
Reid, T.R. 2004. The United States of Europe. The new superpower and the end of American
supremacy. New York, NY: Penguin Books.
Richardson, K. 2000. The making of intelligence. New York, NY: Columbia University Press.
Richerson, P.J., and R. Boyd. 1998. The evolution of human ultra-sociality. In Indoctrinability,
ideology, and warfare: Evolutionary perspective, eds. I. Eibl-Eibesfeldt, and F.K. Salter, 7195.
New York: Berhahn Books.
Richerson, P.J., and R. Boyd. 2005. Not by genes alone: How culture transformed human evolution.
Chicago, IL: Chicago University Press.
Rothschild, K.W. 2001. The reluctant rebel or glamor and poverty of the homo oeconomicus.
Kyklos 54(2/3):445452.
Rubinstein, A. 1998. Modeling bounded rationality. Cambridge, MA: MIT Press.
Schein, E.H. 1990. Organizational culture. American Psychologist 45(2):109119.
Schein, E.H. 1996. Culture: The missing concept in organization studies. Administrative Science
Quarterly 41:229240.
Schein, E.H. 1999. Kurt Lewins change theory in the field and in the classroom: Notes toward a
model of managed learning. Reflections 1(1):5972.
Schein, E.H. 2002. Models and tools for stability and change in human systems. Reflections
4(2):3446.
Scott, W.R. 1995. Institutions and organizations. Thousand Oaks, CA: Sage.
Scott, W.R. 2003. Organizations. Rational, natural, and open systems (5th edition). Englewood
Cliffs, NJ: Prentice-Hall.
Secchi, D. 2007. A theory of docile society: The role of altruism in human behavior. Journal of the
Academy of Business and Economics 7(2):146160.
Secchi, D. 2007. Utilitarian, managerial, and relational theories of corporate social responsibility.
International Journal of Management Reviews 9(4):347373.
Secchi, D. 2010. Banks and Discrimination: How Far Can We Go with Competition? A Reply to
Block, Snow, and Stringham. Submitted to Business and Society Review, March 2009.
Secchi, D. 2009. The Cognitive Side of Social Responsibility. Journal of Business Ethics
88(3):565581.
Secchi, D., and E. Bardone. 2009a. An Organizational Model of Bandwagon. Working Paper
No. 1, College of Business Administration, University of WisconsinLa Crosse.
References 157

Secchi, D., and E. Bardone. 2009b. Super-docility in organizations. International Journal of

Organization Theory and Behavior 12(3):339379.
Selten, R. 1998. Features of experimentally observed bounded rationality. European Economic
Review 42:413436.
Semler, R. 1994. Why my former employees still work for me. Harvard Business Review January
February:6474.
Sent, E.-M. 1997. Sargent versus Simon: Bounded rationality unbound. Cambridge Journal of
Economics 21:323338.
Sent, E.-M. 2005. Simplifying Herbert Simon. History of Political Economy 37(2):227332.
Shakun, M.F. 2001. Unbounded Rationality. Group Decision and Negotiation 10:97118.
Shampanier, K., N. Mazar, and D. Ariely. 2007. Zero as a special price: The true value of free
products. Marketing Science 26(6):742757.
Shier, D.R., and K.T. Wallenius. 2000. Applied mathematical modeling. A multidisciplinary
approach. Boca Raton, FL: Chapman & Hall/CRC.
Silver, W.S., and T.R. Mitchell. 1990. The status quo tendency in decision making. Organizational
Dynamics 18(4):3446.
Simon, H. 1955. A behavioral model of rational choice. Quarterly Journal of Economics LXIX:
99118.
Simon, H.A. 1999. Interview with Herbert A. Simon. Working paper, CISTEMA, Copenhagen
Business School.
Simon, H.A. 1947/1997. Administrative behavior (4th edition). New York, NY: The Free Press.
Simon, H.A. 1957. Models of man. New York, NY: Wiley.
Simon, H.A. 1959. Theories of decision-making in Economics and behavioral science. American
Economic Review 49(3):253283.
Simon, H.A. 1960. The new science of management decision. New York, NY: Harper & Row.
Simon, H.A. 1978. Rationality as process and a product of thought. American Economic Review
68:114.
Simon, H.A. 1990. A mechanism for social selection and successful altruism. Science
250(4988):16651668.
Simon, H.A. 1993b. Altruism and economics. American Economic Review 83(2):156161.
Simon, H.A. 1993a. The human mind: The symbolic level. Proceedings of the American
Philosophic Society 137(4):638647.
Simon, H.A. 1979. Rational decision making in business organizations. American Economic
Review 69:493513.
Slaugher, J.E., and S. Highhouse. 2003. Does matching up features mess up job choice?
Boundary conditions on attribute-salience effects. Journal of Behavioral Decision Making 16:
115.
Small, H. 2004. On the shoulders of Robert Merton: Towards a normative theory of citation.
Scientometrics 60(1):7179.
Sniezek, J.A., G.E. Schrah, and R.S. Dalal. 2004. Improving judgment with prepaid expert advice.
Journal of Behavioral Decision Making 17:173190.
Starbuck, W.H. 1983. Organizations as action generators. American Sociological Review 48(1):
91102.
Staw, B.M., and L.D. Epstein. 2000. What bandwagons bring: Effects of popular management tech-
niques on corporate performance, reputation, and CEO pay. Administrative Science Quarterly
45:523556.
Sternberg, R.J. (ed.). 1999. Handbook of creativity. Cambridge: Cambridge University Press.
Sundaram, A.K., and A.C. Inkpen. 2004. The corporate objective revisited. Organization Science
15(3):350363.
Sunstein, C.R. 2005. Conformity and dissent. Public Law and Legal Theory Working Paper No.
34, University of Chicago.
Sutton, J. 2004. Representation, levels and context in integrational linguistics and distributed
cognition. Language Sciences 26(6):503524.
158 References

Thagard, P. 2007. Introduction to the Philosophy of Psychology and Cognitive Science. In

Philosophy of psychology and cognitive science, ed. P. Thagard. ixxvii. Amsterdam: Elsevier.
Thagard, P. (ed.) 2007. Philosophy of psychology and cognitive science. Amsterdam: Elsevier.
Thaler, R.H., and C.R. Sunstein. 2008. Nudge. Improving decisions about health, wealth, and
happiness. New Haven & London: Yale University Press.
Thomas, D.C., and K. Inkson. 2003. Cultural intelligence. San Francisco, CA: Berrett-Kohler.
Todd, P.M. 2001. Fast and frugal heuristics for environmentally bounded minds. In Bounded ratio-
nality. The adaptive toolbox, eds. G. Gigerenzer, and R. Selten, 5170. Cambridge, MA: MIT
Press.
Todd, P.M., and G. Gigerenzer. 2003. Bounding rationality to the world. Journal of Economic
Psychology 24:143165.
Tversky, A., and D. Kahneman. 1974. Judgment under Uncertainty: Heuristics and Biases. Science
185:11241130.
Van der Veer Martens, B., and A.A. Goodrum 2006. The diffusion of theories: A functional
approach. Journal of the American Society for Information Science and Technology 57(3):
330341.
Verplanken, B., V. Myrbakk, and E. Rudi. 2005. The measurement of habit. In The routines of
decision making, eds. T. Betsch and S. Haberstroh, 231247. Mahwah, NJ & London: Lawrence
Erlbawm.
Vicari, S. 1991. Limpresa vivente. Itinerario in una diversa concezione. Milano: Etas.
Von Foerster, H. 2002. Understanding systems. Conversations on epistemology and ethics. New
York, NY: Kluwer/Plenum.
Von Neumann, J., and O. Morgenstern. 1944. Theory of games and economic behavior. Cambridge,
MA: Princeton University Press.
Wallister, B. 2008. Cognitive economics. Berlin & Heidelberg: Springer.
Walsh, J.P. 1995. Managerial and organizational cognition: Notes from a trip down memory lane.
Organization Science 6(3):280321.
Weber, E.U., and E.J. Johnson. 2009. Mindful judgment and decision making. Annual Review of
Psychology 60:5385.
Weick, K. 1995. Sensemaking in organizations. Thousand Oaks, CA: Sage.
Weick, K., and K.H. Roberts. 1993. Collective mind and organizational reliability: the case of flight
operations on an aircraft carrier deck. Administrative Science Quarterly 38:357381.
Weiss, H.M., K. Suckow, and R. Cropanzano. 1999. Effects of justice conditions on discrete
emotions. Journal of Applied Psychology 84(5):786794.
Werhane, P.H. 1999. Moral imagination and management decision making. New York, NY: Oxford
University Press.
Wilson, R.A. 1994. Wide computationalism. Mind 103:351372.
Wilson, R.A. 2004. Boundaries of the mind. Cambridge: Cambridge University Press.
Wilson, R.A. 2005. Collective memory, group minds, and the extended mind thesis. Cognitive
Processing 6(4):227236.
Woods, J. 2004. The death of argument. Dordrecht: Kluwer.
Yaniv, I., and E. Kleinberger. 2000. Advice taking in decision making: egocentric discounting and
reputation formation. Organizational Behavior and Human Decision Processes 83:260281.
Zafirovski, M. 1999. The unbearable lightness of economic approach to economic behavior in
the social setting: Rational action and the sociology of the economy. Journal for the Theory of
Social Behavior 29(3):301334.
Zajonc, R.B. 1984. On the Primacy of Affect. American Psychologist 39(2):117123.
Zanna, M.P., and J. Cooper. 1974. Dissonance and the pill: An attributional approach to studying
the arousal properties of dissonance. Journal of Personality and Social Psychology 29:703709.
Zelen, M. 2001. Obituary. Herbert A. Simon (19162001). Human Systems Management 20:34.
Zhang, J. 1997. The nature of external representations in problem-solving. Cognitive Science
21(2):179217.
Index

A status quo, 3233

Abduction, 2, 13, 18, 51, 78 Biases paradigm, 5455
Accessibility, 5, 4852, 7576 Boundedly rationality, 20, 24, 37, 52, 63, 77,
Advice 83, 114, 144
active, 98, 122, 127 Bounded rationality, 1, 36, 14, 1925, 2739,
and bandwagon, 82 4159, 6364, 68, 7374, 7678,
giving, 2, 82, 100, 102, 104, 107109, 111, 8183, 85, 93, 95, 113114, 129,
114, 127129 137138, 141, 143144
and information mediums, 97 Brain, 46, 4243, 4648, 55, 6379, 8195,
and information richness scale, 102 97111, 113133, 135145
judge-advisor system (JAS), 99101, isolated, 81
103104
passive, 108110 C
taking, 98111, 113, 115, 119, 128129, Chalmers, David J., 78
133, 137 Change, 14, 11, 2425, 28, 3234, 39, 42,
Altruism, 27, 83, 95, 115116, 45, 5254, 56, 58, 6669, 71, 76,
131133, 137 82, 84, 8694, 97, 100, 117118,
Ariely, Dan, 17, 38, 79, 94 120, 122, 124, 136138, 141,
143144
Clark, Andy, 73, 7879, 90, 95
B Cognition
Bandwagon cognitive process, 74
and docility, 126, 132 distributed (DC), 1, 6, 6478, 81, 8489,
and organizational culture, 124125 9192, 9798, 103105, 110111, 113,
and social relations, 125 115, 117119, 123, 129, 132, 135, 138,
Bandwagon effect, 3435, 39, 50, 119, 142143
123, 132 Cognitive divide, 6, 72, 76
Bardone, Emanuele, 39, 65, 78, 95, 111, 115, Community, 1, 45, 74, 114, 120126,
118, 131132 132133, 142
Behavioral Computation, 16, 22
research, 54 Computational, 1314, 19, 22, 31, 5455, 57,
studies, 27, 54, 79 64, 76, 135
Behaviorism, 141 limits, 31, 5455
Bias Context, 53, 63, 71, 73, 82, 84, 8889, 92, 94,
anchor bias, 3334 97, 104, 117118
endowment effect, 3132 Cooperation, 27, 130131, 133, 137
and errors, 3637 Culture, 45, 71, 79, 8384, 110, 120, 124126,
and prejudices, 31, 3536 130, 132, 139140, 142

159
160 Index

D Externalization, 7073, 75, 8384, 87, 90

Damasio, Antonio, 5758, 9495 External resources
Decision making, 16, 921, 2324, 29, 31, artifacts, 4, 81, 136
3536, 39, 4143, 4647, 49, 5153, non-social, 88, 127
58, 6364, 6970, 7277, 79, 8189, social, 8788, 98, 114, 117119, 121,
92, 9495, 9798, 101, 103, 105106, 127, 139
110, 113117, 119121, 129130, social channels, 127, 144
135136, 138143
Deduction, 13 F
Descartes, Ren, 17, 47, 57 Fallacy
Docile ad hominem, 51
exceptionally docile, 121123 ad populum, 50
non-docile, 114117, 121123, 129, 133 ad verecundiam, 50
ordinarily docile, 121123 composition and division, 51
organization, 113133 gang of eighteen, 4950
Docility hasty generalization, 51
and altruism, 115116, 131133 Foss, Nicholai J., 1, 18, 25, 77, 81, 93
and bandwagon, 123126 Framing, 5, 35, 4851, 7576, 87, 103
community, 120 Fully rational, 1921, 25, 63, 137
and cooperation, 130131, 133
degrees of, 117 G
and environment, 117118 Gabbay, Dov M., 49, 58, 78
in organizations, 121123 Gigerenzer, Gerd, 25, 42, 5457, 59,
prerequisites, 119121 7778, 144
public availability of information, 121 Granovetter, Mark, 39, 108, 111, 123, 132
and social channels, 114, 117, 119123,
126130, 133 H
social responsibility, 126128 Hanoch, Yaniv, 47, 58, 77, 9495
Heuristics
and standards, 120
and biases, 3, 5
and time, 118
and fallacies, 42
Docility effect, 118, 129, 131, 142
fast and frugal (FFH), 4244, 48, 5758, 83
High-technology, 8991, 97
E Homo economicus, 25, 64
Ecological rationality, 55, 144 Hutchins, Edwin, 1, 64, 73, 7879, 119, 132
Emotions
and bounded rationality, 4648 I
and distributed cognition, 71, 84 Induction, 1213, 18, 51
and through doing logic, 91 Information
Environment, 13, 24, 35, 39, 44, 52, 55, discount, 109
6465, 7377, 79, 81, 85, 87, 89, 92, limits, 55
97, 107, 114120, 124125, 136138, and medium, 101103
140142, 144 richness scale, 102
Error Innovation, 2, 18, 41, 49, 84, 8990, 93, 110
central tendency, 37 technology, 93
extreme tendency, 37 Intelligence, 14, 24, 42, 57, 64, 8485, 9495
first impression, 37 Interplay, 66, 70, 75, 79, 82, 88, 9293, 104,
halo, 3637 135137, 143
horn, 37 In vacuum, 46, 52, 77, 104
leniency, 37
recency, 37 K
severity, 37 Kahneman, Daniel, 17, 25, 27, 3839, 54, 58,
spillover, 37 7779, 111
Ethics, 27, 8587, 113, 133, 142 Knudsen, Thorbjrn, 77, 114
Index 161

L possibility violation, 29, 31, 38

Laland, Kevin, 39, 109, 111
Leibniz, Gottfried W., 4647
Less-is-more, 55
Logic of adaptiveness
and change, 89
and context, 88
and fitness, 93
Logic of appropriateness, 5253, 56, 72,
87, 141
Logic of consequences, 52, 56, 87, 141
M
Magnani, Lorenzo, 18, 7879, 8586, 95, 133
Mapping, 45, 56, 83, 137138, 141
Maps, 2739, 4159, 78, 137, 144
March, James G., 18, 2425, 5152, 59,
7778, 94
Medium(s), 33, 6768, 77, 9798, 101107,
110, 117, 119, 123
Mind, 4, 12, 2829, 34, 37, 39, 43, 46, 54, 57,
6471, 73, 75, 7779, 84, 86, 92, 98,
100, 109, 111, 116, 121, 125127, 133
extended, 7374, 77
Morality
and decision making, 86
and distributed cognition, 8587
and through-doing, 8687
Morgenstern, Oscar, 2, 18, 38, 78, 93
N Subjective expected utility (SEU), 2, 3, 63
Neoclassical
economics, 5, 51, 54, 76, 116
model, 14, 21
Newell, Alan, 46, 57, 95
O Through doing
Organization
decision-making, 135
docile, 113133
Organizational
behavior, 1, 3, 14, 39, 45, 5658, 7677,
85, 109110, 123, 125126, 132,
141142
change, 92, 94
culture, 79, 124125, 130, 132, 140
decision making, 69, 73, 138
routine, 10
reflection effect, 30, 38
R
Rationality
adaptive, 86
bounded, 1, 34, 1925, 2739, 4159,
6364, 68, 7374, 7678, 8183, 85,
93, 95, 114, 129, 137138, 143144
ecological, 55, 144
full, 20, 25, 63, 77, 93, 97, 137
limited, 2325
neoclassical model of, 14
procedural, 2022, 24, 41
substantive, 5, 21, 41
Representation, 45, 4852, 56, 66, 72, 7577,
79, 103, 127
Re-projecting, 136
Richardson, Ken, 24, 64, 78
Routine, 10, 18, 33, 7071, 110, 121
S
Scott, W. Richard, 57, 142, 145
Simon, Herbert A., 1, 34, 6, 1319, 21,
2325, 46, 52, 5455, 57, 59, 6379,
81, 93, 95, 113114, 116, 118, 129,
131132, 135, 137, 143144
Smart interplay, 66
Social responsibility, 126128, 131, 133, 141
Sunstein, Cass R., 3839, 110111
T
Thagard, Paul, 57, 7879
Thaler, Richard H., 3839, 108, 110111
logic, 6870, 87, 91
rationality, 78
Time, 3, 9, 1315, 2122, 2425, 32, 3435,
4243, 4546, 4851, 5558, 64,
6672, 77, 82, 84, 86, 88, 90, 98,
100102, 107108, 114118, 130, 133,
136, 142143
Todd, Peter M., 5657, 7778, 144
Toolbox paradigm, 5456
Tversky, Amos, 25, 27, 3839, 54, 77

P V
Peirce, Charles S., 13, 18 Von Neumann, John, 2, 18
Plasticity, 67, 75, 78, 8182, 138
Problem solving, 46, 7475, 82, 143 W
Prospect theory Wilson, Robert A., 73, 7879
certainty effect, 2731, 38 Woods, John, 49, 5758, 78