The

Neuroscience of
RISKY Decision
Making

NeuroscienceTitlePP.indd 1 11/14/13 2:00 PM

 Bronfenbrenner Series on the Ecology
of Human Development

Chaos and Its Influence on Children’s Development: An Ecological Perspective

Edited by Gary W. Evans and Theodore D. Wachs

Research for the Public Good: Applying the Methods of Translational Research to
Improve Human Health and Well-Being
Edited by Elaine Wethington and Rachel E. Dunifon

The Neuroscience of Risky Decision Making

Edited by Valerie F. Reyna and Vivian Zayas

13490-00_FM-3rdPgs.indd 2 11/15/13 1:42 PM

 The
NeuroscieNce of
risKY DecisioN
MaKiNg
Edited by
Valerie F. reyna and Vivian Zayas

American Psychological Association • Washington, DC

NeuroscienceTitlePP.indd 2 11/14/13 2:00 PM

 Copyright © 2014 by the American Psychological Association. All rights reserved. Except as
permitted under the United States Copyright Act of 1976, no part of this publication may
be reproduced or distributed in any form or by any means, including, but not limited to, the
process of scanning and digitization, or stored in a database or retrieval system, without the
prior written permission of the publisher.

Published by To order
American Psychological Association APA Order Department
750 First Street, NE P.O. Box 92984
Washington, DC 20002 Washington, DC 20090-2984
www.apa.org Tel: (800) 374-2721; Direct: (202) 336-5510
Fax: (202) 336-5502; TDD/TTY: (202) 336-6123
Online: www.apa.org/pubs/books
E-mail: [email protected]

In the U.K., Europe, Africa, and the Middle East, copies may be ordered from
American Psychological Association
3 Henrietta Street
Covent Garden, London
WC2E 8LU England

Typeset in Goudy by Circle Graphics, Inc., Columbia, MD

Printer: Edwards Brothers Inc., Ann Arbor, MI

Cover Designer: Naylor Design, Washington, DC

The opinions and statements published are the responsibility of the authors, and such
opinions and statements do not necessarily represent the policies of the American
Psychological Association.

Library of Congress Cataloging-in-Publication Data

The neuorscience of risky decision making / edited by Valerie F. Reyna and Vivian Zayas. —
First edition.
pages cm
Includes bibliographical references and index.
ISBN 978-1-4338-1662-8 — ISBN 1-4338-1662-8 1. Risk-taking (Psychology)
2. Decision making. I. Reyna, Valerie F., 1955- II. Zayas, Vivian.
BF637.R57N48 2014
153.8'3—dc23
2013034272

British Library Cataloguing-in-Publication Data

A CIP record is available from the British Library.

Printed in the United States of America

First Edition

http://dx.doi.org/10.1037/14322-000

13490-00_FM-3rdPgs.indd 4 11/15/13 1:42 PM

 This series is dedicated to the personal memories and lasting theoretical
insights of our friend, colleague, and mentor, Urie Bronfenbrenner. His
thinking about human development has profoundly influenced so many
students and colleagues in multiple areas of enquiry. We hope this series
will provide another vehicle through which Urie’s ideas on the bioecology
of human development can continue to flourish.

13490-00_FM-3rdPgs.indd 5 11/15/13 1:42 PM

13490-00_FM-3rdPgs.indd 6 11/15/13 1:42 PM
 Contents

Contributors.................................................................................................. ix
Foreword....................................................................................................... xi
Colin F. Camerer
Preface.......................................................................................................... xv
Introduction to The Neuroscience of Risky Decision Making........................... 3
Valerie F. Reyna and Vivian Zayas

I. Neuroeconomics....................................................................................... 9
Chapter 1. Reward, Representation, and Impulsivity:
A Theoretical Framework for the Neuroscience
of Risky Decision Making................................................ 11
Valerie F. Reyna and Scott A. Huettel
Chapter 2. Behavioral and Neuroscience Methods for Studying
Neuroeconomic Processes: What We Can Learn
From Framing Effects........................................................ 43
Irwin P. Levin, Todd McElroy, Gary J. Gaeth,
William Hedgcock, and Natalie L. Denburg

vii

13490-00_FM-3rdPgs.indd 7 11/15/13 1:42 PM

 II. Neurodevelopment............................................................................... 71
Chapter 3. Risks, Rewards, and the Developing Brain
in Childhood and Adolescence....................................... 73
Barbara R. Braams, Linda van Leijenhorst,
and Eveline A. Crone
Chapter 4. The Adolescent Sensation-Seeking Period:
Development of Reward Processing
and Its Effects on Cognitive Control............................... 93
Beatriz Luna, Aarthi Padmanabhan, and Charles Geier
Chapter 5. Reward Processing and Risky Decision Making
in the Aging Brain......................................................... 123
Gregory R. Samanez-Larkin and Brian Knutson

III. Neuropsychology............................................................................. 143

Chapter 6. Mind and Brain in Delay of Gratification...................... 145
Vivian Zayas, Walter Mischel, and Gayathri Pandey
Chapter 7. The Neuroscience of Dual (and Triple)
Systems in Decision Making.......................................... 177
Samantha M. W. Wood and Antoine Bechara
Index.......................................................................................................... 203
About the Editors...................................................................................... 213

viii       contents

13490-00_FM-3rdPgs.indd 8 12/9/13 11:51 AM

 Contributors

Antoine Bechara, PhD, Department of Psychology and Brain and Creativity

Institute, University of Southern California, Los Angeles; Department
of Neurology, University of Iowa, Iowa City
Barbara R. Braams, MS, Department of Developmental and Educational
Psychology, Leiden University, Leiden, The Netherlands
Colin F. Camerer, PhD, Department of Behavioral Finance and Econom-
ics, Division of Humanities and Social Sciences, California Institute of
Technology, Pasadena
Eveline A. Crone, PhD, Department of Developmental and Educational
Psychology; Institute for Brain and Cognition, Leiden University,
Leiden, The Netherlands; University of Amsterdam, Amsterdam, The
Netherlands
Natalie L. Denburg, PhD, Department of Psychology, University of Iowa,
Iowa City
Gary J. Gaeth, PhD, Department of Marketing, University of Iowa, Iowa City
Charles Geier, PhD, Department of Human Development and Family Studies,
Pennsylvania State University, State College

ix

13490-00_FM-3rdPgs.indd 9 11/15/13 1:42 PM

 William Hedgcock, PhD, Department of Marketing, University of Iowa,
Iowa City
Scott Huettel, PhD, Department of Psychology and Neuroscience, Duke
University, Durham, NC
Brian Knutson, PhD, Departments of Psychology and Neuroscience, Stanford
University, Stanford, CA
Irwin P. Levin, PhD, Departments of Psychology and Marketing, University
of Iowa, Iowa City
Beatriz Luna, PhD, Departments of Psychiatry and Psychology, University
of Pittsburgh, Pittsburgh, PA
Todd McElroy, PhD, Department of Psychology, Appalachian State Univer-
sity, Boone, NC
Walter Mischel, PhD, Department of Psychology, Columbia University,
New York, NY
Aarthi Padmanabhan, MS, Departments of Psychiatry and Psychology,
University of Pittsburgh, Pittsburgh, PA
Gayathri Pandey, MS, Department of Psychology, Cornell University,
Ithaca, NY
Valerie F. Reyna, PhD, Departments of Human Development and Psychol-
ogy, Human Neuroscience Institute, Cornell Magnetic Resonance
Imaging Facility, Cornell University, Ithaca, NY
Gregory R. Samanez-Larkin, PhD, Department of Psychology, Yale Univer-
sity, New Haven, CT
Linda van Leijenhorst, PhD, Department of Education and Child Studies,
Leiden University, Leiden, The Netherlands
Samantha M. W. Wood, Department of Psychology, University of Southern
California, Los Angeles
Vivian Zayas, PhD, Department of Psychology, Cornell University, Ithaca, NY

x       contributors

13490-00_FM-3rdPgs.indd 10 11/15/13 1:42 PM

 foreword
Colin F. Camerer

The topic of risky decision making is important for both personal deci-
sions and for measurement and control in complex social institutions. (Think
of pricing the mysterious “systemic risk” hidden in our highly interlocked
financial systems, or avoiding pandemics.)
Risky decision making is studied by researchers in a wide variety of
disciplines. This book represents a current snapshot of what is known and a
rough guide for future research on how to make sense of the complexity of
risk. The editors’ recipe: Create as much coherence across evidence from dif-
ferent disciplines as possible, and measure as much about underlying neural
mechanisms as you can; then get top researchers working on their respec-
tive frontiers to say what they know in general language that makes shared
understanding easy.
A striking feature of this volume is how risk taking can be construed as
fundamentally different, according to discipline. For traditional economists,
risk is simply variance of outcomes, with no regard to their sign. Psychologists
have shown the additional importance of loss, compared with gain—losses
are often valued about 1.5 to 2 times as much as equivalent gains (unless you
have amygdala damage, in which case traditional economics applies to you;
DeMartino et al., 2010).

xi

13490-00_FM-3rdPgs.indd 11 11/15/13 1:42 PM

 Much more speculative, and interesting, are the roles emotion and
memory play in risky decision making (see Chapter 1). There is good meta-
analytic evidence that insula cortex encodes financial uncertainty, and other
interocepted discomforts such as pain and disgust. Experienced risks are clearly
encoded dually by verbatim and gist (distilled) memories (e.g., Reyna & Far-
ley, 2006), a distinction that is just now migrating to decision neuroscience.
The economists’ view is that how much risk you take is a personal
matter: Who are we to condemn either cave diving or college cheerleading
(which is more dangerous than you think, Amber!)? But many risks aren’t just
personal, because they expose others to harm (e.g., reckless driving, unsafe
sex) or expose oneself to unknown harm.
But how can we say scientifically that somebody took “too much” risk or
took an “unnecessary” risk? To do so, a concept of inhibitory control becomes
useful. In this mechanistic view, healthy risk takers know when to stop and can
inhibit the last throw of the dice. Shannon et al. (2011) reported a remarkable
finding about youths: In the most impulsive incarcerated youths, premotor
areas were positively connected with “default network” areas and negatively
connected with attention and control areas. These struggling adolescents are
always ready to act, and “don’t attend and control”—adequately.
Regardless of how much you know, something in this book will sur-
prise you. For example, if you know that basic psychological performance
on simple attention, sensory, and memory tasks slowly degrades beginning
in one’s 20s, you might then be surprised to find that there are no reli-
ably established differences in risk taking between young and old people.
Happily—for this 50-something scientist—as people age, they show muted
neural response to anticipated loss, and other “positivity” biases, along with
weakened frontostriatal connectivity (see Chapter 5).
Do you know what the default network is (Raichle & Snyder, 2007)?
It’s a circuit including prefrontal and parietal cortex that is reliably active
during rest or “doing nothing special.” You might be surprised by a recent
finding that the default network is activated when one is simply judging the
probability of a state occurring, independent of reward (d’Acremont et al.,
2013). A string of findings like these are interesting clues that the default
network is not doing nothing . . . instead, it is probably doing some evolu-
tionarily important low-effort work (perhaps thinking about the risky world
or the social world).
As a proud neuroeconomist, I credit the contents of volumes like this
one as the source of much of my optimism. When there is genuine eagerness
to think about the brain computing a wide variety of psychological and eco-
nomic constructs (preferably numbers, please), and to treat all kinds of dif-
ferent evidence seriously, progress will be rapid. Of course, rapid progress will

xii       foreword

13490-00_FM-3rdPgs.indd 12 11/15/13 1:42 PM

 make us wince over time, because it means we will know, hopefully within a
decade or two, which half of what we now think is true is false.

References

d’Acremont, M., Fornari, E., & Bossaerts, P. (2013). Activity in inferior parietal and
medial prefrontal cortex signals the accumulation of evidence in a probability
learning task. PLoS Computational Biology, 9, e1002895. doi: 10.1371/journal.
pcbi.1002895
De Martino, B., Camerer, C. F., & Adolphs, R. (2010). Amygdala damage eliminates
monetary loss aversion. PNAS, 107, 3788–3792.
Raichle, M. E., & Snyder, A. Z. (2007). A default mode of brain function: A brief
history of an evolving idea. NeuroImage, 37, 1083–1090.
Reyna, V. F., & Farley, F. (2006). Risk and rationality in adolescent decision-making:
Implications for theory, practice, and public policy. Psychological Science in the
Public Interest, 7(1), 1–44. doi:10.1111/j.1529-1006.2006.00026.x
Shannon, B. J., Raichle, M. E., Snyder, A. Z., Fair, D. A., Mills, K. L., Zhang, D., . . .
Kiehl, K. A. (2011). Premotor functional connectivity predicts impulsivity in
juvenile offenders. PNAS, 108(27), 11241–11245.

foreword       xiii

13490-00_FM-3rdPgs.indd 13 11/15/13 1:42 PM

13490-00_FM-3rdPgs.indd 14 11/15/13 1:42 PM
 preface

Risky decision making occurs in many domains of life, such as eating,

sexual activity, crime, drugs and alcohol use, finances, warfare, and health
care. In each of these domains, crucial outcomes—developing diabetes, HIV
infection, a terrorist attack, going to jail, being pulled over for drunk driv-
ing, losing one’s lifetime savings in a high-risk investment, or dying from a
preventable cancer—are uncertain. Research on the neuroscience of risky
decision making has surged in recent years, reflecting the intersection of
interests in the brain, decision processes, and the welfare of people—often
young people—who suffer or die needlessly as a consequence of their actions.
A goal of this book is to delineate the precise neural mechanisms that affect
risky decision making and how the brain is affected by momentary tempting
situations, chronic dispositional differences, and developmental periods.
Exciting new discoveries suggest that the neural mechanisms underlying
decision making across domains are simultaneously empirically and theoreti-
cally similar but also meaningfully different depending on the domain. Funda­
mental constructs such as self-control, sensitivity to reward, and preference
for risk predict risky decision making across drastically different domains, and
similar areas of brain activation have been associated with each construct

xv

13490-00_FM-3rdPgs.indd 15 11/15/13 1:42 PM

 across domains. New constructs are being introduced to explain and predict
risky behavior in multiple domains, such as neural representations of the gist
of risky options. From a developmental perspective, risky decision making
changes from childhood to old age, with the greatest risky behavior occurring
during adolescence and young adulthood, and these behavioral changes map
onto massive changes in brain structure. And, individual differences that
were first observed in childhood manifest in differences in brain activation
40 years later. Moreover, the differences in risky decision making across ages
and individuals that are assessed in laboratory behaviors and in brain scanners
correlate with real-world outcomes ranging from income to incarceration.
Despite consistency in brain function and structure across domains,
risky decision making also differs from one domain to another behaviorally
and neurally. The cultural condemnation attached to over­eating, smoking,
and crime varies, as does the availability of different kinds of rewards, and the
brain distinguishes among them. People who take seemingly unreasonable
risks with sex can be careful with money, and vice versa—in spite of the fact
that neurobiological factors such as sensation seeking predict risk taking in
both domains.
Research in the neuroscience of risky decision making has now reached
that productive state in which there is sufficient evidence to take stock and
present frameworks to integrate the seemingly inconsistent, conflicting,
and contradictory findings in this literature. It seems remarkable that this
would be the first book to do so in such an active area of research. However,
this makes sense when one considers that the first study of any kind using
functional magnetic resonance imaging occurred as recently as the 1990s.
Research on risky decision making has benefited from the fact that
scholars from diverse disciplines have been brought together by a shared fas-
cination with these amazing new technologies, such as magnetic resonance
imaging, which allow them to study the functioning brain without disruptive
or invasive methods. For example, the Society for Neuroeconomics, which
had its inaugural meeting in the early 2000s, fosters interaction “among
scholars from the psychological, economic, and neural sciences” (http://www.
neuroeconomics.org/society). Advances in the neuroscience of risky decision
making are also presented increasingly at other national and international
meetings, such as the Society for Neuroscience, the Society for Research in
Child Development, and the American Psychological Association (APA).
This book, the third in the APA Bronfenbrenner Series on the Ecol-
ogy of Human Development, traces its inspiration to Urie Bronfenbrenner’s
1996 book entitled The State of Americans, coauthored with Peter McClelland,
Elaine Wethington, Phyllis Moen, and Stephen Ceci. The 1996 book presented
a compilation of statistics on crime, poverty, family well-being, education, and

xvi       preface

13490-00_FM-3rdPgs.indd 16 11/15/13 1:42 PM

 the like, with the goal of using “hard facts” to translate social scientific insight
into a form that would influence policy and practice. Although this research
is still in its infancy, neural processes of risky decision making (as discussed in
the current volume) have been shown to be related to these kinds of important
social and health statistics.
One assumption of Bronfenbrenner’s 1996 book was that the adop-
tion of policies and practices based on scientific evidence would improve the
well-being of Americans (and others). A second assumption was that social
scientists should be at the table with policymakers in solving practical social
problems because research has practical implications for improving health
and development in the population. Nowhere is that potential more evident
than for research on risky decision making.
In his final book in 2004 (Making Human Beings Human: Bioecological
Perspectives on Human Development), Bronfenbrenner combined biological
perspectives with his ecological framework for human development, paying
particular attention to the different contexts in which biological and brain
development take place—at individual, family, institutional, and societal
levels. In recognition of his lifetime contributions to applied psychology
as a founder of Head Start and to understanding critical social problems,
Bronfenbrenner received the James McKeen Catell Fellow Award from the
Association for Psychological Science (previously the American Psycho-
logical Society) in 1993. In 1996, the APA Award for Lifetime Contribution
to Developmental Psychology in the Service of Science and Society Award
was renamed, in honor of Bronfenbrenner, the Urie Bronfenbrenner Award
for Lifetime Contribution to Developmental Psychology in the Service of
Science and Society.
Urie Bronfenbrenner continues to influence his colleagues at Cornell
University and around the world. In particular, this volume was inspired by a
meeting supported by the Bronfenbrenner Center for Translational Research
at Cornell University focused on this topic that gathered leading scholars from
around the world. To see their talks and be similarly inspired, access http://
mediasite.video.cornell.edu/Mediasite/Catalog/catalogs/BCTR_review.aspx.
The meeting was also a project of the Cornell Judgment, Decision Mak-
ing, and Social Behavior interdisciplinary group and the Center for Behavioral
Economics and Decision Research (http://socialsciences.cornell.edu/judgment-
decision-making-and-social-behavior/). We gratefully acknowledge the assis-
tance of John Eckenrode, The Bronfenbrenner Advisory Committee, Carrie
Chalmers, Karene Booker, and Thomas Craig.
The current book emerges at a key time in understanding the basic
mechanisms underlying risky behavior, a focus of this volume, which is an
essential step in the process of translation of research findings into policy and

preface       xvii

13490-00_FM-3rdPgs.indd 17 11/15/13 1:42 PM

 practice. Without this understanding, policymakers and practitioners can mis-
apply research and waste precious resources, while important problems are left
unaddressed. We hope that this trend of increasing research on basic mecha-
nisms of risky decision making continues and that this volume contributes
to efforts to translate basic science about the neuroscience of risky decision
making into policy and practice.
This area of research is evolving rapidly, and so is the audience for it.
At this moment, we anticipate that the target audience for this book includes
scholars approaching risky decision making from multiple perspectives: neuro­
scientists, neuropsychologists, clinicians, psychologists (developmental, social,
and cognitive), economists and other social scientists, legal scholars and crim-
inologists, and professionals in public health and medicine. May science con-
tinue to supplant superstition when human welfare is at stake.

xviii       preface

13490-00_FM-3rdPgs.indd 18 11/15/13 1:42 PM

 The
Neuroscience of
RISKY Decision
Making

NeuroscienceTitlePP.indd 1 11/14/13 2:00 PM

13490-01_Intro-3rdPgs.indd 2 11/15/13 1:42 PM
 Introduction to the
Neuroscience of Risky
Decision Making
Valerie F. Reyna and Vivian Zayas

The processes by which individuals make decisions under uncertainty has

important implications for real-world outcomes in law, medicine, economics,
clinical psychology, and public policy, to name a few. Despite the explosion
of research on risky decision making over the past four decades, many ques-
tions remain: What are the neurobiological, psychological, and sociocultural
factors that influence risky decision making either in isolation or in com-
bination? How do the effects of these processes on decision making differ
across development? And, what are their implications for problematic behav-
iors and health? The primary goal of this book is to address these questions,
thereby advancing basic understanding and scientific theory about the brain
mechanisms underlying risky decision making across the life span.
Understanding the mechanisms—from brain to behavior—of risky deci-
sion making is essential in paving the way for translation of basic science into
policy and practice. A second goal of this book is to encourage intellectual
integration of existing diverse approaches. To date, the factors that influence

http://dx.doi.org/10.1037/14322-001
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

13490-01_Intro-3rdPgs.indd 3 11/15/13 1:42 PM

 risky decision making have been studied from a number of perspectives that
span the spectrum of neuroscience, psychology, and behavioral economics.
These approaches address the same general topic. However, each focuses on
a different level of analysis (neural, individual, and societal, respectively),
different units of analysis (behaviors, cognitions, affective states, neural pat-
terns of activation), different domains (e.g., problematic behaviors, financial
decision making, medical and health-related decisions), and different types
of questions (e.g., normative processes to individual differences to develop-
mental changes). Despite significant scientific advances, researchers in each
area have remained relatively isolated from one another.
The speed of scientific advances, especially in neuroscience, calls for
integrating findings from various levels of analysis to provide a more compre-
hensive view of the process underlying risky decision making as they operate
over the lifespan. This intellectual integration is another goal of the book.
Building on the recent surge of research on the neuroscience underlying risky
decision making, the book brings together scholars with expertise that blends
approaches from multiple disciplines to promote the development of a com-
prehensive and contextually sensitive model of risky decision making across
the life span.
In the chapters that follow, leading neuroeconomists, neuroscientists,
and social scientists discuss the latest findings and theoretical perspectives
on risky decision making, reviewing such topics as the changing impact of
rewards and punishments at different ages (from early childhood through
old age); the role of emotional regulation and self-control abilities as well as
individual differences in personality in contributing to chronic difficulties in
risky decision making; and the social, cognitive, biological, and developmen-
tal factors that shape risky behavior.
The book is organized into three sections that reflect active areas
of research on the neural underpinnings of risky decision making: neuro­
economics, neurodevelopment, and neuropsychology. Neuroeconomics is
the study of the brain making economic judgments and decisions. We lead off
the book with interdisciplinary approaches to neuroeconomics because of the
foundational role that economics has played in defining the basic phenom-
enon of risky decision making. Hence, from the outset, we compare tenets of
psychology and economics that have provided the context for neuroscience
research on risky decision making over the past decade.
The second section on neurodevelopment describes a special focus of
research on risky decision making, namely, how risky decision making changes
dramatically from childhood to adolescence to adulthood to old age as a func-
tion of maturational and experienced-based changes in the brain. The third
section presents influential research in neuropsychology and individual dif-
ferences, including the famous paradigms known as the “marshmallow test”

4       reyna and zayas

13490-01_Intro-3rdPgs.indd 4 11/15/13 1:42 PM

 (that incorporates the economic notion of temporal or delayed discounting)
and the Iowa Gambling Task. Both tasks call on overlapping neural systems
and elicit behavior that predicts risk taking in the real world. Throughout
the book, many of the authors weave elements of neuroeconomics, neuro­
development, and neuropsychology together.
Specifically, in Part I on neureconomics, fundamental economic phe-
nomena such as risk preferences and framing effects (i.e., shifting from risk-
seeking to risk-avoiding when options are equivalent but worded differently
as gains versus losses) are discussed. In Chapter 1, Reyna and Huettel describe
the implications of risky decision making for law, medicine, and public health
and contrast the definitions of risky decision making in economics versus psy-
chology. They explain that economists focus on risk in the sense of variance
in outcomes (hence, by this definition, risk takers are people who tolerate
uncertainty in outcomes), whereas psychologists focus on risk in the sense of
the particular case in whice uncertain outcomes are bad and detrimental to
well-being (e.g., financially, legally, or medically) and highlight the effects of
emotion and immaturity. Reyna and Huettel then summarize neuroscientific
evidence ranging from ventral striatal reward circuitry to involvement of
the default network in impulsivity. Grounded in this evidence, they intro-
duce a preliminary integrative theoretical framework encompassing neural
substrates of emotional salience, memory representations of options, and
decision conflicts as people experience internal clashes between competing
strategies for making risky decisions.
In Chapter 2, Levin, McElroy, Gaeth, Hedgcock, and Denburg sum-
marize research on framing effects, which challenge axiomatic principles of
economic theories of risk preference. Levin et al. present a process-oriented
perspective that explains why people take risks and why their risk preferences
shift in different circumstances. Summarizing and integrating research using
behavioral and biological/neurological measures, such as neuroimaging, eye-
tracking, circadian rhythms, and life span developmental techniques, they
characterize the cognitive and emotional underpinnings of risky decisions
across tasks and individuals.
In Part II on neurodevelopment, the authors address how developmen-
tal differences in brain functioning—from childhood to old age—are asso-
ciated with risky decision making. In Chapter 3, Braams, van Leijenhorst,
and Crone review the conditions under which developmental differences in
risky behaviors are found among children, adolescents, and adults and how
neuroscience elucidates the neural underpinnings of these differences. They
also describe important neurodevelopmental frameworks of risk and reward
processing, including the dual-processing network, the imbalance model,
and the social information processing network approaches. They explore
such issues as why the striatum is sometimes overactivated and other times

introduction      5

13490-01_Intro-3rdPgs.indd 5 11/15/13 1:42 PM

 underactivated in adolescents; the role of hormones and individual differ-
ences; and the role of social factors, such as peer relations and peer pressure.
They conclude with a new working model that explains how the striatum is
influenced by environmental context and how it connects to other regions
in the brain, which ultimately influence risk-taking behavior. This model is
informative for policymakers and educational practitioners because it identi-
fies what can be expected of people depending on their age.
In Chapter 4, Luna, Padmanabhan, and Geier focus on sensation seek-
ing in adolescence, which is known to increase during the pubertal period
across different societies and different species and often results in risk-taking
behaviors that undermine survival. Despite being a period of peak physical
health, adolescence is a time of increased mortality rates due in great part to
risky behaviors such as substance abuse, unprotected sex, and extreme sports.
To account for the paradox of heightened fitness and increased mortality dur-
ing adolescence, Luna et al. propose that increased sensation seeking in ado-
lescence is an adaptive mechanism. Sensation seeking affords the ability to
explore the environment and expose the individual to information, thereby
modeling the maturing brain. Sensation seeking may also underlie motiva-
tion for seeking independence, which supports an adolescent’s transition to
adult levels of maturity and responsibility.
Drawing on evidence from animal models and human studies, Luna et al.
propose that developmental changes in neurotransmitter availability (along
with pubertal changes) are a possible mechanism underlying the heightened
propensity for sensation seeking in adolescence. Like Braams et al., they explore
discrepancies in neuroimaging studies of engagement of reward-related brain
systems in adolescence, drawing on developmental differences during stages
of reward processing and age-related differences in value assessment. Luna
et al.’s model of increased sensation seeking in adolescence has implica-
tions for juvenile law and education. Concluding the section on neuro-
development, in Chapter 5, Samanez-Larkin and Knutson discuss emerging
research on risky decision making in the aging brain. Despite the aging of the
world population and the importance of decision competence in old age (e.g.,
retirement and end-of-life decisions), remarkably little research has focused
on how aging might influence risk and reward processing. Samanez-Larkin
and Knutson review studies that examine how age influences psychological
and neural responses to financial incentives and risks. Early findings from
this literature suggest that aging may influence the structure and function of
neural circuits implicated in incentive processing and risky decision mak-
ing (e.g., the ventral striatum, the anterior insula, the prefrontal cortex)
and that the consequences of these changes for choice apply to both labo-
ratory and real-world settings. In addition to informing theory about the
impact of affect and cognition on choice, these novel findings imply that

6       reyna and zayas

13490-01_Intro-3rdPgs.indd 6 11/15/13 1:42 PM

 understanding how the aging brain processes incentives may eventually
inform the design of more targeted and effective decision aids for individuals
of all ages.
In Part III on neuropsychology, Zayas, Mischel, and Pandey provide a
compelling review of 40 years of research on delay of gratification. By age 4,
children differ in the ability to delay gratification, and such individual dif-
ferences predict risky decision and problematic behaviors across develop-
ment, such as higher social competence, higher academic achievement (SAT
scores), and lower substance abuse and body mass index (BMI). Recent find-
ings provide empirical evidence that the remarkable long-term continuity in
delay of gratification is rooted in individual differences in prefrontal cortical
activity as well as affect-related brain circuits—a proposition consistent with
a growing body of neuroscientific work in risky decision making.
Specifically, differences in preschool delay of gratification ability were
observed most clearly in adulthood in tasks that involve inhibiting responses
to rewarding stimuli (e.g., smiling face indicating approval) but not to neutral
stimuli. Moreover, in imaging work, the preschool delay predicted greater
activation of the inferior frontal gyrus, a structure in the prefrontal cortex
(PFC) recruited when resolving conflict between representations and motor
responses, on trials that required inhibiting a response. Preschool delay abil-
ity also predicted less activation of the ventral striatum in response to the
rewarding stimulus. These authors conclude by arguing that a key benefit of
enacting effortful self-control is that it lessens the tempting aspects of the
cues to be inhibited and thus makes the very act of delaying gratification
easier, essentially lessening the need to exert effortful self-control.
In the final chapter on neuropsychology, Wood and Bechara weigh in on
the debate about “single system” versus “dual system” models for valuation in
risky decision making. Using clinical evidence, they argue that not only are
the traditional two processes supported by the clinical facts, but, in addition,
the evidence points to a third process, thus calling for the notion of “triple pro-
cess” models. This third system, involving the insula, translates homeostatic,
bodily signals into feelings of craving.
That is, Wood and Bechara present evidence that many clinical condi-
tions associated with poor impulse control and poor decision making are the
product of an imbalance between two separate but interacting neural systems:
(a) an impulsive, amygdala and striatum-dependent, neural system that pro-
motes automatic and habitual behaviors and (b) a reflective, PFC-dependent,
neural system for decision making. These neural systems map onto the psycho­
logical systems named “System 1” and “System 2,” respectively. System 1 is
defined as quick, automatic, and associative in its response, while System 2
is slow, effortful, reflective, and “rational.” The reflective system controls
the impulsive system via several mechanisms. However, this control is not

introduction      7

13490-01_Intro-3rdPgs.indd 7 11/15/13 1:42 PM

 absolute; hyperactivity within the impulsive system can override the reflec-
tive system. The impulsive system is reminiscent of the described properties
of System 1, which require a long time to build associations, but once those
associations are made, they are rapid and difficult to override.
However, critically, going beyond standard dual-processes models, Wood
and Bechara suggest that the insula plays a key role in modulating the dynam-
ics of these two systems. While most prior research has focused on the impul-
sive versus reflective or System 1 versus System 2, they suggest that the insula
plays a key role in modulating the dynamics of these two systems. More specifi-
cally, the insula or “craving” system potentiates the activity of the impulsive
system, weakens the goal-driven cognitive resources that are needed for the
normal operation of the reflective system, or does both. Thus, when physi-
ological states that involve deprivation, withdrawal, stress, anxiety, or any
condition associated with homeostatic perturbation are considered, a third
process (the insula) comes to the fore with direct impact on the functionality
of the traditional dual systems.
As a whole, the book provides a comprehensive and up-to-date over-
view of major approaches to the neuroscience of risky decision making. The
authors summarize cutting-edge research on the neuroeconomic, neurode-
velopmental, and neuropsychological factors that explain and predict risky
decision making. Diverse findings from structural and functional neuroim-
aging, as well as behavioral and neurophysiological studies, on populations
ranging from young children to old age, are integrated to provide a scientific
framework for understanding causal mechanisms underlying risky decision
making across the life span. This work has important implications not only
for reconceptualizing and reforming the next phase of research on the neuro-
science of risky decision making but also for informing practice and policy in
law, medicine, and public health.

8       reyna and zayas

13490-01_Intro-3rdPgs.indd 8 11/15/13 1:42 PM

 I
Neuroeconomics

13490-02_PT1_Ch01-3rdPgs.indd 9 11/15/13 1:42 PM

13490-02_PT1_Ch01-3rdPgs.indd 10 11/15/13 1:42 PM
 1
Reward, Representation, and
Impulsivity: A Theoretical
Framework for the
Neuroscience of Risky
Decision Making
Valerie F. Reyna and Scott A. Huettel

We provide an overview of neuroscience research on risky decision making,

organizing findings in an integrative theoretical framework aimed at elucidating
mechanisms that drive behavior. The concept of risk has been used to describe
a variety of influences on decisions—including both the variance of outcomes
and the potential for a negative outcome—each of which may have a distinct
influence on neural processing. Armed with these distinctions, we examine
neural substrates of reward and valuation, reviewing evidence that the ventro­
medial prefrontal cortex (vmPFC) computes a common currency signal that
allows comparison of rewards across domains (e.g., food and money). This com­
mon currency signal is modulated by the variables that shape decision making,
such as gains, losses, and their probabilities. We review evidence that subjective
feelings about the uncertainty and valence of outcomes (e.g., risk and loss aver­
sion) follow from signals in the insula and that signals associated with uncertainty

Preparation of this manuscript was supported in part by the National Institutes of Health under award
number R21CA149796 and R01NR014368-01 to the first author. The content is solely the responsibility
of the authors and does not necessarily represent the official views of the National Institutes of Health.
http://dx.doi.org/10.1037/14322-002
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

11

13490-02_PT1_Ch01-3rdPgs.indd 11 11/15/13 1:42 PM

 can be distinguished from signals of emotional salience in the amygdala. Rep­
resentations of options in vmPFC/medial orbitofrontal cortex serve as inputs
to a comparison process in anterior cingulate cortex (ACC)/ dorsomedial pre­
frontal cortex (dmPFC), which reflects decisional conflict. Activation in ACC/
dmPFC is greater when choices conflict with otherwise dominant strategies,
such as gist-based simplification versus verbatim-based trading off, triggering
cognitive control mechanisms in dorsolateral PFC. When value signals are
translated into actions, prefrontal signals influence processing of neurons in
the posterior parietal cortex whose activity is consistent with drift-accumulator
models of choice. This tentative process model differentiates several indepen­
dent contributors to risk-taking behavior and identifies levers of behavioral
change that could be used to prevent unhealthy decisions.
People often die from risky choices. Choices to initiate smoking, take
illicit drugs, eat unhealthy foods, drive recklessly, and drink excessively con-
tribute to cancer, heart disease, and trauma from traffic accidents (Reyna &
Farley, 2006). Lifetime prevalence of alcohol dependence alone approaches
20% (Bloom, 2010). Crime is less a phenomenon of psychopathy (although
psychopaths commit a disproportionate percentage of crime) than of age-
related risk taking, peaking in adolescence and young adulthood. Indeed,
homicide is the second leading cause of death among 15- to 19-year-olds, after
unintentional injury; firearms are the instrument of death in 85% of cases
(Centers for Disease Control and Prevention, 2012). Therefore, the “bur-
den of illness”—the severity and prevalence of harm to oneself and others—
produced by risky decision making is enormous (Arnett, 1992; Reyna &
Rivers, 2008).
Risky decision making is a major public health problem, as illustrated
by this brief litany of statistics, but it is also a problem of law, medicine,
social relations, and personal finance. In the domain of personal finance, for
example, people commit errors of risk aversion and risk seeking, avoiding
riskier investments with better lifetime returns and spending down reserves
rather than saving enough for retirement (Benartzi & Thaler, 1995; Rick,
Cryder, & Loewenstein, 2008). Although risky behaviors can differ across
domains of life, they also correlate with one another, peak at roughly simi-
lar ages, and activate a common neural circuitry of risk and reward valua-
tion (Jessor, 1991; D. J. Levy & Glimcher, 2011; Porcelli & Delgado, 2009;
Reyna, Chapman, Dougherty, & Confrey, 2012). As we discuss in this chap-
ter, understanding the risk and reward mechanisms in the brain, and their
development, is key to unraveling the mystery of irrational risk taking in
real life.
Laboratory experiments are an essential element in this understand-
ing. Risky decisions made in the laboratory predict real-world decisions,

12       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 12 11/15/13 1:42 PM

 although far from perfectly (e.g., Galvan et al., 2006; Galvan, Hare, Voss,
Glover, & Casey, 2007; Lejuez, Aklin, Zvolensky, & Pedulla, 2003; Parker
& Fischhoff, 2005; Pleskac, 2008; Reyna et al., 2011; Steinberg, Cauffman,
Woolard, Graham, & Banich, 2009; Stout, Rock, Campbell, Busemeyer, &
Finn, 2005). Alternatively, using real-world behaviors for prediction of dif-
ferent risky behaviors does not solve the problem of imperfect prediction,
and typically such relationships are difficult to interpret because they are
confounded (i.e., multiple correlated factors contribute to risky behaviors).
(Using some real-world risky behaviors to “predict” other risky behaviors
is satisfactory for actuarial purposes, but it sheds little light on causation or
mechanisms.) Consequently, when the proper methodological controls are
used, laboratory tasks provide crucial insight into the mechanisms of risky
decision making. The main challenge in developing effective practice and
public policy is understanding these mechanisms of risky decision making to
determine (a) whether they apply in a given context and (b) how they com-
bine to produce behavior.
Progress in this endeavor hinges on distinguishing among related
concepts such as impulsivity, reward sensitivity, risk perception, risk pref-
erence, sensation seeking, and how rewards are represented and remem-
bered; many tasks used in research conflate these concepts. One thesis of
our broader argument is that behavioral evidence that distinguishes these
concepts is essential to inform the design of tasks and analyses, and, hence,
the inter­pretation of neuroscience data. Despite generally impressive rigor,
the amount of neuroscience data regarding any one question is relatively
limited, and what data there are is sometimes not sufficiently theoreti-
cally grounded to be interpreted unambiguously. In addition, neuroscience
research can be enhanced through a greater emphasis on hypothesis testing
and by experimental designs that complement correlational approaches, as
found in relevant behavioral research (e.g., see Huettel, Song, & McCarthy,
2009).
Furthermore, although it is possible to wait until more data arrive, research
on both brain and behavior will progress more expeditiously if they are brought
together. More fundamentally, the functions of the brain cannot be under-
stood without a process analysis of the behavioral tasks they support, and a
deep understanding of behavior cannot be achieved in ignorance of the rich
new world of neuroscience research. In short, this chapter is a prolegomenon
to a fully verified theoretical framework for the neuroscience of risky decision
making. The interpretations we offer are preliminary, but they respond to the
challenge of developing a functional taxonomy that maps decision behavior
onto its underlying processes, both psychological and neuro­scientific (Huettel,
2010; Poldrack et al., 2011).

reward, representation, and impulsivity      13

13490-02_PT1_Ch01-3rdPgs.indd 13 11/15/13 1:42 PM

 Definitions and Distinctions

There are two main approaches to defining risky decision making: eco-
nomic and psychological. For economists, risk is about variance of known
outcomes (i.e., uncertain events, not uncertainty in the epistemic sense).
Within the expected utility framework of economics, risk preference (some-
times called risk attitude) refers to the shape of the utility function estimated
from a series of risky choices, with the choice between a sure versus a risky
option of equal expected value being a canonical example (von Winterfeldt
& Edwards, 1986). Expected utility is a nonlinear transformation of expected
value into subjective value.
For example, consider a choice between (a) winning $1,000 for sure
versus (b) a .50 probability of winning $2,000 and .50 probability of winning
nothing. Risk-averse people would prefer the sure thing (a) over the risky
gamble (b). Comparison of option (a) with option (b) indicates that the
second option has more variance in outcomes than the first. (It may be useful
to think of the sure option as .50 × $1,000 + .50 × $1,000, illustrating that
outcomes do not vary when they are sure.) When the expected values of each
option are equal to one another, as in this example, sure options ($1,000)
are mathematically closer to the origin (i.e., closer to zero) than non–zero
gamble outcomes ($2,000). Thus, if people prefer the sure option, their “util-
ity” function for money is assumed to have a concave curvature—$2,000
deviates more (in a downward direction) from its objective value than $1,000
does from its objective value.
In neuroscience research on decision making, the values of decision
options are talked about in at least two ways: the value of the outcomes
($1,000 and $2,000 in our example) and the value of the whole option,
known as expected value, which is a weighted average of outcomes and prob-
abilities (i.e., $1,000 × 1.0 = $1,000 or .5 × $2,000 + .5 × 0 = $1,000) (e.g.,
D. J. Levy & Glimcher, 2012; Zhang & Hirsch, 2013). When outcomes are
positive, they are referred to as rewards; expected value captures the idea that
probability and reward trade off to produce overall value (e.g., stocks have
greater expected value but greater risk than bonds).
Expected value corresponds roughly to people’s intuitions about proba-
bilities and outcomes. If people did not weight outcomes by their probabilities,
they would feel the same way about a lottery ticket that paid off with a $1,000
prize for sure and one that paid off $1,000 with a .50 probability. Even if a per-
son has never heard of the concept of expected value, she would not prefer a
.5 chance (e.g., flipping a coin) to win $1,000 over a sure $1,000. Animals and
children who have not yet learned to multiply also make such trade-offs of risk
and reward intuitively (Reyna & Brainerd, 1994). Thus, the formal concepts
of variance and expected value have some psychological reality.

14       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 14 11/15/13 1:42 PM

 For economists, none of the outcomes need to be “bad” for an option
to be considered risky. The second option (or gamble) in our example could
be a .50 chance of winning $1,200 and a .50 chance of winning $800, and it
would still be riskier in the economics sense compared with a sure $1,000.
For psychologists, however, behavioral choices that lead to the possibility of
loss or harm are referred to as risky (Fox & Tannenbaum, 2011). As in the
economists’ definition of risky, these bad outcomes (losses or harms) are uncer­
tain: Crimes go unpunished, most people who drink never become alcoholics,
unprotected sex does not necessarily result in HIV infection, and many pieces
of pie do not necessarily produce death or disease. Individuals differ in their
willingness to tolerate this uncertainty.
Taking undue risks can be considered pathological and, thus, classified in
the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM–IV–TR;
e.g., eating disorders or antisocial personality disorder; American Psychiatric
Association, 2000) or the International Statistical Classification of Diseases and
Related Health Problems (10th ed.; ICD–10; World Health Organization, 2010).
However, risky choices can also be socially approved, for example, when foot-
ball players risk serious injury, or they can be an ordinary part of daily life, as in
high-risk jobs such as logging or washing windows of skyscrapers. A question
that has guided neuroscience research is how the brain processes this variance
or uncertainty, in particular, how brain systems support compensatory trade-
offs between a safer, lower value option and a riskier, higher value option
(De Martino, Kumaran, Seymour, & Dolan, 2006; Huettel, 2006; Tom, Fox,
Trepel, & Poldrack, 2007; Venkatraman, Payne, Bettman, Luce, & Huettel,
2009). Given the uncertainty inherent in risk taking, it can be difficult to draw
a clear line between healthy and unhealthy risk taking, but that distinction is
the crux of adaptive behavior (Reyna & Farley, 2006).
The psychological definition of risky behavior conflates two aspects of
decision making, however, that are separable theoretically and functionally
in the brain: risk and loss attitudes (e.g., Yechiam & Telpaz, 2013). The
example of the type that we have discussed (a sure gain vs. a gamble of equal
expected value) is frequently used to illustrate risk aversion because most
people prefer the sure option to the gamble of equal expected value. People
find the uncertainty in the gamble aversive despite its mathematical equiva-
lence over repeated trials (i.e., if the gamble were enacted repeatedly, the
average winnings or expectation would be $1,000). However, risk aversion
is not the same as loss aversion.
Loss aversion does not just mean that losses (e.g., losing money, people
dying rather than being saved) are negative, but that losses hurt more than
gains of similar magnitude feel good (Kahneman & Tversky, 2000; there is also
evidence that people pay closer attention to losses; see Yechiam & Telpaz,
2013). So, most people will not accept a bet on the flip of a coin in which

reward, representation, and impulsivity      15

13490-02_PT1_Ch01-3rdPgs.indd 15 11/15/13 1:42 PM

 the amount to win is equal to the amount to lose (e.g., win $10 if heads, lose
$10 if tails). They usually require about a 2:1 ratio of wins to losses to accept
the bet (e.g., win $20 if heads, lose $10 if tails). Loss aversion refers to an
asymmetry in the impact of gain versus loss outcomes, rather than risk or
uncertainty (Willemsen, Böckenholt, & Johnson, 2011).
In sum, risk aversion can be defined as preferring a sure thing to a roughly
mathematically equivalent (or superior) gamble, and risk taking as the oppo-
site preference. Economists focus on people’s taste for variance, which classic
theories identify with nonlinear functions of outcomes, such as money. In the
economists’ view, two gambles can differ in risk if one has higher variance in
outcomes than another. Psychologists focus on tolerance for uncertain bad
outcomes, such as losses, and the implications of risk attitudes for life out-
comes. However, risk and loss aversion can be distinguished empirically—and
loss aversion can be demonstrated when options are certain. Both economists’
and psychologists’ definitions of risk attitude have influenced neuroscience
research (e.g., Schonberg, Fox, & Poldrack, 2011).
According to either economists’ or psychologists’ definitions, most peo-
ple have an aversion to uncertainty. When people demonstrate a high toler-
ance for uncertain outcomes, whether inside or outside of the laboratory, what
are the brain processes that underlie their risk preferences? Do they process
risk or reward differently, fail to trade them off properly, or lack the ability to
control their attraction to rewards? Moreover, are the nature of thinking and
underlying brain processes qualitatively different among those who seek versus
avoid risks, especially in ways that go beyond traditional dichotomies between
sensation seeking and self-control (Chick & Reyna, 2012; Reyna et al., 2011)?
With these definitions and distinctions in mind, we can begin to assemble the
neural building blocks of risky decision making.

Neural Substrates of Reward and Valuation

The neural substrates of reward have been well characterized, building

on carefully controlled research with animals (Platt & Huettel, 2008). The
reward circuit of the brain consists of the midbrain dopamine areas (the ven-
tral tegmental area and substantia nigra) and the basal ganglia structures to
which they project (the ventral striatum, where the nucleus accumbens is
located, and the dorsal striatum). Axons from the midbrain areas also project
broadly to the prefrontal cortex, but particularly to the ventromedial pre­
frontal cortex (vmPFC; Galvan, 2012). Dopaminergic activity in these areas
has been linked to current and anticipated rewards (e.g., Glimcher, Camerer,
Fehr, & Poldrack, 2009).

16       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 16 11/15/13 1:42 PM

 Rewards are often appetitive in studies of animals (e.g., juice or food),
but studies of humans have shown generalization of value signals to include
monetary, social, and other rewards (Delgado, Nystrom, Fissell, Noll, & Fiez,
2000). The idea that there is a common currency of rewards in the brain
is consistent with economic notions of utility in which the subjective val-
ues of different types of rewards all project onto a dimension of reward value
(Montague & Berns, 2002; Smith et al., 2010). (We initially focus on rewards,
such as appetitive stimuli, but discuss aversive stimuli below.)
To test the common-currency hypothesis, D. J. Levy and Glimcher
(2011) mapped neural circuits for reward valuation by scanning subjects who
made choices for money, food, and water (see also Chib, Rangel, Shimojo,
& O’Doherty, 2009; FitzGerald, Seymour, & Dolan, 2009; Kim et al., 2010).
For example, a subject might choose between a sure win of five M&Ms ver-
sus a risky option offering a 38% chance of winning 20 M&Ms (and a 62%
chance of winning nothing). Subjects also received trade-off trials in which
they chose between a sure win of a small amount of money ($0.50) versus
some probability of either winning a fixed amount of food or water (or getting
nothing).
D. J. Levy and Glimcher (2011) found that risk preferences across
reward types were correlated: The level of risk aversion when choosing among
monetary options predicted risk aversion for food and water. In other words,
a subject who was more risk averse for money was generally also more risk
averse for food and water (although substantial variation among individu-
als was observed). The common areas of neural activation that varied with
valuation across domains were the vmPFC and striatum. These results are
reminiscent of those of other studies implicating the vmPFC and striatum in
representing values in risky choice tasks (Daw, O’Doherty, Dayan, Seymour,
& Dolan, 2006; De Martino, Kumaran, Holt, & Dolan, 2009; Huettel, Stowe,
Gordon, Warner, & Platt, 2006; Kable & Glimcher, 2007; Knutson, Fong,
Bennett, Adams, & Hommer, 2003; I. Levy, Snell, Nelson, Rustichini, &
Glimcher, 2010).
Interestingly, vmPFC is close to, or part of, the default mode network,
which is more active at rest than during task performance (Raichle & Snyder,
2007). This suggests that task-related activations in vmPFC might correspond
to different degrees of deactivation, relative to rest (Rushworth, Noonan,
Boorman, Walton, & Behrens, 2011). However, traditional default mode acti-
vation is anterior to the posterior region of the vmPFC most often linked to
common-currency activation.
In addition to common areas of activation in vmPFC, D. J. Levy and
Glimcher (2011) found that distinct neural networks represented monetary
and food rewards: the dorsal hypothalamic region responded mainly to the
reward value of food, whereas the posterior cingulate cortex responded mainly

reward, representation, and impulsivity      17

13490-02_PT1_Ch01-3rdPgs.indd 17 11/15/13 1:42 PM

 to the value of money. Within the vmPFC itself, there were overlapping but
also distinct areas of activation for money and food (see also Clithero, Carter,
& Huettel, 2009, for similar conclusions from pattern classification of brain
activation). However, only the vmPFC represented the value of money and
food on what appeared to be a common scale, as predicted by expected utility
and neurobiological approaches (Glimcher, Dorris, & Bayer, 2005; Glimcher,
2011; von Neumann & Morgenstern, 1944). That is, the relative levels of
activity in vmPFC to food and money rewards reflected the relative values
of food and money rewards to that individual and predicted trade-off choices
between these different types of rewards.
In contrast to this common-currency hypothesis, Weber, Blais, and
Betz (2002) and others have argued that an individual’s risk taking in one
domain, such as finances, need not be reflected in her risk taking for health
or other outcomes. Weber et al.’s Domain-Specific Risk-Taking (DOSPERT)
scale assesses risk-taking separately for decisions about monetary gambling
and investment, ethical, health/safety, social, and recreational domains (for
an updated scale, see Blais & Weber, 2006). As in classic expected-utility
approaches, they distinguish between perceptions of risks and benefits (i.e.,
rewards), and relate those to risk taking.
More specifically, the updated DOSPERT’s items include “Having an
affair with a married man/woman” (Ethical), “Investing 10% of your annual
income in a new business venture” (Financial), “Engaging in unprotected sex”
(Health/Safety), “Disagreeing with an authority figure on a major issue” (Social),
and “Taking a weekend sky-diving class” (Recreational). Weber et al. (2002)
and other studies show that risk attitudes are not the same across these domains
within individuals (which seems to contradict D. J. Levy & Glimcher’s, 2011,
results, but see below). Content varies across items, which makes them rich
descriptors of real-life behavior, but they do not isolate risk preference per se. In
other words, each item taps factors unrelated to risk preference, such as moral
compunctions or athletic interests, as well as risk preference.
Despite finding domain differences, Weber et al. (2002) reported that
sensation seeking (and its various subscales) correlated significantly with risky
behavior in all five domains of the DOSPERT scale, with the highest correla-
tion between the thrill-and-adventure-seeking subscale and recreational risk
taking (r = .56), echoing decades of similar results in which sensation seek-
ing predicts a broad spectrum of risky behaviors (e.g., Arnett, 1990a, 1990b;
Hoyle, Fejfar, & Miller, 2000; Zuckerman, 1994). In addition, Weber et al.
found that scores on each domain’s risk attitudes scale significantly predicted
frequency of risky behaviors for the other domains (with the exceptions of the
social subscale and social risk taking). Thus, D. J. Levy and Glimcher (2011)
used a common procedure to elicit risk attitudes across reward domains, rather
than comparing apples to oranges (i.e., noncommensurate financial vs. health

18       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 18 11/15/13 1:42 PM

 risk preferences), but Weber et al.’s results nevertheless confirm that prefer-
ences across reward types that reflect realistic decisions are correlated within
individuals. That is, a person who is more risk averse in a given reward type is
likely be more risk averse in another reward type. A recent meta-analysis using
data from 13 different functional magnetic resonance imaging (fMRI) stud-
ies corroborated the conclusion that this common reward representation is
located in a subregion of the vmPFC and adjacent medial orbitofrontal cortex
(mOFC; D. J. Levy & Glimcher, 2012).

Neural Substrates Differentiating Gains,

Losses, and Probability

Consistent with research that we have reviewed, the vmPFC/mOFC

blood oxygen level dependent (BOLD) signal has been shown to be corre-
lated with the reward value of choices in multiple studies (e.g., Rushworth
et al., 2011). For example, Plassmann, O’Doherty, and Rangel (2007) presented
subjects with a series of pictures of food items and asked them how much they
were willing to pay for each item. They used a Becker–DeGroot–Marschak
method to ensure that subjects had an incentive to provide their true valu-
ation; if their valuation exceeded a randomly drawn number, they forfeited
money and accepted the food instead. The vmPFC/mOFC signal increased
with participants’ valuation of the food.
The balance of evidence suggests that the vmPFC/mOFC signal responds
to losses as well as gains, decreasing proportionately to the magnitude of antic-
ipated losses or negative outcomes (Rushworth et al., 2011; Tom et al., 2007).
For example, Plassman, O’Doherty, and Rangel (2010) showed that willing-
ness to pay to avoid eating an unpleasant food inversely correlated with signal
in this region. The signal also decreases with other dimensions of choice
that lower the overall value of an option, such as lower probability of gains
or greater delay to receive gains (Kable & Glimcher, 2007; Peters & Buchel,
2009; Prévost, Pessiglione, Me´te´reau, Cléry-Melin, & Dreher, 2010). Hunt
et al. (2012) advanced a plausible model of neuronal activity in this region
in which signals initially correspond to the sum of the values of potential
choices, but, later in the trial, reflect the value difference between choices.
Using magnetoencephalography, Hunt et al. found that such a sequence of
signals occurred in vmPFC/mOFC and also superior parietal cortex near
the intraparietal sulcus. Thus, conflicting results supporting both the pres-
ence of overall value and value-difference signals in vmPFC/mOFC can
be reconciled by assuming that these outputs occur at different points in
time (Rushworth, Kolling, Sallet, & Mars, 2012; Wunderlich, Rangel, &
O’Doherty, 2010).

reward, representation, and impulsivity      19

13490-02_PT1_Ch01-3rdPgs.indd 19 11/15/13 1:42 PM

 Therefore, across studies, neural valuation subsumes both the reward
value of outcomes and overall expected value, which includes probability
(or risk) and outcomes. However, little consensus exists about the neural
substrates of probability or risk (Huettel, 2010). Indeed, information about
probability or risk has no meaning without information about outcomes; the
probability must refer to the probability of some outcome to elicit prefer-
ences or choices. (This psychological reality about probability does not imply
that probability cannot be varied orthogonally to outcomes; see d’Acremont,
Fornari, & Bossaerts, 2013; Reyna et al., 2011).
D’Acremont et al. (2013) varied probability independently of outcome
value using a probability learning task (subjects learned the probabilities of
stimuli representing various payoffs through repetitive sampling). They found
that activation in the medial prefrontal cortex and parietal cortex (angular
gyri) increased linearly with the probability of the currently observed stimu-
lus. Connectivity analyses during rest and task revealed that these regions
were part of the default mode network. (Outcome values were encoded out-
side of the default network, in the striatum, when net rewards were realized
at the end of the decision phase.)
Thus, contrary to the usual characterization, the default mode net-
work was active during a task requiring attention to external stimuli (see
also Spreng, 2012). Neuroimaging studies have identified these areas of the
network, parietal cortex and medial prefrontal cortex, as involved in success-
ful recognition (Cabeza, Ciaramelli, Olson, & Moscovitch, 2008; Wagner,
Shannon, Kahn, & Buckner, 2005). D’Acremont et al. (2013) argue that a
major function of the default mode network is to represent memory strength
for traces of past events, consistent with contemporary memory-based theo-
ries of decision making (Reyna & Brainerd, 2011; Weber & Johnson, 2009).
As we discuss in greater detail below, activation in dorsal anterior cingulate
cortex (ACC) was related to choice uncertainty (i.e., uncertainty in making a
choice, or decisional conflict). When choices were made but outcomes were
unknown (the risky phase of the task), the right anterior insula and bilateral
caudate were activated; the insula signaled outcome uncertainty.
To evaluate the evidence about neural substrates of probability (or risk),
Mohr, Biele, and Heerkeren (2010) conducted a quantitative meta-analysis of
fMRI experiments on risk processing. Mohr et al. examined how risk process-
ing is influenced by emotions, how it differs between choice and nonchoice
situations, and how it changes when losses are possible. Over a range of para-
digms, risk was consistently associated with activation in the anterior insula,
a brain region known to process aversive emotions, such as anxiety. The ante-
rior insula was predominantly active in the presence of potential losses. The
authors interpreted these results as evidence that risk processing is influenced
by emotions and as indicating that potential losses modulate risk processing.

20       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 20 11/15/13 1:42 PM

 Reflecting the nature of the literature, Mohr et al.’s (2010) conclu-
sion spans the concepts of risk and loss aversion, which, as we have already
discussed, are distinct (e.g., Kahneman & Tversky, 2000). However, these
results raise the interesting question of whether risk and loss aversion share an
underlying commonality—a subjective feeling of aversion—despite the fact
that this feeling is elicited by different causes (i.e., uncertainty vs. loss, both
of which many people find distasteful; Reyna & Brainerd, 2011; Willemsen
et al., 2011). This subjective feeling of aversion, in turn, is one avenue through
which risk taking can be inhibited (Clark et al., 2008).
An interpretation of Mohr et al.’s (2010) meta-analysis, then, one
that is consistent with studies cited by Wood and Bechara in this volume, is
that the insula signals the phenomenology of aversion, common to risk and
loss aversion. In other words, the insula provides an interoceptive signal of
subjective feeling, not only of aversion (or disgust) but also of cravings and
other homeostatic signals, as well as more abstract feelings such as admira-
tion, love, and indignation (Chapter 7, this volume). Physiological states
activated in the posterior insula seem to be re-represented in the anterior
insula, the latter mapping onto subjective feelings. Craig, Chen, Bandy,
and Reiman (2000), for example, found that activity in the posterior insula
was linearly related to the actual temperature of a stimulus (applied to the
hand), but activity in the anterior insula correlated with subjective ratings
of temperature. Thus, processing of the magnitudes of potential losses in
vmPFC/mOFC should be distinguished from subjective twinges associated
with feeling those losses in the insula, which may help people avoid risk
(Kuhnen & Knutson, 2005).
The function of the insula should also be distinguished from that of the
amygdala, although they are connected reciprocally (the anterior insula is
also connected to the ventral striatum, vmPFC, and ACC). As Wood and
Bechara (Chapter 7, this volume) explain, “Although the amygdala is respon-
sible for associative learning and emotions, the insula governs the conscious
feeling of those emotions” (p. 189). Some patients with amygdala lesions do
not show loss aversion, which is consistent with other research showing such
patients’ impaired processing of negative stimuli (De Martino, Camerer, &
Adolphs, 2010; Murray, 2007). However, other patients with amygdala lesions
have shown impairments for risky decisions involving gains rather than losses
(Weller, Levin, Shiv, & Bechara, 2007). Basten, Biele, Heekeren, and Fiebach
(2010) found that ventral striatum activated more in response to expected
rewards or benefits in a decision task, whereas amygdala activated more to
expected losses (and cost–benefit differences were correlated with an inferred
comparison region in the vmPFC). As suggested above, results for vmPFC
were consistent with computing expected reward by trading off amygdalar
“costs” and ventral striatal “benefits.”

reward, representation, and impulsivity      21

13490-02_PT1_Ch01-3rdPgs.indd 21 11/15/13 1:42 PM

 Although stimulating the amygdala produces fear, anxiety, and vigilance
(Phillips, Drevets, Rauch, & Lane, 2003), the amygdala activates more in
response to positive, negative, and unusual or interesting stimuli than in response
to neutral stimuli (Cunningham & Brosch, 2012). Therefore, the amygdala
may signal emotional salience, with riskiness and losses being among those
features of decision options that have functional significance (Cunningham,
Van Bavel, & Johnsen, 2008; De Martino et al., 2006; Roiser et al., 2009). De
Martino et al. (2010) speculated that an initial negative anticipatory response
is generated in the amygdala to potential losses, which influences the striatal
computation of the gamble’s net value, and consequently leads to loss aversion.
Thus, in this view, the amygdala is not the locus of memory or choice, and its
effect on loss aversion is indirect. Although patients with amygdala lesions
would be less likely to be loss averse, those with an intact ventral striatum and
vmPFC/mOFC—and especially insula—might be able to avoid disadvanta-
geous (in the sense of lower expected value) gambles involving losses (Weller
et al., 2007).
These areas of the brain are dynamically connected, as seen when people
regulate emotional affect. Studies of emotion regulation suggest that activa-
tion of the vmPFC and the amygdala are related, and coupling of the amygdala
and vmPFC is related to trait anxiety in both task-related and resting-state
studies (Burghy et al., 2012). For example, individuals high in trait anxi-
ety exhibit reduced amygdala-vmPFC resting-state connectivity (Kim, Gee,
Loucks, Davis, & Whalen, 2011), as though the lack of connectivity reflected
habitual inability to down-regulate anxiety (but see Poldrack, 2006, for cave-
ats regarding “reverse inference”).

Comparison and Conflict

As we have discussed, the evidence suggests that values of gains, losses,

and their probabilities are represented in the vmPFC/mOFC, informed by
the striatum and the rest of the reward circuit. The options also give rise
to subjective feelings and are infused with emotional significance, reflected
in the insula and amygdala. At some point in the decision process, the rep-
resentations of options are input to another comparison process that takes
place in dorsal ACC or adjacent (e.g., Hare, Schultz, Camerer, O’Doherty,
& Rangel, 2011).
In contrast to signals in the vmPFC/mOFC, the ACC/dmPFC BOLD
signal increases when the difference between the values of potential choices
decreases. For example, Pochon, Riis, Sanfey, Nystrom, and Cohen (2008)
had subjects choose between two attractive faces; activity in the ACC/dmPFC
was higher when faces were similar in attractiveness, provoking decisional

22       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 22 11/15/13 1:42 PM

 conflict. Similarly, De Martino et al. (2006) observed greater ACC/dmPFC
activation when subjects made choices that were inconsistent with their dom-
inant preference of framing effects. Increased activation in the dmPFC seems
to represent a conflict between the generally preferred gist-based response and
a compensatory analytical choice. In other words, the difficulty of a compari-
son between options varies with signal in the ACC/dmPFC as though this
area captures disparity detection (e.g., error monitoring), cognitive conflict,
or decisional conflict (e.g., Brown & Braver, 2008; Taren, Venkatraman, &
Huettel, 2011; Venkatraman et al., 2009; see Venkatraman & Huettel, 2012).
The output of this comparison process then determines activation in the motor
system, guiding which response should be made to satisfy the reward goal, and
culminating in the response (Cai & Padoa-Schioppa, 2012; Rushworth et al.,
2012). Hence, the vmPFC/mOFC can be thought of as “choosing” (or favor-
ing) a reward goal, whereas the ACC/dmPFC adjudicates potentially antago-
nistic actions or rules for decision making (Brown & Braver, 2008; Rushworth
et al., 2012; Venkatraman & Huettel, 2012).

Representation and Response Preference

While the vmPFC/mOFC signal increases with the difference in value

between possible choices, the BOLD signal in the parietal cortex and some
other motor association areas increases as the choice selection becomes
more difficult, as indexed by reaction time. The parietal signal, therefore,
often has characteristics that are the opposite of the vmPFC/mOFC signal.
Like the ACC/dmPFC, the size of the parietal signal (e.g., posterior parietal
cortex, the medial intraparietal sulcus) is negatively correlated with the dif-
ference in value between choices (Basten et al., 2010; Rushworth et al.,
2011). According to Basten et al. (2010), for example, a neural representa-
tion of the difference between rewards (ventral striatal benefit signal) and
losses (amygdalar cost signal) is evaluated in the vmPFC and then accumu-
lates in the parietal cortex (intraparietal sulcus) until a decision threshold
is reached.
Thus, one way to think of the accumulator is as an index of response pref-
erence; difficult choices are those for which the difference in value between
response options is small. The accumulator can operate within trials, as people
process information about risks and rewards, or costs and benefits, with longer
response latency and lower confidence for difficult choices. The accumula-
tor can also be thought of as operating across trials, as information becomes
acquired through experiencing outcomes of risky decisions. As examples of
learning from experience, one can have unprotected sex and either experience
pregnancy or not over the course of a year; one can leave a car unlocked each

reward, representation, and impulsivity      23

13490-02_PT1_Ch01-3rdPgs.indd 23 11/15/13 1:42 PM

 night for a month and find out each morning whether items in the car have
been stolen or not (e.g., Yechiam, Barron, & Erev, 2005). Alternatively, one
can read about the probability of pregnancy given unprotected sex over a year
on a contraceptive label or look up the rates at which cars are broken into in
a neighborhood, learning from description as opposed to experience (Reyna
& Adam, 2003; Reyna & Farley, 2006).
Many of the decision paradigms we have cited involve learning about
outcomes and probabilities through experience rather than direct description,
as in the Iowa Gambling Task (IGT). In the IGT, subjects select cards from
“good” and “bad” decks, discovering that some decks yield small rewards and
losses but net gains, whereas other decks yield large rewards but still larger
losses (e.g., Hochman, Yechiam, & Bechara, 2010). Therefore, subjects learn
to anticipate outcomes by virtue of experience, a process that relies heav-
ily on memory for outcomes and their probabilities (Stout et al., 2005).
Developmental differences in risk taking in the IGT, for example, virtually
disappear when children are provided with ongoing tallies of frequencies of
outcomes (Van Duijvenvoorde, Jansen, Bredman, & Huizenga, 2012; see also
Reyna & Brainerd, 1994). As we discuss, in addition to the medial temporal
lobe (MTL), encompassing the hippocampus and parahippocampal regions,
the parietal cortex plays an important role in memory representation and
retrieval (Cabeza et al., 2008).

Memory Representations

Memory representations are generally encoded simultaneously in two

formats: verbatim (e.g., the number of dollars to be won) and gist (e.g., the
categorical or ordinal qualitative essence of the amount to be won, such as
“some money” or “a lot of money”; Kühberger & Tanner, 2010; Reyna &
Brainerd, 1995, 2011). Verbatim representations support precise processing
(e.g., trading off the magnitudes of risk and reward, as in expected value or
utility), but gist representations support the fuzzy, impressionistic processing
of intuition (Reyna, 2013). Although neuroscience research has focused on
representing quantitative valuation of rewards or expected value (even when
options are nonnumerical, such as food; Hare et al., 2011), behavioral research
demonstrates that qualitative, fuzzy gist representations of decision options,
along with retrieval of associated broad mores (e.g., that saving lives, gaining
money, or helping family is good), often govern risk preferences (Reyna, 2012;
Reyna et al., 2011).
Specifically, consistent with neuroscience research, decision makers
appear to estimate expected value or expected utility (e.g., Glimcher, 2011;
Hare, Camerer, & Rangel, 2009), a conclusion that, at first blush, contradicts
the literature ruling out expected value or utility as viable descriptive theories

24       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 24 11/15/13 1:42 PM

 of behavior (e.g., Tversky & Kahneman, 1986). However, these views can be
reconciled by positing parallel systems that process both verbatim (symbolic,
but superficial) values of options and their qualitative gist. Qualitative gist,
in turn, is responsible for effects of meaning and context that can be said to
“bias” decisions, such as framing effects, used as evidence against expected
value and utility theories (Reyna & Brainerd, 2011).
Furthermore, these verbatim–gist parallel systems are required to account
for predicted double dissociations and nonmonotonic effects in the behav-
ioral literature, and they are supported by mathematical models tested for
goodness-of-fit with behavioral data (e.g., see Brainerd, Reyna, & Howe,
2009; Brainerd, Reyna, & Mojardin, 1999; Reyna, 2012; Reyna & Brainerd,
1995; Singer & Remillard, 2008). Evidence that people encode both verba-
tim (literal) and gist (simple meaning) representations comes from research
on memory, reasoning, and decision making, and has been studied across the
life span (e.g., for reviews, see Reyna, 1995, 2012; Reyna & Mills, 2007).
In particular, many experiments have compared verbatim-based true
memory to so-called false memory, the latter usually based on memory for
the gist of experience (Kim & Cabeza, 2007; Slotnick & Schacter, 2004).
While true recollection (verbatim memory is vivid or recollective) has been
associated with neural activity in the MTL (hippocampus and posterior para-
hippopcampal gyrus), medial prefrontal cortex (mPFC), lateral parietal cor-
tex, and posterior cingulate, vague familiarity has been associated with activity
in lateral PFC regions, the MTL (anterior parahippocampal gyrus and rhinal
cortex), and the superior parietal cortex (e.g., Daselaar, Fleck, & Cabeza, 2006;
Spaniol et al., 2009; Yonelinas, Otten, Shaw, & Rugg, 2005; for reviews, see
Diana, Yonelinas, & Ranganath, 2007; Eichenbaum, Yonelinas, & Ranganath,
2007; Eichenbaum, Sauvage, Fortin, Komorowski, & Lipton, 2012).
These memory studies provide clues about the neural basis of gist ver-
sus verbatim representations and processing that explain risky decision mak-
ing, as we presently discuss. However, many of the memory studies confound
phenomenology—vivid recollection versus vague familiarity—with verba-
tim or gist representation. Although verbatim representations (e.g., memory
for frequencies of outcomes of draws in the IGT) are vivid, gist representa-
tions can be either vague (experienced as global similarity or familiarity) or
vivid (experienced as “phantom recollection”; Brainerd, Payne, Wright, &
Reyna, 2003).
Using a recognition task with pictures, Dennis, Bowman, and Vandekar
(2012) separated recollection from familiarity by asking subjects to judge
recognition-test pictures as “remember,” “know,” or “new” (“remember”
responses were designated as recollection and “know” responses as familiar-
ity). Test items were previously presented pictures (true), semantically simi-
lar but not identical pictures (false or gist-consistent), or unrelated pictures.

reward, representation, and impulsivity      25

13490-02_PT1_Ch01-3rdPgs.indd 25 11/15/13 1:42 PM

 Directly comparing true with false recollection revealed that true recol-
lection uniquely involved hippocampus and early visual cortex, consistent
with verbatim memory supporting remember responses to presented pictures
(remember hits; Brainerd et al., 2003). Gist memory is assumed to be com-
mon to both true and false recollection and thus may be reflected in conjunc-
tion results for true and false recollection (e.g., precentral gyrus and superior
parietal cortex).
Dennis et al. (2012) also assessed whole brain functional connectivity
for an MTL region in the anterior parahippocampal gyrus that was activated
for true (true remember > true know) and for false recollection (false remem-
ber > false know). For true recollection (compared with false recollection), the
anterior parahippocampal gyrus showed greater functional connectivity with
inferior regions, including the bilateral hippocampus, the ACC, the orbito-
frontal cortex, and the occipital cortex. In contrast, false recollection (com-
pared with true recollection) showed greater functional connectivity between
the anterior parahippocampal gyrus and superior regions including the bilat-
eral pre- and post-central gyrus, the superior PFC, and the bilateral parietal
cortex. This inferior–superior dissociation may support a distinction between
bottom-up, lower-order verbatim processes in true recollection and top-down,
higher order gist and control processes in false recollection (Cabeza, 2008).

Memory Representations of Risk Preference in Decision Making

In decision making as in other information-processing tasks, people

encode verbatim and gist representations of options, and in experiential
paradigms (e.g., IGT), they encode both types of representations of the
probabilities and outcomes of choices. People extract multiple gist repre-
sentations of options but typically rely on the simplest (categorical) gist,
distinguishing between some quantity versus nothing, outputting a deci-
sion if the choice is not contradicted by parallel verbatim processing. For
example, the simplest gist of the example presented earlier is (a) winning
some money versus (b) either winning some money or winning nothing.
Because most people value money, some money is generally preferred to
nothing, favoring the sure option.
The same kind of categorical (some–none) gist applies to choices
involving losses: Thus, a choice between (c) losing $1,000 for sure ver-
sus (d) a .5 probability of losing $2,000 and .5 probability of losing nothing
boils down to (c) losing some money versus (d) either losing some money or
losing nothing. Because most people value losing nothing more than losing
money, they generally prefer the risky option. The shift from risk avoiding for
gains to risk seeking for losses is an example of a framing effect (Kahneman &
Tversky, 2000). Critical tests have been conducted showing that categorical

26       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 26 11/15/13 1:42 PM

 gist accounts for a variety of framing effects and ruling out expected utility and
prospect theory as explanations for these effects (e.g., Kühberger & Tanner,
2010; Reyna, 2012; Reyna & Brainerd, 1991, 1995).
To take one example, removing the mathematically redundant zero com-
plement of the gamble (.5 probability of winning/losing nothing = 0) removes
the categorical some–none contrast and eliminates framing effects, although
all of the causal elements of framing effects, according to utility and pros-
pect theory, remain. These and other empirical results, such as larger framing
effects when numbers are removed, experimentally manipulating attention to
the zero outcome to increase framing effects, and the growth of framing effects
from childhood to adulthood, all support a fuzzy-trace theory account of ver-
batim and gist processing in risky decision making (e.g., Kühberger & Tanner,
2010; Reyna & Brainerd, 1991, 1995; Reyna & Ellis, 1994; Reyna et al., 2011;
Reyna & Farley, 2006). Although framing effects themselves divulge the use
of categorical gist (per the explanation above), parallel verbatim processing
is revealed through people’s sensitivity to expected value: People are usually
indifferent between options when their expected values are equal—provided
that the zero complement is removed—but they favor the option with the
larger expected value when options are unequal (e.g., Reyna & Brainerd,
1995; Weller et al., 2007).
Venkatraman et al. (2009) explored the use of these alternative strate-
gies of verbatim trading off of expected value versus categorical some–none
gist processing by presenting subjects with a series of five-outcome gambles
containing gain and loss outcomes (probabilities are shown in parentheses),
such as $80 (.20), $40 (.25), $0 (.20), -$25 (.15), -$70 (.20). However, sub-
jects could improve the gambles, for example, by adding $15 to either the
$0 outcome (changing that outcome to $15) or to the -$70 (changing that
outcome to -$55). The choices subjects made to change gambles diagnosed
their information-processing strategies.
Altogether, Venkatraman et al. (2009) assessed three strategies: increas-
ing the magnitude of the highest gain (Gmax), decreasing the magnitude of
the worst loss (Lmin), or improving the probability of winning something by
adding money to the middle outcome (e.g., by eliminating the categorical
possibility of winning nothing; Pmax). The Gmax and Lmin choices repre-
sent a compensatory verbatim analytical strategy consistent with standard
models of risky choice, whereas Pmax represents a simplifying gist-based
strategy. Thus, the tasks used by Venkatraman et al. (2009), like those used by
Kühberger and Tanner (2010); Mills, Reyna, and Estrada (2008); and Reyna
(2012), make it possible to discriminate between gist- and verbatim-based
strategies in risky decision making.
Activation in the posterior parietal cortex and dorsolateral prefrontal
cortex (dlPFC) predicted gist-based, simplifying choices, whereas activation

reward, representation, and impulsivity      27

13490-02_PT1_Ch01-3rdPgs.indd 27 12/9/13 2:06 PM

 in the vmPFC and anterior insula predicted verbatim analytical, compen-
satory choices, maximizing gains (i.e., Gmax) and minimizing losses (i.e.,
loss aversion or Lmin), respectively (see also, Hedgcock, Denburg, Levin, &
Halfmann, 2012). Functional connectivity (psychophysiological interaction)
analyses showed positive correlations between the dmPFC and the dlPFC for
simplifying choices and between the dmPFC and insula for compensatory
choices. Although many scholars assume that compensatory strategies are
more adaptive, research has shown that gist-based simplifying strategies are
associated with greater development (from childhood to adulthood), greater
expertise in adulthood, and better health outcomes (e.g., Reyna et al., 2011;
Reyna & Farley, 2006; Reyna & Lloyd, 2006).
Consistent with our earlier discussion of the ACC/dmPFC, activation
in this area was greater when subjects made choices that conflicted with
their dominant strategy, such as when people who generally preferred the
gist-based, simplifying choice made a compensatory choice and vice versa.
Greater ACC/dmPFC activation occurs for higher decisional conflict even
when subjects do not have to respond overtly on a given trial and when deci-
sion and response phases are separated (Pochon et al., 2008; Venkatraman
et al., 2009). Further, Venkatraman et al. (2009) showed, using resting state
connectivity analyses, that the dmPFC was connected to the dlPFC, the lat-
ter associated with response selection, in an anterior to posterior organization
(e.g., anterior dmPFC to rostral dlPFC; posterior dmPFC to caudal dlPFC
and premotor cortex; Bunge, Hazeltine, Scanlon, Rosen, & Gabrieli, 2002;
Taren et al., 2011). Venkatraman and Huettel (2012) argued that the ante-
rior dmPFC regulates activity in the anterior dlPFC when control demands are
highly abstract (e.g., choosing among conflicting strategies), but the posterior
dmPFC regulates the posterior dlPFC and premotor cortices when control
demands merely choose between conflicting motor responses.

Impulsivity/Inhibition of Responses

The dlPFC, often in conjunction with the ACC, has long been associ-
ated with cognitive control, which can be manifested as response inhibition,
cognitive distraction (distancing), or reappraisal of the meaning of a stimulus
(Ochsner & Gross, 2008; Venkatraman & Huettel, 2012; see also Chapter 6,
this volume). The dlPFC modulates the value signal encoded in the vmPFC,
and dlPFC activity is correlated with successful self-control (e.g., in go/no-go
tasks; Casey et al., 2011; or when choosing between healthy and unhealthy
foods; Hare et al., 2009).
Several studies have reported a link between higher dlPFC activity and
lower risk taking. For example, Gianotti et al. (2009) found a negative cor-
relation between tonic activity in the dlPFC (i.e., cortical hypoactivity) as

28       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 28 11/15/13 1:42 PM

 measured using electroencephalogram at rest and risk-taking propensity in
a laboratory task. These results are consistent with findings from patients
with lesions in this area and healthy subjects with “virtual lesions” (created
using transmagnetic cranial stimulation, or TMS), who showed increased risk
taking relative to controls (Knoch et al., 2006). Conversely, Fecteau et al.
(2007) reduced risk taking in a laboratory task by increasing dlPFC activity
using transcranial direct current stimulation. Schonberg et al. (2012) inter-
preted their results showing increasing dlPFC activation as risk mounts in a
repeated sampling task as reflecting the increased engagement of self-control,
which prompts subjects to terminate the trial.
Finally, Hutcherson, Plassman, Gross, and Rangel (2012) instructed
subjects to down-regulate (“distance”) or up-regulate (“indulge”) their desire
for foods (or react naturally) while being imaged. During down-regulation,
activation decreased in the dlPFC but not in the vmPFC, and the relative
contribution of the two value signals to behavior shifted toward the dlPFC.
The opposite pattern was observed during up-regulation; activation increased
in the vmPFC but not the dlPFC, and the relative contribution to behavior
shifted toward the vmPFC. Although the direction of activation seems to con-
tradict other research on cognitive regulation of emotion, the authors inter-
pret the reduction in activation in the dlPFC as indicating value modulation
(see Hutcherson et al.’s, 2012, discussion), a finding consistent with recent
research on emotional value in vmPFC (Winecoff et al., 2013). However,
in both regulation conditions relative to “natural” responding, Hutcherson
et al. observed significant activation in the ventrolateral PFC and the poste-
rior parietal cortex, areas also found in other studies. Taken together, these
studies suggest that dlPFC (and other areas’) activity reflects cognitive con-
trol that reigns in risk taking, generally through inhibition or reevaluation of
emotional, rewarding, or prepotent responses.
Going beyond the influence of a single area or region such as the dlPFC,
Shannon et al. (2011) correlated resting state connectivity with a measure of
impulsivity for adolescents incarcerated in a high-security facility. The impul-
sivity measure included “need for stimulation” (e.g., “I enjoy gambling for large
stakes”). Shannon et al. showed that prediction of impulsivity peaked using
premotor (i.e., motor planning regions) functional connectivity (i.e., selecting
or adding other brain regions of interest degraded prediction of impulsivity). A
typical young adult sample displays positive correlations between the pre-
motor area and both the dorsal attention and executive networks, and negative
correlations with the default mode network. A similar pattern was observed in
less impulsive incarcerated youth as well as normal controls; premotor areas
were connected to networks involving attention and executive processes.
In contrast, premotor areas were positively correlated with the default net-
work among the impulsive incarcerated youth, parts of which (e.g., vmPFC),

reward, representation, and impulsivity      29

13490-02_PT1_Ch01-3rdPgs.indd 29 11/15/13 1:42 PM

 as we have discussed, are involved in task-related reward valuation and prob-
ability learning in normally developing adults. Premotor areas were nega-
tively correlated with attention and control networks. Hence, impulsive
incarcerated youth displayed the opposite pattern shown with normal adults.
Moreover, functional connectivity varied with age among another sample
of normal controls ranging in age from 7 to 31. As age increased, functional
connectivity with the premotor area migrated from the default network
(significantly decreasing with age) to the attention and control networks
(significantly increasing with age). Thus, the relationship of the networks to
impulsivity, though stronger in incarcerated youth, was recapitulated by age,
as though impulsive offenders were developmentally delayed.

Overview and Implications for Risky Decision Making

Our review of the literature on the neuroscience of risky decision making

has ranged from reward circuitry to response inhibition. We have reviewed
activations in different brain regions, but also the interconnections among
areas, for example, between the amygdala and the vmPFC (e.g., the input of
the amygdala to vmPFC valuation comparisons). The hypotheses offered
about function represent an effort to make sense of neuroscientific find-
ings, but they can only be tentative as the necessary process models remain
underspecified.
In particular, the simultaneous and countervailing interactions among
different brain processes challenge interpretation (e.g., Shannon et al., 2011,
which demonstrates network-level interactions). For example, if the dlPFC
downregulates the global valuation of a risky option, then activation of the
vmPFC might appear misleadingly small. By misleading, we mean that the
causal pathway for risky behavior differs for an individual who has mastered
self-control (but for whom self-regulation is imperfect) versus someone who
was never tempted to begin with (DeYoung, 2010). Understanding these
causal pathways is crucial for designing interventions and public policies to
reduce unhealthy risk taking.
For example, the levers for behavioral change include shaping the encod-
ing of the functionally significant features of decisions (e.g., consequential finan-
cial losses or major health threats; Reyna, 2008). Very little is understood
about “salience” and how the amygdala categorizes features of decisions,
such as their riskiness, as significant or not. The presence of risk, by itself, is
not a contraindication to choosing an option; driving in most cities repre-
sents a nonnegligible risk. (Phobias about driving, flying, or otherwise engag-
ing in everyday risks can severely impair adaptive decision making.) Thus,
despite the widespread faith in expected utility, decision trees, and other
rational approaches to decision making, adaptive risk taking is generally not

30       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 30 11/15/13 1:42 PM

 characterized by a focus on such details, but, rather, on the “big picture” of
what is important.
That is, adaptive decision making involves more than applying a rote
formula, as suggested by fuzzy trace theory. Understanding functional signifi-
cance (i.e., what the individual represents as important about options that
involve risk) is a deep and highly contextual judgment. The concept of gist,
which dates back to research on summarizing narratives, sheds some light
on how people distill the substance of decisions (e.g., Clark & Clark, 1977).
Although the representation of the gist of options—their meaning—seems
to shape risk preferences, most neuroscience research employs tasks that fail
to capture or even assess that meaning. Naturally, researchers who do not
measure gist processing will not find it in the brain, omitting a major mode of
real-world decision making.
The concept of self-control is also fraught with ambiguity in the cur-
rent literature, but our review suggests some basic distinctions that may
be helpful. As Casey et al. (2011) emphasized, responsiveness to enticing
rewards represents a distinct vulnerability for impulsive individuals, and,
presumably, for those who take risks (here, impulsivity is defined narrowly
as failure to inhibit responses, but see Zuckerman, 1994). Response inhibi-
tion in the sense of restraint seems to be discriminable from reward sensi-
tivity (i.e., temptation by valuation). The inability to engage in cognitive
distraction or reappraisal of meaning because of low intelligence or poor
coping skills is yet another source of vulnerability. The failure to learn the
probabilities of events from experience or to detect conflict among deci-
sion strategies, and the absence of internal feeling cues, are additional risk
factors for maladaptive decision making. How each of these factors oper-
ates to promote risk taking (alone or in combination)—and their neural
circuitry—is poorly understood.
One impediment to progress in this area is the mistaken belief that
laboratory risk-taking tasks that unconfound causal factors do not predict
real-life risk taking. Many studies have refuted that misconception, and yet
it persists. However, this does not imply that a narrow focus on college stu-
dents of a certain age with little variation in culture or individual differences
is adequate; representative samples of subjects are also needed. Rather than
measure large numbers of people using poorly conceived tasks (and powerful
statistics), however, progress in the neuroscience of risky decision making
requires a greater focus on tasks motivated by process-oriented theory.
To some decision neuroscience researchers, theory harkens back to tra-
ditional expected utility models and simple Skinnerian behaviorism. Is risky
decision making merely a matter of computing gains (rewards), losses, and
probabilities (with perhaps a touch of Spence’s notion of behavioral inhi-
bition)? Certainly, at one level, such simple concepts successfully predict

reward, representation, and impulsivity      31

13490-02_PT1_Ch01-3rdPgs.indd 31 11/15/13 1:42 PM

 approach and avoidance behavior, including some aspects of risky decision
making. However, behaving logically and trading off risk and reward do not
seem to be sufficient for adaptive decision making. Children (by about 6 years
of age), adolescents, and people with mild autism are less prone to classic
judgment-and-decision-making biases, compared with neurodevelopmen-
tally typical adults, but they nevertheless have deficits in real-world risky
decision making (Reyna & Brainerd, 2011). Acknowledging “risk as feelings”
is also not sufficient to predict or improve risky decision making to date (for a
discussion of the shortcomings of this approach, see Sunstein, 2008). Instead,
an integrative approach is needed that emphasizes prediction and hypothesis
testing, an approach that incorporates motivational, emotional, and cogni-
tive elements but goes beyond them in designing new tasks that capture the
richness and relevance of risky decision making.

References

American Psychiatric Association. (2000). Diagnostic and Statistical Manual of Mental

Disorders (4th ed., text rev.). Washington, DC: Author.
Arnett, J. (1990a). Contraceptive use, sensation seeking, and adolescent egocen-
trism. Journal of Youth and Adolescence, 19, 171–180. doi:10.1007/BF01538720
Arnett, J. (1990b). Drunk driving, sensation seeking and egocentrism among
adolescents. Personality and Individual Differences, 11, 541–546. doi:10.1016/
0191-8869(90)90035-P
Arnett, J. (1992). Reckless behavior in adolescence: A developmental perspective.
Developmental Review, 12, 339–373. doi:10.1016/0273-2297(92)90013-R
Basten, U., Biele, G., Heekeren, H. R., & Fiebach, C. J. (2010). How the brain
integrates costs and benefits during decision making. PNAS Proceedings of the
National Academy of Sciences of the United States of America, 107, 21767–21772.
doi:10.1073/pnas.0908104107
Benartzi, S., & Thaler, R. H. (1995). Myopic loss aversion and the equity premium
puzzle. The Quarterly Journal of Economics, 110(1), 73–92. doi:10.2307/2118511
Blais, A., & Weber, E. U. (2006). A domain-specific risk-taking (DOSPERT) scale
for adult populations. Judgment and Decision Making, 1, 33–47. Retrieved from
http://journal.sjdm.org/06005/jdm06005.htm
Bloom, D. (2010). Programs and policies to assist high school dropouts in the transi-
tion to adulthood. The Future of Children, 20, 89–108. doi:10.1353/foc.0.0039
Brainerd, C. J., Payne, D. G., Wright, R., & Reyna, V. F. (2003). Phantom recall. Jour­
nal of Memory and Language, 48, 445–467. doi:10.1016/S0749-596X(02)00501-6
Brainerd, C. J., Reyna, V. F., & Howe, M. L. (2009). Trichotomous processes in early
memory development, aging, and neurocognitive impairment: A unified theory.
Psychological Review, 116, 783–832. doi:10.1037/a0016963

32       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 32 11/15/13 1:42 PM

 Brainerd, C. J., Reyna, V. F., & Mojardin, A. H. (1999). Conjoint recognition.
Psychological Review, 106, 160–179. doi:10.1037/0033-295X.106.1.160
Brown, J. W., & Braver, T. S. (2008). A computational model of risk, conflict, and
individual difference effects in the anterior cingulated cortex. Brain Research,
1202, 99–108. doi:10.1016/j.brainres.2007.06.080
Bunge, S. A., Hazeltine, E., Scanlon, M. D., Rosen, A. C., & Gabrieli, J. D. (2002).
Dissociable contributions of prefrontal and parietal cortices to response selec-
tion. NeuroImage, 17, 1562–1571. doi:10.1006/nimg.2002.1252
Burghy, C. A., Stodola, D. E., Ruttle, P. L., Molloy, E. K., Armstrong, J. M., Oler,
J. A., . . . Birn, R. M. (2012). Developmental pathways to amygdala-prefrontal
function and internalizing symptoms in adolescence. Nature Neuroscience, 15,
1736–1741. doi:10.1038/nn.3257
Cabeza, R. (2008). Role of parietal regions in episodic memory retrieval: The dual
attentional processes hypothesis. Neuropsychologia, 46, 1813–1827. doi:10.1016/j.
neuropsychologia.2008.03.019
Cabeza, R., Ciaramelli, E., Olson, I., & Moscovitch, M. (2008). The parietal cortex
and episodic memory: An attentional account. Nature Reviews Neuroscience, 9,
613–625. doi:10.1038/nrn2459
Cai, X., & Padoa-Schioppa, C. (2012). Neuronal encoding of subjective value in
dorsal and ventral anterior cingulate cortex. The Journal of Neuroscience, 32,
3791–3808. doi:10.1523/JNEUROSCI.3864-11.2012
Casey, B. J., Somerville, L. H., Gotlib, I. H., Ayduk, O., Franklin, N. T., Askren.
M. K. . . . Shoda, Y. (2011). Behavioral and neural correlates of delay of gratifica-
tion 40 years later. Proceedings of the National Academy of Sciences of the United
States of America, 108, 14998–15003. doi:10.1073/pnas.1108561108
Centers for Disease Control and Prevention. (2012). Web-Based Injury Statistics Query
and Reporting System (WISQARS). National Center for Injury Prevention and
Control, Centers for Disease Control and Prevention (producer). Retrieved
from http://www.cdc.gov/injury/wisqars/fatal.html
Chib, V. S., Rangel, A., Shimojo, S., & O’Doherty, J. P. (2009). Evidence for a
common representation of decision values for dissimilar goods in human ven-
tromedial prefrontal cortex. The Journal of Neuroscience, 29, 12315–12320.
doi:10.1523/JNEUROSCI.2575-09.2009
Chick, C. F., & Reyna, V. F. (2012). A fuzzy trace theory of adolescent risk taking:
Beyond self-control and sensation seeking. In V. F. Reyna, S. Chapman, M.
Dougherty, & J. Confrey (Eds.), The adolescent brain: Learning, reasoning, and deci­
sion making (pp. 379–428). Washington, DC: American Psychological Associa-
tion. doi:10.1037/13493-013
Clark, H. H., & Clark, E. V. (1977). Psychology and language: An introduction to
psycho­linguistics. New York, NY: Harcourt Brace Jovanovich.
Clark, L., Bechara, A., Damasio, H., Aitken, M. R. F., Sahakian, B. J., & Robbins,
T. W. (2008). Differential effects of insular and ventromedial prefrontal cortex

reward, representation, and impulsivity      33

13490-02_PT1_Ch01-3rdPgs.indd 33 11/15/13 1:42 PM

 lesions on risky decision-making. Brain: A Journal of Neurology, 131, 1311–
1322. doi:10.1093/brain/awn066
Clithero, J. A., Carter, R. M., & Huettel, S. A. (2009). Local pattern classification
differentiates processes of economic valuation. NeuroImage, 45, 1329–1338.
doi:10.1016/j.neuroimage.2008.12.074
Craig, A. D., Chen, K., Bandy, D., & Reiman, E. (2000). Thermosensory activation
of insular cortex. Nature Neuroscience, 3, 184–190. doi:10.1038/72131
Cunningham, W. A., & Brosch, T. (2012). Motivational salience: Amygdala tuning
from traits, needs, values, and goals. Current Directions in Psychological Science,
21, 54–59. doi:10.1177/0963721411430832
Cunningham, W. A., Van Bavel, J. J., & Johnsen, I. R. (2008). Affective flexibility:
Evaluative processing goals shape amygdala activity. Psychological Science, 19,
152–160. doi:10.1111/j.1467-9280.2008.02061.x
d’Acremont, M., Fornari, E., & Bossaerts, P. (2013). Activity in inferior parietal and
medial prefrontal cortex signals the accumulation of evidence in a probability
learning task. PLoS Computational Biology, 9, e1002895. doi:10.1371/journal.
pcbi.1002895
Daselaar, S. M., Fleck, M. S., & Cabeza, R. E. (2006). Triple dissociation in the
medial temporal lobes: Recollection, familiarity, and novelty. Journal of Neuro­
physiology, 96, 1902–1911. doi:10.1152/jn.01029.2005
Daw, N. D., O’Doherty, J. P., Dayan, P., Seymour, B., & Dolan, R. J. (2006). Cor-
tical substrates for exploratory decisions in humans. Nature, 441, 876–879.
doi:10.1038/nature04766
Delgado, M. R., Nystrom, L. E., Fissell, C., Noll, D. C., & Fiez, J. A. (2000). Tracking
the hemodynamic responses to reward and punishment in the striatum. Journal
of Neurophysiology, 84, 3072–3077.
Dennis, N. A., Bowman, C. R., & Vandekar, S. N. (2012). True and phantom
recollection: An fMRI investigation of similar and distinct neural correlates
and connectivity. NeuroImage, 59, 2982–2993. doi:10.1016/j.neuroimage.
2011.09.079
De Martino, B., Camerer, C. F., & Adolphs, R. (2010). Amygdala damage eliminates
monetary loss aversion. Proceedings of the National Academy of Sciences of the
United States of America, 107, 3788–3792. doi:10.1073/pnas.0910230107
De Martino, B., Kumaran, D., Holt, B., & Dolan, R. J. (2009). The neurobiology
of reference-dependent value computation. The Journal of Neuroscience, 29,
3833–3842. doi:10.1523/JNEUROSCI.4832-08.2009
De Martino, B., Kumaran, D., Seymour, B., & Dolan, R. J. (2006). Frames, biases,
and rational decision-making in the human brain. Science, 313, 684–687.
doi:10.1126/science.1128356
DeYoung, C. G. (2010). Impulsivity as a personality trait. In K. D. Vohs & R. F.
Baumeister (Eds.), Handbook of self-regulation: Research, theory, and applications
(2nd ed., pp. 485–502). New York, NY: Guilford Press.

34       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 34 11/15/13 1:42 PM

 Diana, R. A., Yonelinas, A. P., & Ranganath, C. (2007). Imaging recollection and
familiarity in the medial temporal lobe: A three-component model. Trends in
Cognitive Sciences, 11, 379–386. doi:10.1016/j.tics.2007.08.001
Eichenbaum, H., Sauvage, M., Fortin, N., Komorowski, R., & Lipton, P. (2012).
Towards a functional organization of episodic memory in the medial tempo-
ral lobe. Neuroscience and Biobehavioral Reviews, 36, 1597–1608. doi:10.1016/
j.neubiorev.2011.07.006
Eichenbaum, H., Yonelinas, A. P., & Ranganath, C. (2007). The medial temporal
lobe and recognition memory. Annual Review of Neuroscience, 30, 123–152.
doi:10.1146/annurev.neuro.30.051606.094328
Fecteau, S., Pascual-Leone, A., Zald, D. H., Liguori, P., Theoret, H., Boggio, P. S., &
Fregni, F. (2007). Activation of prefrontal cortex by transcranial direct current
stimulation reduces appetite for risk during ambiguous decision making. The Jour­
nal of Neuroscience, 27, 6212–6218. doi:10.1523/JNEUROSCI.0314-07.2007
FitzGerald, T. H., Seymour, B., & Dolan, R. J. (2009). The role of human orbito-
frontal cortex in value comparison for incommensurable objects. The Journal of
Neuroscience, 29, 8388–8395. doi:10.1523/JNEUROSCI.0717-09.2009
Fox, C. R., & Tannenbaum, D. (2011). The elusive search for stable risk preferences.
Frontiers in Psychology, 2(298). doi:10.3389/fpsyg.2011.00298
Galvan, A. (2012). Risky behavior in adolescents: The role of the developing
brain. In V. F. Reyna, S. B. Chapman, M. R. Dougherty, & J. Confrey (Eds.),
The adolescent brain: Learning, reasoning, and decision making (pp. 267–289).
Washington, DC: American Psychological Association.
Galvan, A., Hare, T., Voss, H., Glover, G., & Casey, B. J. (2007). Risk-taking
and the adolescent brain: Who is at risk? Developmental Science, 10, F8–F14.
doi:10.1111/j.1467-7687.2006.00579.x
Galvan, A., Hare, T. A., Parra, C. E., Penn, J., Voss, H., Glover, G., & Casey, B. J.
(2006). Earlier development of the accumbens relative to orbitofrontal cortex
might underlie risk-taking behavior in adolescents. The Journal of Neuroscience,
26, 6885–6892. doi:10.1523/JNEUROSCI.1062-06.2006
Gianotti, L. R., Knoch, D., Faber, P. L., Lehmann, D., Pascual-Marqui, R. D., Diezi,
C., . . . Fehr, E. (2009). Tonic activity level in the right prefrontal cortex pre-
dicts individuals’ risk taking. Psychological Science, 20(1), 33–38. doi:10.1111/
j.1467-9280.2008.02260.x
Glimcher, P. W. (2011). Understanding dopamine and reinforcement learning: The
dopamine reward prediction error hypothesis. Proceedings of the National Acad­
emy of Sciences of the United States of America, 108, 15647–15654. doi:10.1073/
pnas.1014269108
Glimcher, P. W., Camerer, C. F., Fehr, E., & Poldrack, R. A. (Eds.). (2009). Neuro­
economics: Decision making and the brain. New York, NY: Academic Press.
Glimcher, P. W., Dorris, M. C., & Bayer, H. M. (2005). Physiological utility theory
and the neuroeconomics of choice. Games and Economic Behavior, 52, 213–256.
doi:10.1016/j.geb.2004.06.011

reward, representation, and impulsivity      35

13490-02_PT1_Ch01-3rdPgs.indd 35 11/15/13 1:42 PM

 Hare, T. A., Camerer, C. F., & Rangel, A. (2009). Self-control in decision-making
involves modulation of the vmPFC valuation system. Science, 324, 646–648.
doi:10.1126/science.1168450
Hare, T. A., Schultz, W., Camerer, C. F., O’Doherty, J. P., & Rangel, A. (2011).
Transformation of stimulus value signals into motor commands during simple
choice. Proceedings of the National Academy of Sciences of the United States of
America, 108, 18120–18125. doi:10.1073/pnas.1109322108
Hedgcock, W., Denburg, N., Levin, I. P., & Halfmann, K. (November, 2012). Why
older adults are impaired on some decision making tasks but not on others–Behavioral
and neuroimaging evidence. Paper presented at the Annual Meeting of the
Society for Judgment and Decision Making, Minneapolis, MN.
Hochman, G., Yechiam, E., & Bechara, A. (2010). Recency gets larger as lesions
move from anterior to posterior locations within the ventromedial prefrontal
cortex. Behavioural Brain Research, 213, 27–34. doi:10.1016/j.bbr.2010.04.023
Hoyle, R. H., Fejfar, M. C., & Miller, J. D. (2000). Personality and sexual risk-taking:
A quantitative review. Journal of Personality, 68, 1203–1231. doi:10.1111/1467-
6494.00132
Huettel, S. A. (2006). Behavioral, but not reward, risk modulates activation of pre-
frontal, parietal, and insular cortices. Cognitive, Affective & Behavioral Neuro­
science, 6, 141–151. doi:10.3758/CABN.6.2.141
Huettel, S. A. (2010). Ten challenges for decision neuroscience. Frontiers in Decision
Neuroscience, 4(171). doi:10.3389/fnins.2010.00171
Huettel, S. A., Song, A. W., & McCarthy, G. W. (2009). Functional magnetic reso­
nance imaging (2nd ed.). Sunderland, MA: Sinauer Associates Inc.
Huettel, S. A., Stowe, C. J., Gordon, E. M., Warner, B. T., & Platt, M. L. (2006).
Neural signatures of economic preferences for risk and ambiguity. Neuron, 49,
765–775. doi:10.1016/j.neuron.2006.01.024
Hunt, L. T., Kolling, N., Soltani, A., Woolrich, M. W., Rushworth, M. F., & Behrens,
T. E. (2012). Mechanisms underlying cortical activity during value-guided choice.
Nature Neuroscience, 15, 470–476. doi:10.1038/nn.3017
Hutcherson, C. A., Plassmann, H., Gross, J. J., & Rangel, A. (2012). Cognitive reg-
ulation during decision-making shifts behavioral control between ventrome-
dial and dorsolateral prefrontal value systems. Journal of Neuroscience, 32(39),
13543–13554.
Jessor, R. (1991). Risk behavior in adolescence: A psychosocial framework for under-
standing and action. Journal of Adolescent Health, 12, 597–605. doi:10.1016/
1054-139X(91)90007-K
Kable, J. W., & Glimcher, P. W. (2007). The neural correlates of subjective value
during intertemporal choice. Nature Neuroscience, 10, 1625–1633. doi:10.1038/
nn2007
Kahneman, D., & Tversky, A. (Eds.). (2000). Choices, values, and frames. New York,
NY: Cambridge University Press.

36       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 36 11/15/13 1:42 PM

 Kim, H., & Cabeza, R. (2007). Trusting our memories: Dissociating the neural cor-
relates of confidence in veridical and illusory memories. The Journal of Neuro­
science, 27, 12190–12197. doi:10.1523/JNEUROSCI.3408-07.2007
Kim, M. J., Gee, D. G., Loucks, R. A., Davis, F. C., & Whalen, P. J. (2011). Anxiety
dissociates dorsal and ventral medial prefrontal cortex functional connectivity
with the amygdala at rest. Cerebral Cortex, 21, 1667–1673. doi:10.1093/cercor/
bhq237
Kim, M. J., Loucks, R. A., Neta, M., Davis, F. C., Oler, J. A., Mazzulla, E. C., & Wha-
len, P. J. (2010). Behind the mask: The influence of mask-type on amygdala
response to fearful faces. Social Cognitive and Affective Neuroscience, 5, 363–368.
doi:10.1093/scan/nsq014
Knoch, D., Gianotti, L. R., Pascual-Leone, A., Treyer, V., Regard, M., Hohmann,
M., & Brugger, P. (2006). Disruption of right prefrontal cortex by low-frequency
repetitive transcranial magnetic stimulation induces risk-taking behavior. The
Journal of Neuroscience, 26, 6469–6472. doi:10.1523/JNEUROSCI.0804-06.2006
Knutson, B., Fong, G. W., Bennett, S. M., Adams, C. M., & Hommer, D. (2003).
A region of mesial prefrontal cortex tracks monetarily rewarding outcomes:
Characterization with rapid event-related fMRI. NeuroImage, 18, 263–272.
doi:10.1016/S1053-8119(02)00057-5
Kühberger, A., & Tanner, C. (2010). Risky choice framing: Task versions and a com-
parison of prospect theory and fuzzy-trace theory. Journal of Behavioral Decision
Making, 23, 314–329. doi:10.1002/bdm.656
Kuhnen, C. M., & Knutson, B. (2005). The neural basis of financial risk taking.
Neuron, 47, 763–770. doi:10.1016/j.neuron.2005.08.008
Lejuez, C. W., Aklin, W. M., Zvolensky, M. J., & Pedulla, C. M. (2003). Evaluation
of the balloon analogue risk task (BART) as a predictor of adolescent real-world
risk-taking behaviours. Journal of Adolescence, 26, 475–479. doi:10.1016/S0140-
1971(03)00036-8
Levy, D. J., & Glimcher, P. W. (2011). Comparing apples and oranges: Using reward-
specific and reward-general subjective value representation in the brain.
The Journal of Neuroscience, 31, 14693–14707. doi:10.1523/JNEUROSCI.
2218-11.2011
Levy, D. J., & Glimcher, P. W. (2012). The root of all value: A neural common cur-
rency for choice. Current Opinion in Neurobiology, 22, 1027–1038. doi:10.1016/
j.conb.2012.06.001
Levy, I., Snell, J., Nelson, A. J., Rustichini, A., & Glimcher, P. W. (2010). Neural
representation of subjective value under risk and ambiguity. Journal of Neuro­
physiology, 103, 1036–1047. doi:10.1152/jn.00853.2009
Mills, B. A., Reyna, V. F., & Estrada, S. M. (2008). Explaining contradictory rela-
tions between risk perception and risk taking. Psychological Science, 19(5), 429–
433. doi:10.1111/j.1467-9280.2008.02104.x
Mohr, P. N., Biele, G., & Heerkeren, H. R. (2010). Neural processing of risk. Journal
of Neuroscience, 30, 6613–6619. doi:10.1523/JNEUROSCI.0003-10.2010

reward, representation, and impulsivity      37

13490-02_PT1_Ch01-3rdPgs.indd 37 11/15/13 1:42 PM

 Montague, P. R., & Berns, G. S. (2002). Neural economics and biological substrates
of valuation. Neuron, 36, 265–284. doi:10.1016/S0896-6273(02)00974-1
Murray, E. A. (2007). The amygdala, reward, and emotion. Trends in Cognitive Sci­
ences, 11, 489–497. doi:10.1016/j.tics.2007.08.013
Ochsner, K. N., & Gross, J. J. (2008). Cognitive emotion regulation: Insights from
social cognitive and affective neuroscience. Current Directions in Psychological
Science, 17, 153–158. doi:10.1111/j.1467-8721.2008.00566.x
Parker, A., & Fischhoff, B. (2005). Decision-making competence: External valida-
tion through an individual-differences approach. Journal of Behavioral Decision
Making, 18, 1–27. doi:10.1002/bdm.481
Peters, J., & Buchel, C. (2009). Overlapping and distinct neural systems code for
subjective value during intertemporal and risky decision making. The Journal of
Neuroscience, 29, 15727–15734. doi:10.1523/JNEUROSCI.3489-09.2009
Phillips, M. L., Drevets, W. C., Rauch, S. L., & Lane, R. (2003). Neurobiology of
emotion perception II: Implications for major psychiatric disorders. Biological
Psychiatry, 54, 515–528. doi:10.1016/S0006-3223(03)00171-9
Plassmann, H., O’Doherty, J., & Rangel, A. (2007). Orbitofrontal cortex encodes
willingness to pay in everyday economic transactions. The Journal of Neuroscience,
27, 9984–9988. doi:10.1523/JNEUROSCI.2131-07.2007
Plassmann, H., O’Doherty, J. P., & Rangel, A. (2010). Appetitive and aversive
goal values are encoded in the medial orbitofrontal cortex at the time of
decision making. The Journal of Neuroscience, 30, 10799–10808. doi:10.1523/
JNEUROSCI.0788-10.2010
Platt, M. L., & Huettel, S. A. (2008). Risky business: The neuroeconomics of deci-
sion making under uncertainty. Nature Neuroscience, 11, 398–403. doi:10.1038/
nn2062
Pleskac, T. J. (2008). Decision making and learning while taking sequential risks.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 167–
185. doi:10.1037/0278-7393.34.1.167
Pochon, J. B., Riis, J., Sanfey, A. G., Nystrom, L. E., & Cohen, J. D. (2008). Func-
tional imaging of decision conflict. The Journal of Neuroscience, 28, 3468–3473.
doi:10.1523/JNEUROSCI.4195-07.2008
Poldrack, R. A. (2006). Can cognitive processes be inferred from neuroimaging data?
Trends in Cognitive Sciences, 10, 59–63. doi:10.1016/j.tics.2005.12.004
Poldrack, R. A., Kittur, A., Kalar, D., Miller, E., Seppa, C., Gil, Y., . . . Bilder, R. M.
(2011). The cognitive atlas: Towards a knowledge foundation for cognitive neuro­
science. Frontiers in Neuroinformatics, 5(17). doi:10.3389/fninf.2011.00017
Porcelli, A. J., & Delgado, M. R. (2009). Neural systems of reward processing in
humans. In J. C. Dreher & L. Tremblay (Eds.), Handbook of reward and decision
making (pp. 165–184). Oxford, England: Academic Press. doi:10.1016/B978-0-
12-374620-7.00007-8

38       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 38 11/15/13 1:42 PM

 Prévost, C., Pessiglione, M., Météreau, E., Cléry-Melin, M.L., & Dreher, J.C. (2010).
Separate valuation subsystems for delay and effort decision costs. The Journal of
Neuroscience, 30, 14080–14090. doi:10.1523/JNEUROSCI.2752-10.2010
Raichle, M. E., & Snyder, A. Z. (2007). A default mode of brain function: A brief
history of an evolving idea. NeuroImage, 37, 1083–1090, discussion 1097–1099.
doi:10.1016/j.neuroimage.2007.02.041
Reyna, V. F. (1995). Interference effects in memory and reasoning: A fuzzy-trace
theory analysis. In F. N. Dempster & C. J. Brainerd (Eds.), Interference and inhi­
bition in cognition (pp. 29–59). San Diego, CA: Academic Press. doi:10.1016/
B978-012208930-5/50003-9
Reyna, V. F. (2008). Theories of medical decision making and health: An evidence-
based approach. Medical Decision Making, 28(6), 829–833. doi:10.1177/027298
9X08327069
Reyna, V. F. (2012). A new intuitionism: Meaning, memory, and development in
fuzzy-trace theory. Judgment and Decision Making, 7, 332–359.
Reyna, V. F. (2013). Intuition, reasoning, and development: A fuzzy-trace theory
approach. In P. Barrouillet & C. Gauffroy (Eds.), The development of thinking and
reasoning (pp. 193–220). Hove, England: Psychology Press.
Reyna, V. F., & Adam, M. B. (2003). Fuzzy-trace theory, risk communication, and
product labeling in sexually transmitted diseases. Risk Analysis, 23, 325–342.
doi:10.1111/1539-6924.00332
Reyna, V. F., & Brainerd, C. J. (1991). Fuzzy-trace theory and framing effects in
choice: Gist extraction, truncation, and conversion. Journal of Behavioral Deci­
sion Making, 4, 249–262. doi:10.1002/bdm.3960040403
Reyna, V. F., & Brainerd, C. J. (1994). The origins of probability judgment: A review
of data and theories. In G. Wright & P. Ayton (Eds.), Subjective probability
(pp. 239–272). New York, NY: Wiley.
Reyna, V. F., & Brainerd, C. J. (1995). Fuzzy-trace theory: An interim synthesis. Learn­
ing and Individual Differences, 7(1), 1–75. doi:10.1016/1041-6080(95)90031-4
Reyna, V. F., & Brainerd, C. J. (2011). Dual processes in decision making and devel-
opmental neuroscience: A fuzzy-trace model. Developmental Review, 31, 180–
206. doi:10.1016/j.dr.2011.07.004
Reyna, V. F., Chapman, S., Dougherty, M., & Confrey, J. (Eds.). (2012). The adoles­
cent brain: Learning, reasoning, and decision making. Washington, DC: American
Psychological Association. doi:10.1037/13493-000
Reyna, V. F., & Ellis, S. C. (1994). Fuzzy-trace theory and framing effects in children’s
risky decision making. Psychological Science, 5, 275–279. doi:10.1111/j.1467-
9280.1994.tb00625.x
Reyna, V. F., Estrada, S. M., DeMarinis, J. A., Myers, R. M., Stanisz, J. M., & Mills,
B. A. (2011). Neurobiological and memory models of risky decision making in
adolescents versus young adults. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 37, 1125–1142. doi:10.1037/a0023943

reward, representation, and impulsivity      39

13490-02_PT1_Ch01-3rdPgs.indd 39 11/15/13 1:42 PM

 Reyna, V. F., & Farley, F. (2006). Risk and rationality in adolescent decision-making:
Implications for theory, practice, and public policy. Psychological Science in the
Public Interest, 7(1), 1–44. doi:10.1111/j.1529-1006.2006.00026.x
Reyna, V. F., & Lloyd, F. J. (2006). Physician decision-making and cardiac risk:
Effects of knowledge, risk perception, risk tolerance, and fuzzy processing.
Journal of Experimental Psychology: Applied, 12, 179–195. doi:10.1037/1076-
898X.12.3.179
Reyna, V. F., & Mills, B. A. (2007). Converging evidence supports fuzzy-trace the-
ory’s nested sets hypothesis (but not the frequency hypothesis). Behavioral and
Brain Sciences, 30, 278–280. doi:10.1017/S0140525X07001872
Reyna, V. F., & Rivers, S. E. (2008). Current theories of risk and rational decision
making. Developmental Review, 28, 1–11. doi:10.1016/j.dr.2008.01.002
Rick, S. I., Cryder, C. E., & Loewenstein, G. (2008). Tightwads and spendthrifts.
Journal of Consumer Research, 34, 767–782. doi:10.1086/523285
Roiser, J. P., De Martino, B., Tan, G. C. Y., Kumaran, D., Seymour, B., Wood, N. W.,
& Dolan, R. J. (2009). A genetically mediated bias in decision making driven
by failure of amygdala control. The Journal of Neuroscience, 29, 5985–5991.
doi:10.1523/JNEUROSCI.0407-09.2009
Rushworth, M. F., Kolling, N., Sallet, J., & Mars, R. B. (2012). Valuation and decision-
making in frontal cortex: One or many serial or parallel systems? Current Opin­
ion in Neurobiology, 22, 946–955. doi:10.1016/j.conb.2012.04.011
Rushworth, M. F., Noonan, M. P., Boorman, E. D., Walton, M. E., & Behrens, T. E.
(2011). Frontal cortex and reward-guided learning and decision-making. Neuron,
70, 1054–1069. doi:10.1016/j.neuron.2011.05.014
Schonberg, T., Craig, F. R., Mumford, J. A., Congdon, E., Trepel, C., & Poldrack,
R. A. (2012). Decreasing ventromedial prefrontal cortex activity during sequen-
tial risk-taking: An fMRI investigation of the balloon analog risk task. Frontiers
in Decision Neuroscience, 6(80). doi:10.3389/fnins.2012.00080
Schonberg, T., Fox, C. R., & Poldrack, R. A. (2011). Mind the gap: Bridging eco-
nomic and naturalistic risk-taking with cognitive neuroscience. Trends in Cogni­
tive Sciences, 15(1), 11–19. doi:10.1016/j.tics.2010.10.002
Shannon, B. J., Raichle, M. E., Snyder, A. Z., Fair, D. A., Mills, K. L., Zhang,
D., . . . Kiehl, K. A. (2011). Premotor functional connectivity predicts impul-
sivity in juvenile offenders. Proceedings of the National Academy of Sciences of
the United States of America, 108, 11241–11245. doi:10.1073/pnas.1108241108
Singer, M., & Remillard, G. (2008). Veridical and false memory for text: A mul-
tiprocess analysis. Journal of Memory and Language, 59, 18–35. doi:10.1016/
j.jml.2008.01.005
Slotnick, S. D., & Schacter, D. L. (2004). A sensory signature that distinguishes
true from false memories. Nature Neuroscience, 7, 664–672. doi:10.1038/nn1252
Smith, D. V., Hayden, B. Y., Truong, T. K., Song, A. W., Platt, M. L., & Huettel,
S. A. (2010). Distinct value signals in anterior and posterior ventromedial

40       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 40 11/15/13 1:42 PM

 prefrontal cortex. The Journal of Neuroscience, 30, 2490–2495. doi:10.1523/
JNEUROSCI.3319-09.2010
Spaniol, J., Davidson, P. S., Kim, A. S., Han, H., Moscovitch, M., & Grady, C. L.
(2009). Event-related fMRI studies of episodic encoding and retrieval: Meta-
analyses using activation likelihood estimation. Neuropsychologia, 47, 1765–
1779. doi:10.1016/j.neuropsychologia.2009.02.028
Spreng, R. N. (2012). The fallacy of a “task-negative” network. Frontiers in Psychol­
ogy, 3(145). doi:10.3389/fpsyg.2012.00145
Steinberg, L., Cauffman, E., Woolard, J., Graham, S., & Banich, M. (2009). Are ado-
lescents less mature than adults? Minors’ access to abortion, the juvenile death
penalty, and the alleged APA “flip-flop.” American Psychologist, 64, 583–594.
doi:10.1037/a0014763
Stout, J. C., Rock, S. L., Campbell, M. C., Busemeyer, J. R., & Finn, P. R. (2005).
Psychological processes underlying risky decisions in drug abusers. Psychology of
Addictive Behaviors, 19, 148–157. doi:10.1037/0893-164X.19.2.148
Sunstein, C. R. (2008). Adolescent risk-taking and social meaning: A commentary.
Developmental Review, 28, 145–152. doi:10.1016/j.dr.2007.11.003
Taren, A. A., Venkatraman, V., & Huettel, S. A. (2011). A parallel functional topog-
raphy between medial and lateral prefrontal cortex: Evidence and implications
for cognitive control. The Journal of Neuroscience, 31, 5026–5031. doi:10.1523/
JNEUROSCI.5762-10.2011
Tom, S. M., Fox, C. R., Trepel, C., & Poldrack, R. A. (2007). The neural basis of loss
aversion in decision-making under risk. Science, 315, 515–518. doi:10.1126/
science.1134239
Tversky, A., & Kahneman, D. (1986). Rational choice and the framing of decisions.
The Journal of Business, 59, S251–S278. doi:10.1086/296365
Van Duijvenvoorde, A. C. K., Jansen, B. R. J., Bredman, J., & Huizenga, H. M.
(2012). Age-related changes in decision-making: Comparing informed and
noninformed situations. Developmental Psychology, 48, 192–203. doi:10.1037/
a0025601
Venkatraman, V., & Huettel, S. A. (2012). Strategic control in decision making under
uncertainty. European Journal of Neuroscience, 35, 1075–1082. doi:10.1111/
j.1460-9568.2012.08009.x
Venkatraman, V., Payne, J. W., Bettman, J. R., Luce, M. F., & Huettel, S. A. (2009).
Separate neural mechanisms underlie choices and strategic preferences in risky
decision making. Neuron, 62, 593–602. doi:10.1016/j.neuron.2009.04.007
von Neumann, J., & Morgenstern, O. (1944). Theory of games and economic behavior.
Princeton, NJ: Princeton University Press.
von Winterfeldt, D., & Edwards, W. (1986). Decision analysis and behavioral research.
Cambridge, England: Cambridge University Press.

reward, representation, and impulsivity      41

13490-02_PT1_Ch01-3rdPgs.indd 41 11/15/13 1:42 PM

 Wagner, A. D., Shannon, B., Kahn, I., & Buckner, R. (2005). Parietal lobe contri-
butions to episodic memory retrieval. Trends in Cognitive Sciences, 9, 445–453.
doi:10.1016/j.tics.2005.07.001
Weber, E. U., Blais, A.-R., & Betz, N. (2002). A domain-specific risk-attitude scale:
Measuring risk perceptions and risk behaviors. Journal of Behavioral Decision
Making, 15, 263–290. doi:10.1002/bdm.414
Weber, E. U., & Johnson, E. J. (2009). Mindful judgment and decision making. Annual
Review of Psychology, 60, 53–85. doi:10.1146/annurev.psych.60.110707.163633
Weller, J. A., Levin, I. P., Shiv, B., & Bechara, A. (2007). Neural correlates of adap-
tive decision making for risky gains and losses. Psychological Science, 18, 958–
964. doi:10.1111/j.1467-9280.2007.02009.x
Willemsen, M. C., Böckenholt, U., & Johnson, E. (2011). Choice by value encod-
ing and value construction: Processes of loss aversion. Journal of Experimental
Psychology: General, 140, 303–324. doi:10.1037/a0023493
Winecoff, A., Clithero, J. A., Carter, R. M., Bergman, S. R., Wang, L., & Huettel,
S. A. (2013). Ventromedial prefrontal cortex encodes emotional value. Journal
of Neuroscience, 33(27), 11032–11039
World Health Organization. (2010). International statistical classification of diseases and
related health problems (10th ed.). Geneva, Switzerland: Author.
Wunderlich, K., Rangel, A., & O’Doherty, J. P. (2010). Economic choices can be
made using only stimulus values. Proceedings of the National Academy of Sciences of
the United States of America, 107, 15005–15010. doi:10.1073/pnas.1002258107
Yechiam, E., Barron, G., & Erev, I. (2005). The role of personal experience in con-
tributing to different patterns of response to rare terrorist attacks. Journal of
Conflict Resolution, 49, 430–439. doi:10.1177/0022002704270847
Yechiam, E., & Telpaz, A. (2013). Losses induce consistency in risk taking even with-
out loss aversion. Journal of Behavioral Decision Making, 26, 31–40. doi:10.1002/
bdm.758
Yonelinas, A. P., Otten, L. J., Shaw, K. N., & Rugg, M. D. (2005). Separating the brain
regions involved in recollection and familiarity in recognition memory. The Jour-
nal of Neuroscience, 25, 3002–3008. doi:10.1523/JNEUROSCI.5295-04.2005
Zhang, X., & Hirsch, J. (2013). The temporal derivative of expected utility: A
neural mechanism for dynamic decision-making. NeuroImage, 65, 223–230.
doi:10.1016/j.neuroimage.2012.08.063
Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking.
New York, NY: Cambridge University Press.

42       reyna and huettel

13490-02_PT1_Ch01-3rdPgs.indd 42 12/9/13 2:07 PM

 2
Behavioral and Neuroscience
Methods for Studying
Neuroeconomic Processes:
What We Can Learn From
Framing Effects
Irwin P. Levin, Todd McElroy, Gary J. Gaeth,
William Hedgcock, and Natalie L. Denburg

In this chapter, we illustrate how framing effects can be used to examine the
complementary contributions of behavioral and neuroscience research in
understanding neuroeconomic decision processes, particularly those involving
risk. Framing effects are ubiquitous in everyday life, yet they can be studied
under controlled conditions with simple manipulations (e.g., describing a
medical treatment as either having a 50% success rate or a 50% failure rate).
Objectively equivalent information presented in different frames has been shown
to lead to substantially different preference ratings that are accompanied by
different brain activation patterns. An array of multidisciplinary tools across a
growing body of literature shows that framing effects play a fundamental role
in decision making. In this chapter, we explore this recent research using dif-
ferent framing paradigms (attribute and risky-choice framing) in conjunction
with recently developed neuroeconomic risky-choice tasks such as the Iowa
Gambling Task and the “cups” task, which differentially tap into affective
and cognitive systems. Studies include brain scanning, eye tracking, circadian

http://dx.doi.org/10.1037/14322-003
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

43

13490-03_Ch02-3rdPgs.indd 43 11/15/13 1:43 PM

 rhythms, and life-span developmental techniques. We discuss results in terms
of current dual-process theories of decision making. Evidence is converging to
allow us to better answer the basic question of “Why (and when) do you and
I make different decisions?”
Decision making is a complex process that is affected by many factors,
and sometimes the inputs are conflicting. For example, in economic decisions
the most desirable outcome may be the one that involves the most risk. Human
decision makers are characterized by a correspondingly complex, intricately
related system that is bound as much by its basic physiological underpinnings
as it is by its dependency on willful thought. Every psychological and biologi-
cal process operates in conjunction with other processes. Decision making,
like any other task, functions within this synergistic environment and is reli-
ant on both affect and cognition. We focus on framing effects in decision mak-
ing in this chapter because evidence is accumulating at both the behavioral
and the neural levels that framing effects represent an appropriate setting for
examining the individual and complementary roles of affect and cognition in
neuroeconomic decision making. Framing effects are ubiquitous in everyday
life and thus have been a popular topic for the laboratory. We find that frames
affect judgments about gambles, consumer goods and services, medical proce-
dures, and political candidates. We concentrate on valence framing effects,
which are particularly interesting because they represent different responses
to objectively equivalent information by simply varying whether the informa-
tion is expressed in positive or negative terms.
The history of studying framing effects is long and is often traced to
Tversky and Kahneman’s (1981) seminal paper on the Asian disease problem.
New research tools have allowed us to go beyond the early demonstration of
biased responding, that is, departure from normative principles, to examine
in more detail when and why framing effects occur. This in turn provides
deeper insight into basic decision-making processes, especially the ability to
separate cognitive and affective or emotional systems; the separate impact of
these two systems is at the core of our chapter.

Introduction

Evidence is accumulating that common decision-making biases such

as framing effects can be more thoroughly understood at the neural level,
particularly through the activation of affective systems in the brain (De
Martino, Kumaran, Seymour, & Dolan, 2006; Huettel, Stowe, Gordon,
Warner, & Platt, 2006; Kuhnen & Knutson, 2005; Sanfey, Rilling, Aronson,
Nystrom, & Cohen, 2003; Tom, Fox, Trepel, & Poldrack, 2007; Weller, Levin,
Shiv, & Bechara, 2007). Neuroscientific research suggests that because of

44       levin et al.

13490-03_Ch02-3rdPgs.indd 44 11/15/13 1:43 PM

 the evolutionary importance of avoiding negative consequences, the mere
presence of uncertainty induces a primary “fear” response elicited by the
amygdala, which has been associated specifically with fear processing and
avoidance of negative consequences (LeDoux, 2000; Phelps, 2006; Trepel,
Fox, & Poldrak, 2005). This fear response activates the ventromedial pre-
frontal cortex (vmPFC) whose function is to mediate decision making and
to allow for more careful deliberative processes by linking together working
memory and emotional systems (Damasio, 1994). Structures such as the insu-
lar cortex, which are independent of the amygdala, are also likely to impact
decision making under uncertainty by providing complementary systems for
dealing with potential losses (Kuhnen & Knutson, 2005; Weller, Levin, Shiv,
& Bechara, 2009). This account parallels the proposed dual System 1/System 2
approach of decision making (Kahneman, 2003, 2011), which represents an
interaction between emotional intuitive processing and more deliberative
processing of choice options.
The idea that humans process information in either an automatic fash-
ion that is fast or a thoughtful manner that is more deliberative traces its
roots to William James and has been applied in many areas of psychology that
directly relate to decision making (e.g., Epstein, Lipson, Holstein, & Huh, 1992;
Gilbert, 1991; Reyna & Brainerd, 1991; Shiffrin & Schneider, 1977; Sloman,
1996; Stanovich & West, 2000). More recently, Kahneman and Frederick (2002;
see also Kahneman, 2003, 2011) adapted a two-system approach that mirrors
much of the prior contribution by Stanovich and West (2000) and creates a new
vision for the two-process model. In this approach an automatic System 1 oper-
ates without attention or control processes using basic intuitive associations
and emotional intensity and in the end supplies us with quick preferences
that can guide decision making. In contrast to System 1, System 2 is delib-
erative and thoughtful, and its hallmark is that it requires effort. According
to the dual-process approach, much of our decision making is driven by
System 1, which allows for choices to be made with cognitive ease and fac-
tors such as emotional intensity to become the guiding principles for choice
preference. However, if the situation sufficiently motivates us and cognitive
resources are available, then the more thoughtful System 2 is employed, and
more elaborative, effortful, and time-consuming processing occurs. In System
2 processing, attention is directed to the task, alternatives are generated from
memory, thoughtful (e.g., mathematical) comparisons are made, and alterna-
tives may then be weighed in a deliberative fashion.
System 1 is relatively automatic as well as holistic, leads to an automatic
contextualization of problems, and brings potential solutions to consider-
ation relatively quickly. System 2 involves a controlled and analytic process-
ing style that serves to decontextualize and depersonalize problems. From this
route, people will normally rely on individual elements of the task and their

methods for studying neuroeconomic processes      45

13490-03_Ch02-3rdPgs.indd 45 11/15/13 1:43 PM

 logical consistencies. Stanovich and West (2000) further argue that intel-
lect can predispose people to this route of processing, such that people with
higher levels of analytic intelligence will be more likely to rely on System 2
processing.
Across many different approaches we see that cognition and emotion
play complementary roles, with the emotionally laden System 1 quickly pull-
ing us toward an alternative and the cognitively heavy System 2 having the
potential for overriding the initial impulse. The significance of avoiding losses
leads to the belief that the mere framing of equivalent outcomes in terms of
gains or losses could influence the activation of these different systems. In
a recent review, Levin et al. (2012) did in fact show that different neural
systems appear to be activated in dealing with potential gains and potential
losses, especially when neural activation was assessed at the crucial prechoice
stage when positively or negatively framed options were being considered.
A distinction has been made in the literature between reflection effects, a
term referring to tasks involving literal gains and losses (e.g., gambling tasks),
and framing effects, a term referring to equivalent outcomes expressed in terms
of gains or losses (Levin, Schneider, & Gaeth, 1998). Because of the apparent
commonality of processes involved, we will not make this distinction here.
In this chapter, we discuss framing effect differences across tasks and
individuals and what they tell us about why decisions differ both between and
within groups. We stress the complementarity of behavioral and biological/
neurological measures, and we describe studies in which age-related deficits
in cognitive and emotional functioning sometimes lead to impaired decisions
and sometimes to unimpaired decisions. We are aided in this pursuit by subtle
but important distinctions between different framing manipulations.
We include both recently published data and current unpublished data
that highlight the role of framing in illuminating decision processes. Specifically,
we will show the following:
77 Different types of framing manipulations evoke different levels
of emotional reliance.
77 These differences are revealed both in behavioral measures and
biological/neurological indicators.
77 Normal metabolic and biological fluctuations related to circa-
dian rhythms affect decision making and are revealed by different
levels of framing effects at different times of the day.
77 Naturally occurring hemispheric laterality differences are revealed
by handedness impact framing effects.
77 Gaze duration as revealed by eye-tracking techniques provides
further insight into attentional processes underlying framing
effects.

46       levin et al.

13490-03_Ch02-3rdPgs.indd 46 11/15/13 1:43 PM

 77 The most basic of all biological processes, aging, results in changes
in the balance between emotional and cognitive processes that
can lead to impaired performance on some tasks but improved
performance on other tasks in the form of resistance to some
forms of framing.
We then integrate these various findings to better evaluate theories of framing
effects and their more widespread application to understanding basic deci-
sion processes.

Background: Framing Effects

Based on confusions in the literature at the time as to why framing effects

differed across studies and investigators, we suggested that “all frames are not
created equal” (Levin et al., 1998; see also Kühberger, 1998 and Levin, Gaeth,
Schreiber, & Lauriola, 2002) and laid out a scheme for classifying framing stud-
ies in terms of their operational definitions, typical findings, and possible expla-
nations. In this chapter, we focus on two types of framing effects, risky-choice
framing and attribute framing, selected because of their potential for separating
the relative roles of System 1 and System 2 processing in decision making.
Risky-choice framing, as exemplified by the Asian disease problem, requires
a choice between a risky and a riskless option of equal expected value, where
outcomes are alternatively expressed in positive terms (e.g., lives saved by a
medical procedure) or negative terms (lives lost). In the classic example,
respondents are offered a treatment choice for dealing with a disease that
is expected to kill 600 persons. For half the respondents (the positive frame
condition) the choice is between an option that offers a sure saving of 200
lives and an option that offers a 1/3 chance of saving all 600 lives and a 2/3
chance of saving no lives. For the other half (the negative frame condition)
the objectively identical choice is between an option that offers a sure loss
of 400 lives and an option that offers a 1/3 chance of losing no lives and a
2/3 chance of losing all 600 lives. The most common result is that subjects
are more apt to make the risky choice (selecting the option that offers a 1/3
possibility of no lives lost) in the negative condition but choose the riskless
option (a sure saving of 1/3 of the lives) in the positive condition. Later work
revealed a fourfold pattern of risk aversion for potential gains and risk seek-
ing for potential losses of high probability, as in the Asian disease problem,
but the opposite for losses of low probability (Tversky & Kahneman, 1992).
These results are explained in terms of separate value functions for gains and
losses and nonlinear weighting of probabilities in Kahneman and Tversky’s
(1979) prospect theory.

methods for studying neuroeconomic processes      47

13490-03_Ch02-3rdPgs.indd 47 11/15/13 1:43 PM

 Attribute framing, as exemplified by the percentage lean versus percent-
age fat labeling of ground beef (Levin & Gaeth, 1988), is a simpler form of
framing because it involves only the alternative labeling of a single attribute
of an object or event and does not require comparisons based on the element
of risk. The typical finding is more favorable evaluations when alternatives
are labeled in positive terms (percentage lean, success rate) rather than nega-
tive terms (percentage fat, failure rate). These results were explained in terms
of associative networks primed by positive or negative labels (Levin, 1987).
Levin and Gaeth (1988) found that even after consuming an identical
sample of ground beef, those who were told that it was 75% lean ground beef
gave higher ratings of healthiness and quality than did those who were told
that they were tasting 25% fat ground beef. Later, Levin et al. (2002) found
that economic decisions were also affected, with participants willing to spend
more for a package of ground beef with the percentage lean label. In an exten-
sion of this research, Braun, Gaeth, and Levin (1997) investigated attribute
framing effects in a realistic environment where the label was embedded in
an actual product package. Following a taste test, 80% fat-free chocolate was
given higher ratings than 20% fat chocolate, especially by female consumers
for whom the framed attribute was particularly salient.

Applied Attribute Framing

Much of the research on attribute framing effects has emerged from

controlled conditions carried out in the laboratory; however, a number of
studies have demonstrated the relevance for applied decision making. For
example, using practicing physicians, researchers showed that when the
option of surgery was expressed in terms of “survival rate” as opposed to “mor-
tality rate,” significantly more surgeons selected surgery in the survival condi-
tion (McNeil, Pauker, Sox, & Tversky, 1982).
Also in the domain of health, Jasper, Goel, Einarson, Gallo, and Koren
(2001) examined the effects of attribute framing on pregnant women’s per-
ceptions of fetal risk and their subsequent intentions to use a particular drug
during pregnancy. Half the women received positively framed information
(97%–99% chance of having a normal child) and half received the negative
version (1%–3% chance of having a malformed child). Participants in the
negative condition had a significantly higher perception of risk and were
significantly less likely to contemplate taking the drug.
In a study of real consumers, Diamond and Sanyal (1990) showed that
consumers were more likely to redeem coupons when the value was framed
as a gain than when it was framed as a “reduced loss.” Because this was a
situation in which a gain (product plus free item) was compared with a loss

48       levin et al.

13490-03_Ch02-3rdPgs.indd 48 11/15/13 1:43 PM

 (two paid products) and a gain (discount), consumers avoided the coupon
expressed as having a loss and preferred coupons expressed as a gain. In fact,
research has shown that attribute framing plays a profoundly influential role in
many important economic decisions, such as fairness in health care decisions
(Gamliel & Peer, 2010) and allocation of resources (Gamliel & Peer, 2006).
Recently, in a study involving life annuities, a team of researchers looked
at how framing the question of life expectancy affects consumers’ subsequent
estimate of their expected life span and concomitant economic decisions
(Payne, Sagara, Shu, Appelt, & Johnson, 2012). The positive frame asks
people to provide probabilities of living to a certain age or older; the negative
frame asks people to provide probabilities of dying by a certain age or younger.
These two answers should be complements, but Payne et al. (2012) found that
estimated probabilities differed significantly in the two conditions. People
in the live-to frame reported that they have a 55% chance of being alive at
age 85, whereas people in the die-by frame reported that they have a 68%
chance of being dead at age 85. Overall, estimated mean life expectancies,
across three studies and over 2,300 respondents, were between 7.29 and
9.17 years longer when solicited in the live-to frame.
Importantly, Payne et al. (2012) showed that frame-induced life expec-
tancies predicted economic behavioral intentions. Differences in life expec-
tancies were predictive of stated preferences for life annuities, a product that
provides insurance against outliving one’s savings. On a process level, the
authors showed that the attribute framing effect on judgments was partially
mediated by the relative number of thoughts in favor of being alive at that
age. This is consistent with Levin’s (1987) associative model of attribute
framing effects.
Sparked largely by the work of Stanovich and West (Stanovich, 1999;
Stanovich & West, 1998a, 1998b; see also Levin, 1999), there has been a
growing movement to extend aggregate-level analysis of behavioral decision
making to focus on individual differences in rational thought. For example,
although we talk about “typical” findings in studies of risky choice and attri-
bute framing, there are systematic variations in the magnitude and existence
of these effects as a function of specific task and situational characteristics and
their interaction with individual difference factors (Lauriola & Levin, 2001;
Levin et al., 2002; Mahoney, Buboltz, Levin, Doverspike, & Svyantek, 2011;
McElroy & Seta, 2003; Peters & Levin, 2008; Simon, Fagley, & Halleran,
2004; Smith & Levin, 1996). We rely heavily on such systematic sources of
variance when presenting and interpreting the data provided in this chapter.
Before proceeding, we will briefly describe how we think the distinction
between risky-choice framing and attribute framing is important for our pur-
poses. Based on its simpler form and our associative model of its effects, attri-
bute framing seems particularly amenable to System 1 processing. Specifically,

methods for studying neuroeconomic processes      49

13490-03_Ch02-3rdPgs.indd 49 11/15/13 1:43 PM

 attribute framing problems are comparatively simple and can be processed
with relative ease. And because System 1 is the default processing style for
cognitively easy tasks, it seems likely that System 1 processing will dominate
for most individuals.
Conversely, because risky-choice framing has the added element of pro-
cessing numeric risk information and comparing risky and riskless options,
both elements associated with System 2, we presume that System 2 process-
ing is more likely to occur in risky-choice framing. To preview what comes
later, we use this distinction to better understand why some individuals show
biased responses on a risky-choice framing task but not on an attribute fram-
ing task and why different brain mechanisms are involved.
Because we are interested in the process by which framing guides choice
and how the emotional component of this process differs across types of fram-
ing, for some of the more recent studies we constructed slightly modified
versions of the standard risky-choice framing and attribute framing tasks
in which the response scale is equated across tasks. This serves to remove
any confound between the deliberation of choice options and the mechani-
cal process of response elicitation, which becomes particularly important in
brain imaging studies. As illustrated below, a 4-point preference scale is used
for each. The positive framing conditions are shown.

Example of Risky-Choice Framing

Because of changes in tax laws, you may get back as much as $1,200 in
income tax. Your accountant has been exploring two ways to take advantage
of this situation:
77 If Plan A is adopted, you will get back $400 of the possible
$1,200.
77 If Plan B is adopted, you have a 33% chance of getting back
all of the $1,200, and a 67% chance of getting back no money.
77 Which option do you prefer?
77 Strongly Prefer A Weakly Prefer A
Strongly Prefer B Weakly Prefer B

Example of Attribute Framing

As R&D manager, two project teams have come to you requesting addi-
tional funds for a project you instituted several months ago. You can only
fund one team.
77 Team A is proposing a project that has historically been suc-
cessful in 24 of the last 50 attempts. It will cost $100,000.

50       levin et al.

13490-03_Ch02-3rdPgs.indd 50 11/15/13 1:43 PM

 77 Team B is proposing a project that has historically been success-
ful in 20 of the last 50 attempts. It will cost $65,000.
77 Which option do you prefer?
77 Strongly Prefer A Weakly Prefer A
Strongly Prefer B Weakly Prefer B
Note that in the attribute framing example, Team A has a better record
but higher cost than Team B. Across conditions, only success versus failure
rate is manipulated; cost does not change. Thus, based on the associative
model of attribute framing effects (Levin, 1987), positive framing is expected
to lead to greater preference for A, because focusing on successes will make
Team A seem especially attractive compared with Team B, even with the
higher cost. In contrast, by focusing on failures in the negative condition,
Team A will seem less attractive and not worth the extra cost, which will lead
to greater preference for B.

Emotion Suppression

One way to examine the differential role of affect or emotion in deci-

sion tasks is to manipulate the extent of emotional processing through
instructional variations, including emotion suppression. Gross and Levenson
(1993) defined emotion suppression as the conscious inhibition of emotional
expressive behavior while emotionally aroused. In their study, participants
who watched a disgust-eliciting film and were subject to emotion suppres-
sion instructions showed decreased somatic activity and decreased heart rate
compared with controls. Using this manipulation, we provide preliminary
data on the comparative role of emotion in the two types of framing tasks.
A 2 (instructions) × 2 (frame type: risk/attribute) × 2 (valence: positive/
negative) × 10 (domains) design was employed where the first factor was
manipulated between subjects and the rest within. Within-subject manipula-
tion of frame type and valence allows for each type of framing effect to be mea-
sured and compared at the level of the individual respondent. Domain refers
to the varied content areas of the problems (financial, medical, recreational,
etc.) included to increase generality.
The baseline condition instructed participants to indicate their prefer-
ence using the 4-point scale described earlier. The emotion control condition
was identical except for the additional instruction to adopt a “detached and
unemotional attitude.” Risky-choice framing resulted in the usual finding of
significantly greater risk taking in the negative framing condition than in the
positive condition, both in the baseline condition and in the suppression con-
dition. By contrast, attribute framing resulted in the typical finding of lower

methods for studying neuroeconomic processes      51

13490-03_Ch02-3rdPgs.indd 51 11/15/13 1:43 PM

 evaluations for negatively framed stimuli only in the baseline condition but
not in the suppression condition. The same participants who exhibited risky-
choice framing effects in the emotion suppression condition were resistant to
attribute framing effects.
Given that resistance to framing is seen as a component of decision-
making competence (Bruine de Bruin, Parker, & Fischhoff, 2007; Parker &
Fischhoff, 2005; Weller, Levin, Rose, & Bossard, 2012), our results indicate
that in the case of attribute framing, suppression of emotions is associated with
reducing or eliminating biased decision making. Because attribute framing was
more affected by emotion suppression than risky-choice framing, the results
are consistent with the hypothesis that attribute framing is more influenced by
emotional (System 1) processes than risky-choice framing, and in that sense,
the sources of the two types of framing differ at a basic level.
Apart from task and individual characteristics that promote delibera-
tive versus intuitive processing during choice, there is evidence that the same
person performing the same task may vary in processing choices simply due to
natural biological processes that vary across times of the day. Again, framing
effects shed light on such processes.

Brain Metabolic Levels Influence Processing

of Risky-Choice Framing Tasks

The body’s metabolic activity influences all aspects of physiological and

psychological functioning. Factors that affect metabolic activity should either
enhance or inhibit System 2–type processing. This change in cognitive pro-
cessing should, in turn, provide clues about the extent to which deliberative
System 2 processing is involved in risky choice and attribute framing. To explore
this, we focus on two main contributors to metabolic activity: circadian rhythms
and glucose. Both of these factors have recently drawn the interest of decision
researchers, largely because of their effects on cognitive processing.
We live in an environment that is filled with many externally regu-
lated changes, but the most profound is the day–night cycle. Not surprisingly,
our bodies have developed a host of biological fluctuations that accompany
variations in the diurnal cycle, all geared toward giving us specific temporal
advantages. On any normal day, our internal biological clock cycles through
changes in a variety of physiological functions that include corporal tempera-
ture, hormones, heart rate, and blood pressure. And the list is not limited to
biological functions; cognitive abilities also vary with the circadian cycle.
These circadian variations occur on a roughly 24-hour cycle and appear to
be controlled by an area deep within our brain, the suprachiasmatic nucleus
(SCN), which acts as a circadian oscillator.

52       levin et al.

13490-03_Ch02-3rdPgs.indd 52 11/15/13 1:43 PM

 Although a great deal of evidence supports the SCN as the major player
in circadian rhythm, perhaps the most convincing evidence for its central
role was put forth by Ralph, Foster, Davis, and Menaker (1990) when they
grafted the SCN of a mutant strain of hamster with very short circadian
cycles onto hamsters with a nonfunctioning SCN. Ralph et al. found that the
recipient hamster always took on the circadian cycle of the mutant donor,
even though it was different from their previous cycles. Thus, it appears that
the SCN is the controller of circadian cycles.
Given the important biological functions that are regulated by the cir-
cadian rhythm cycle, it seems evident that diurnal cycles will have important
influences on many aspects of human behavior, including decision making.
Studies investigating this topic have shown that circadian mismatched par-
ticipants (e.g., morning type in the evening) rely more on judgmental heu-
ristics, which lead to greater stereotyping (Bodenhausen, 1990), and they
have lower levels of strategic reasoning when making decisions (Dickinson
& McElroy, 2010, 2012). Both findings suggest attenuation in deliberative
thought.
McElroy and Dickinson (2010) investigated how changes in deliberative
thought, observed across the circadian cycle, influenced risky-choice and attri-
bute framing. To achieve this, they used online survey software and randomly
assigned participants to specific time slots across the full 24-hour cycle. Overall,
these findings showed that risky-choice framing effects were significantly stron-
ger during circadian off-times than during circadian on-times and the effect was
most pronounced in the losses condition where risk-taking was especially high
during off times. These changes in framing effects were interpreted through a
dual-process approach wherein more automatic processing is hypothesized to
occur during circadian off-times and more thoughtful, deliberative processing
is hypothesized to occur during on-times. Importantly, circadian variation had
no effect on attribute framing.
In tune with circadian rhythm, a variety of metabolic and physiological
processes operate in conjunction with SCN oscillation. One of the key meta-
bolic factors that varies is glucose. The SCN has been shown to regulate daily
cycles in blood glucose levels to create an interactive circadian arousal system
that runs in parallel with the glucose fuel source (e.g., La Fleur, 2003). The key
benefactor of glucose utilization is the brain—it accounts for roughly 25% of
the body’s total glucose utilization.
Researchers have begun to focus on how glucose levels may influence
decision making. For example, Masicampo and Baumeister (2008) used an
attraction task wherein participants are attracted to an option because it is
similar to a decoy that is not desirable. When choosing between two apart-
ments based on their size and distance from school, the introduction of a
third option (the decoy), which is inferior to the others on both attributes,

methods for studying neuroeconomic processes      53

13490-03_Ch02-3rdPgs.indd 53 11/15/13 1:43 PM

 can nevertheless affect the choice between the two most viable options.
They found that glucose-deficient participants were more likely to make less
optimal choices (rely on the decoy) than glucose-enriched participants. In
another investigation, McMahon and Scheel (2010) looked at probability
learning and found that participants relied more on simple probability rules
when they were glucose deprived, and as a result, mimicked the percent-
age of occurrence for each event (a less advantageous strategy) rather than
choosing the event of greater likelihood. Further research by Wang and
Dvorak (2010) showed that glucose enrichment reduced the rate of future
discounting, suggesting a better ability to regulate expected rewards. Thus,
these findings seem to indicate that glucose deprivation impedes a decision
maker’s ability to ignore irrelevant information, apply more complex rules,
and thoughtfully consider future rewards. This seems to suggest that glucose
acts as circadian fuel either to inhibit or to encourage more thoughtful pro-
cessing of decisions.
These circadian rhythm and glucose findings provide further evidence
that processing differences have pronounced effects on risky-choice framing.
It is well established that cognitive processing is constrained by the cognitive
resources currently available (Hasher & Zacks, 1988). Behavioral research has
shown that during circadian off-times cognitive resources are depleted (e.g.,
Schmidt, Collette, Cajochen, & Peigneux, 2007). Furthermore, neurologi-
cal evidence indicates that during circadian off-times cognitive resources are
constrained because of reduced frontal lobe functioning (e.g., Manly, Lewis,
Robertson, Watson, & Datta, 2002). Together, both behavioral and neuro-
logical evidence suggest that circadian off-times decrease glucose levels and
inhibit cognitive resources. Consequently, deliberative, effortful processing
should be impeded during circadian off-times and advanced during on-times.
This parallels circadian effects on risky-choice framing wherein framing
effects are strongest during off-times and weakest during on-times. However,
circadian rhythm had no effect on attribute framing. Taken together, these
findings suggest that deliberative processing is a central component in risky-
choice framing but plays little or no role in attribute framing, thus solidifying
the distinction between risky-choice and attribute framing as representing
different balances of cognitive and affective processing.

Handedness and the Hemispheres:

Connections and Processing Styles

Another important link between processing styles and biological pro-

cesses is handedness as an observable element of right brain/left brain func-
tions. There has been a great deal of research investigating how handedness

54       levin et al.

13490-03_Ch02-3rdPgs.indd 54 11/15/13 1:43 PM

 differences represent underlying anatomical differences in the brain. A grow-
ing body of research has been focusing on how these differences may affect
decision making (e.g., Christman, Jasper, Sontam, & Cooil, 2007; Jasper,
Barry, & Christman, 2008; Jasper & Christman, 2005). At the heart of this
literature lies the belief that mixed-handed people have greater interaction
between the left and right hemispheres and consequently greater access to
right-hemisphere processing than “strong-handers” (those with a dominant
hand). Research based on this premise has shown that this access leads “mixed-
handers” to demonstrate more risk propensity and stronger informational
framing effects (Jasper, Fournier, & Christman, 2013).
How processing differences associated with the respective hemispheres
may influence risky-choice framing has also been a topic of research. Historically,
hemispheric research suggests that the left hemisphere processes informa-
tion in an analytic fashion, whereas the right hemisphere tends to process
information in a more holistic style (e.g., Ornstein, 1972). The outcome of
these respective processing styles is that left hemisphere processing is reliant
on numerically derived information and the right hemisphere is especially
sensitive to context cues.
McElroy and Seta (2004) manipulated hemispheric activation to capi-
talize on the different hemispheric processing styles and test their effects
on a traditional risky-choice framing task. Their findings revealed that the
respective hemispheric processing styles had significant effects on risky-choice
framing, with pronounced framing effects under holistic processing and no
framing effects when analytic processing was induced. Thus, both manipu-
lated hemispheric activation and naturally occurring differences provide con-
verging evidence of the role of processing style on risky-choice framing effects.

How Physiological Measures Can Complement

Behavioral Measures in Testing Theories
of Decision Making

The following examples also illustrate how the use of framing tasks in
conjunction with physiological/biological indicators can help us understand
and refine our theories about decision processes. Traced back to the early work
of Kahneman and Tversky (1979), prospect theory represents a psychophysical
approach to describing rational choice; yet at its core, the approach maintains
a theoretical basis that is consistent with traditional approaches to decision
making. The traditional approach assumes that decision makers mathemati-
cally combine quantitative information by transposing numerical amounts
and probabilities into some quantitative value or utility that is then compared
and contrasted across alternatives. This assumption has its foundation in some

methods for studying neuroeconomic processes      55

13490-03_Ch02-3rdPgs.indd 55 11/15/13 1:43 PM

 of the earliest beginnings of rational decision making (Bernoulli, 1738/1954)
and during its formative stages (e.g., Edwards, 1954).
The more recent fuzzy trace theory (Reyna, 2012; Reyna & Brainerd,
2011), like earlier dual process theories, involves two complementary styles
that can lead decision makers to different choice preferences. However, fuzzy
trace theory differs from prospect theory and other decision-making theories
first and foremost because under most decision-making conditions it does not
take a quantitative transformation approach. Rather, fuzzy trace theory is a
dual process, memory-based theory in which one memory system involves
precise, verbatim representations of information and extracts a level of detail
critical for more analytic comparisons. The other memory system extracts only
vague “fuzzy” impressions but is the source of meaning and captures gener-
alized “gist” appeal that directs an intuitive inclination toward an alterna-
tive. These two memory systems operate in parallel fashion, each individually
extracting representations from a target stimulus and then encoding and stor-
ing the information independently. The end result for decision making is that
these functionally independent memory systems often lead people to different
choice preferences.
According to fuzzy trace theory, adults normally process in as fuzzy and
impressionistic a way as is reasonable, and consequently they normally make
decisions using only vague representations that capture the gist of the task’s
information. Thus, in gist processing, numeric information is extracted in a
very simplified form and precise quantities are often neglected. For example,
33% might be extracted as “some,” 67% as “most,” and 0% as “none.” As a
consequence of this processing style, alternatives are perceived in an ordinal
gist and individuals are directed by a sense of “intuition” toward the alterna-
tive with the greatest extracted value. Applying this to framing effects, in the
positive framing condition of the Asian disease problem, saving some lives is
more attractive than taking a chance that no lives will be saved, and in the
negative condition, losing some lives is less attractive than having a chance of
losing no lives. Likewise, in the ground beef attribute framing problem, “fat”
is less desirable than “lean.”
Recently, a good deal of research has set out to test fuzzy trace theory
with prospect theory and other utility models. A means for testing these theo-
ries has emerged that involves the importance of presenting “0” (null) infor-
mation to the decision maker. Because prospect theory and other standard
utility models rely on a numerical transformation of the information, the
presence or absence of the zero complement (e.g., 2/3 chance of saving no
lives) should be irrelevant because the quantitative value of the alternative
(1/3 chance of saving all 600) does not change with its presence. However, in
fuzzy trace theory, null information is vital in gist processing as it represents a
basic categorical distinction that the decision maker will use for comparison.

56       levin et al.

13490-03_Ch02-3rdPgs.indd 56 11/15/13 1:43 PM

 For example, in gist processing the alternative “1/3 chance of saving all 600
lives and a 2/3 chance of saving no lives” should be encoded as “some peo-
ple saved and none saved,” which creates a strikingly different representa-
tion than when the zero complement is removed resulting in “some saved.”
Kühberger and Tanner (2010) tested this theoretical assumption by manipu-
lating whether participants had access to the zero complement. Kühberger
and Tanner found that framing effects were present when the zero comple-
ment was presented but were not observed when the zero complement was
removed, and thus supported fuzzy trace theory assumptions.
In a recent study, McElroy, Dickinson, Corbin, and Beck (2013) used
eye-tracking technology during a risky-choice task to test the more quantita-
tive processing approach of prospect theory and the memory-based process-
ing of fuzzy trace theory. Eye-tracking technology allows comparative analysis
of discrete eye movements and can measure gaze duration, thus allowing
researchers to operationalize “processing” by associating inward cognitive
processing with observational differences in eye gaze duration. McElroy et al.
hypothesized that because of prospect theory’s traditional decision-making
approach, numerical transformation of the quantitative information deter-
mines value, and no difference in gaze duration should be observed between
those who demonstrate a risky-choice framing effect and those who do not.
Conversely, because zero-complement information is so important in gist pro-
cessing, they predicted from fuzzy trace theory that gaze duration of the zero
complement (i.e., the amount of time the pupil rests on the null information)
should be greater for those who demonstrate risky-choice framing effects.
McElroy et al.’s (2013) findings were consistent with fuzzy trace theory
in the losses condition where participants who demonstrated framing effects
had significantly longer gaze duration for the zero complement. However, in
the gains condition no significant differences were observed, suggesting that
the gains condition may lack sufficient intensity. This is consistent with the
special role of loss aversion (Weller et al., 2009). Thus, in this example eye-
tracking technology is used to corroborate behavioral data and show how
physiological techniques and behavioral data can complement one another.
Furthermore, it shows how processing differences underlie risky-choice fram-
ing effects.

Decision Making in Older Adults

For the closing sections of this chapter, we turn to a basic biological

process—aging—and its relation to decision making in framing and other
neuroeconomic tasks. Studies of the life-span trajectory of brain functions pro-
vide insight into age-related differences in decision making, which, in turn, are

methods for studying neuroeconomic processes      57

13490-03_Ch02-3rdPgs.indd 57 11/15/13 1:43 PM

 associated with changes in the brain. Much of this research has focused on the
early preadulthood ages when incomplete maturation of some brain regions
leads to decision-making deficits. For example, functional maturation of the
prefrontal cortex has been shown to be protracted compared with the devel-
opment of subcortical reward-processing structures, and this limits top-down
control processes in the processing of fear and reward signals (Casey, Giedd,
& Thomas, 2000; Crone & van der Molen, 2004; Galvan et al., 2006; Hare
& Casey, 2005). The result is often inappropriate risk taking at the younger
ages. These findings suggest a neurobiological basis for age-related differences
in decision making.
At the other end of the spectrum, there is growing support for the frontal
lobe hypothesis of aging (Brown & Park, 2003; Pardo et al., 2007; Resnick,
Pham, Kraut, Zonderman, & Davatzikos, 2003; West, 1996, 2000) whereby
age-related structural declines in the prefrontal cortex, particularly the orbito-
frontal and lateral prefrontal cortices, lead to declines in executive function-
ing tasks in the elderly that are mediated by frontal lobe functioning (West,
Murphy, Armillo, Craik, & Stuss, 2002). However, results are not always
clear-cut as to when older adults do and do not exhibit decision deficits.
Some researchers have found that older adults were less sensitive to risk
level than younger adults (Deakin, Aitken, Robbins, & Shahakian, 2004;
Denburg, Tranel, & Bechara, 2005; Weller, Levin, & Denburg, 2011), but
other researchers have found that older adults performed as well as younger
adults on risky decision-making tasks (Kovalchik, Camerer, Grether, Plott,
& Allman, 2005; Wood, Busemeyer, Kolings, Cox, & Davis, 2005). Bruine
de Bruin, Parker, and Fischhoff (2012) explained some of these mixed results
by showing that some decision-making skills decline with age while others
remain unchanged or improve. One theme of the current chapter is that the
net effect of this age-related mixture of skill levels depends on task demands.
In studying age-related differences in decision making, Weller et al. (2011)
used a relatively new task, the cups task (Levin & Hart, 2003), designed to
mimic the risky-choice framing task but with actual gains and losses. Analogous
to the separate gain- and loss-framed versions of the risky-choice framing task,
the cups task includes separate trials involving risky gains and risky losses.
Because of its economic consequences and its potential usefulness in scanner
research, the cups task is ideally suited for neuroeconomic research.
Gain trials require the choice between an option that offers a sure gain
of one quarter and another option that offers a designated probability of win-
ning multiple quarters or no quarters, where probability information is con-
veyed simply by the number of cups from which to choose. This property
allows the task to be administered to all age groups. Loss trials require the
choice between a sure loss of one quarter and a designated probability of
losing multiple quarters or no quarters.

58       levin et al.

13490-03_Ch02-3rdPgs.indd 58 11/15/13 1:43 PM

 On each trial an array of two, three, or five cups is shown on each side
of a computer screen. One array is identified as the certain or riskless side
where one quarter would be gained (lost) for whichever cup was selected.
The other array is designated as the risky side where the selection of one cup
would lead to a designated number of quarters gained (lost) and the other
cups would lead to no gain (loss). Participants select one cup from either the
riskless or the risky side. By manipulating the number of cups from which
to choose and the possible outcome magnitude, some trials represent “risk
advantageous” choices because the expected value of a risky choice is more
favorable than the sure gain or loss (e.g., one out of three chances of winning
five quarters vs. winning one quarter for sure, or one out of five chances of
losing three quarters vs. losing one quarter for sure). Conversely, other trials
are “risk disadvantageous” because the expected value of a risky choice is less
favorable than the sure gain or loss (e.g., one out of five chances of winning
three quarters, or one out of three chances of losing five quarters).
These design features allow for comparisons across age groups of overall
risk taking to achieve gains or avoid losses and of the tendency to make risk-
advantageous or -disadvantageous choices for risky gains and losses. Overall
level of risk taking to achieve gains was shown to decrease steadily across the
life span, while risk seeking to avoid a loss was remarkably constant across age
levels, a result attributed to the pervasiveness of loss aversion (Weller et al.,
2011). All but the youngest groups showed the classic risky-choice framing
effect of more risk taking to avoid a loss than to achieve a gain of the same
magnitude. More interestingly, age-related differences were found in the ten-
dency to make risk-advantageous/disadvantageous choices. The difference in
the number of risk-advantageous and risk-disadvantageous choices increased
from childhood through young and middle adulthood but decreased for those
65 years and older. This demonstrates that the sensitivity to risk-relevant
information decreases in older adults in a manner consistent with the frontal
lobe hypothesis.

Framing Effects for “Impaired” and “Unimpaired”

Older Decision Makers

The goal of this line of research is to identify when older adults do and do
not display impaired decision making and to identify the neurological corre-
lates. Earlier we showed that emotion suppression was associated with reduced
attribute framing effects. Another task in which emotional involvement
plays a key role is the Iowa Gambling Task (IGT; Bechara, 2007; Bechara,
Damasio, Tranel, & Damasio, 1997). Because of its wide use with brain-
damaged patients and in scanner research, the IGT has perhaps become the

methods for studying neuroeconomic processes      59

13490-03_Ch02-3rdPgs.indd 59 11/15/13 1:43 PM

 prototypical “neuroscience” task. Successful performance on the most widely
used version of this task requires the avoidance of choices providing imme-
diate reward but larger subsequent losses. In a recent study using this task,
Bauer et al. (2013) found age-related deficits in performance consistent with
the notion that decisions by older persons are disproportionately influenced
by prospects of receiving rewards, irrespective of the degree of punishment.
Participants are asked to select cards from four different decks, each
with its own distribution of gains and losses. No information about differ-
ences between decks is given ahead of time, and participants have to learn
from repeated trials which are the two “good” decks and which are the two
“bad” decks. Repeated sampling from the two good decks will lead to greater
gains than losses in the long run, whereas sampling from the bad decks, while
offering initial gains, will lead to greater losses than gains in the long run.
Bechara, Damasio, Damasio, & Anderson (1994) and Bechara et al. (1997)
showed that patients with lesions to the vmPFC were unable to learn in this
task because the activation of brain mechanisms that signal potential loss
based on prior experiences was dampened.
In a current preliminary study of IGT impairment and framing effects
in older decision makers (Hedgcock, Denburg, Levin, & Halfmann, 2012),
adults over age 65 were administered both the IGT and the framing problems.
They were classified as “unimpaired” (n = 16) if they selected significantly
more cards from the good decks than from the bad decks or as “impaired” (n =
16) if they selected significantly more cards from the bad decks than the good
decks. (“Unimpaired” older adults are indistinguishable from healthy young
adults on the IGT.) Functional magnetic resonance imaging (fMRI) activa-
tion patterns were examined as a function of IGT score and type of framing
effect. The difference between brain activation for the positive and negative
versions of each type of framing manipulation was correlated with IGT score.
This allowed us to determine which parts of the brain were most sensitive to
individual differences in performing the two types of framing problems.
For attribute framing difference scores, IGT scores correlated with activ-
ity in cortical midline structures including the vmPFC, the dorsomedial pre-
frontal cortex, and the anterior cingulate cortex. Of particular interest, the
vmPFC has been implicated in processing emotional stimuli (Damasio, 1994;
Northoff & Bermpohl, 2004). (See, however, Huettel’s 2010 cautionary note
that some brain regions can be labeled as making cognitive or emotional con-
tributions in different contexts.) Perhaps even more important, these areas
have been implicated in age-related declines. Pardo et al. (2007) found that
the largest declines in brain activity with normal aging were localized in a
medial network including the anterior cingulate/medial prefrontal cortex.
The only area associated with risky-choice framing in the present study was
the parietal cortex.

60       levin et al.

13490-03_Ch02-3rdPgs.indd 60 11/15/13 1:43 PM

 These results provide evidence, backed by earlier reports in this chapter,
that attribute framing and risky-choice framing are different in the extent to
which they evoke emotional processing. Furthermore, this work shows that
different levels of emotional processing can help or hurt rational decision
making depending on task demands. Finally, they help explain why the aging
process can simultaneously hurt performance on some tasks but improve per-
formance on others as structural changes in the brain influence response to
cues that are sometimes relevant and sometimes not.

Summary, Conclusions, and Future Research

In this chapter, we focused on framing effects to help readers understand

neuroeconomic processes for several reasons:
77 framing effects represent a well-known decision bias with impor-
tant real-world consequences
77 framing effects have a long history of basic research in behavioral
decision making and a shorter, but growing, history of research
in decision neuroscience and neuroeconomics
77 framing effects are not an isolated phenomenon but are tied to
broader principles such as loss aversion or risk aversion
77 the simplicity of tasks that demonstrate framing effects allows
for use with populations ranging from young children to the
elderly and persons with known decision impairments
77 they can easily be studied using technologically current methods
such as brain scanning or eye tracking as well as in the traditional
laboratory or online setting.
New features illustrated in the applications sampled here include the
identification of distinct types of framing effects with a different balance of
cognitive processing and emotional involvement. This then allows the differ-
ent paradigms to be used to address issues of broader scope, such as age-related
changes in the balance of skills that lead to impaired decision making on some
tasks but not on others. Another important feature is that with proper experi-
mental design features, framing effects can be studied not just in the aggregate
but at the level of the individual decision maker. This allowed us, for example,
to show which areas of the brain are differentially activated as a function of
the size of the framing effect displayed by different decision makers.
The evidence provided about the concordance of biological and behav-
ioral factors associated with the circadian rhythm cycle came from the use
of a risky-choice framing task administered at different points in the cycle.
Susceptibility to risky-choice framing effects was significantly greater during

methods for studying neuroeconomic processes      61

13490-03_Ch02-3rdPgs.indd 61 11/15/13 1:43 PM

 circadian off-times than during circadian on-times and this was related to the
actions of an area deep in the brain, the SCN, which governs glucose level
and metabolic activity, a physiological correlate of effort. Equally insightful
was the revelation that circadian cycle had no impact on attribute framing,
a finding supporting our hypothesis of different processes for attribute and
risky-choice framing.
Related findings were that eye-fixation patterns were predictive of
the presence versus absence of risky-choice framing effects and that right-
hemispheric versus left-hemispheric processing was also related to risky-
choice framing effects, each result being indicative of variations in cognitive
processing. The important point for the present purposes is that these demon-
strations of the relation between brain functions and behavior using a fram-
ing paradigm serve as a tool for understanding such relations. McElroy et al.
(2013) were able to use their measure of gaze duration to assess the dual pro-
cessing assumptions of fuzzy trace theory.
In a similar vein, the emotion suppression study revealed that the two
types of framing manipulations—attribute framing and risky-choice framing—
do in fact call into play fundamentally different forms of decision making.
Attribute framing appears to involve a purer form of emotionally driven
choice, similar to System 1 processing, whereas risky-choice framing involves
more deliberative analysis, similar to System 2 processing. This was confirmed
in an fMRI study in which attribute framing was more closely linked to activa-
tion of brain areas associated with the emotional system than was risky-choice
framing. These results then provide impetus for using this imbalance to tar-
get specific populations, in this case the older decision maker, whose deci-
sion making in different domains is likely affected by varying levels of specific
functions. The fMRI study with older adults not only affirmed the relation
between types of framing and differential activation of brain systems but also
showed that those systems exhibiting age-related declines can sometimes lead
to impaired performance, such as on the IGT where learning from past errors
is key, but can sometimes lead to improved performance, such as resistance to
the emotional lure of attribute framing.
The illustrations focusing on framing effects in this chapter provide only
a thin slice of the fascinating phenomena revealed in a systematic study of the
neuroscience of human judgment and decision making. One suggestion we
have for future researchers is to not consider framing effects as a unified con-
cept but to specify how they are operationalized in ways that can account for
different results in different studies. It is our hope that the key features illus-
trated here will serve future researchers investigating other phenomena that
separate normatively rational behavior from the reality of the human experi-
ence. Among these features are the use of brain imaging, eye-tracking, and
other biological measures to complement traditional behavioral measures;

62       levin et al.

13490-03_Ch02-3rdPgs.indd 62 11/15/13 1:43 PM

 inclusion of both variables manipulated in the laboratory and individual
difference factors that account for why you and I make different decisions,
particularly as these individual differences interact with task demands; and,
finally, strategic selection of participant populations with distinctive charac-
teristics to provide tests of theories as well as applications of those theories.

References

Bauer, A. S., Timpe, J. C., Edmonds, E. C., Bechara, A., Tranel, D., & Denburg,
N. L. (2013). Myopia for the future or hyposensitivity to reward? Age-related
changes in decision making on the Iowa Gambling Task. Emotion, 13(1), 19–24.
doi:10.1037/a0029970
Bechara, A. (2007). Iowa Gambling Task (IGT) professional manual. Lutz, FL: Psycho-
logical Assessment Resources.
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994). Insensitivity
to future consequences following damage to human prefrontal cortex. Cognition,
50, 7–15. doi:10.1016/0010-0277(94)90018-3
Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1997). Deciding advanta-
geously before knowing the advantageous strategy. Science, 275, 1293–1295.
doi:10.1126/science.275.5304.1293
Bernoulli, D. (1954). Exposition of a new theory on the measurement of risk. Econo-
metrica, 22, 23–36. (Original work published 1738)
Bodenhausen, G. Y. (1990). Stereotypes as judgmental heuristics: Evidence of circa-
dian variations in discrimination. Psychological Science, 1, 319–322. doi:10.1111/
j.1467-9280.1990.tb00226.x
Braun, K. A., Gaeth, G. J., & Levin, I. P. (1997). Framing effects with differential
impact: The role of attribute salience. In M. Brucks & D. J. Mac Innis (Eds.),
Advances in consumer research, Vol. 24 (pp. 405–411). Provo, UT: Association
for Consumer Research.
Brown, S. C., & Park, D. C. (2003). Theoretical models of cognitive aging and impli-
cations for translational research. The Gerontologist, 43, 57–67. doi:10.1093/
geront/43.suppl_1.57
Bruine de Bruin, W., Parker, A. M., & Fischhoff, B. (2007). Individual differences in
adult decision-making competence. Journal of Personality and Social Psychology,
92, 938–956. doi:10.1037/0022-3514.92.5.938
Bruine de Bruin, W., Parker, A. M., & Fischhoff, B. (2012). Explaining adult age dif-
ferences in decision-making competence. Journal of Behavioral Decision Making,
25, 352–360. doi:10.1002/bdm.712
Casey, B. J., Giedd, J. N., & Thomas, K. M. (2000). Structural and functional brain
development and its relation to cognitive development. Biological Psychology,
54, 241–257. doi:10.1016/S0301-0511(00)00058-2

methods for studying neuroeconomic processes      63

13490-03_Ch02-3rdPgs.indd 63 11/15/13 1:43 PM

 Christman, S. D., Jasper, J. D., Sontam, V., & Cooil, B. (2007). Individual differ-
ences in risk perception versus risk taking: Handedness and interhemispheric
interaction. Brain and Cognition, 63, 51–58. doi:10.1016/j.bandc.2006.08.001
Crone, E. A., & van der Molen, M. W. (2004). Developmental changes in real life
decision making: Performance on a gambling task previously shown to depend
on the ventromedial prefrontal cortex. Developmental Neuropsychology, 25, 251–
279. doi:10.1207/s15326942dn2503_2
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the human brain. New York,
NY: Putnam.
Deakin, J., Aitken, M., Robbins, T., & Sahakian, B. J. (2004). Risk taking during
decision-making in normal volunteers changes with age. Journal of the Interna-
tional Neuropsychological Society, 10, 590–598. doi:10.1017/S1355617704104104
De Martino, B., Kumaran, D., Seymour, B., & Dolan, R. J. (2006). Frames, biases, and
rational decision-making in the human brain. Science, 313, 684–687. doi:10.1126/
science.1128356
Denburg, N. L., Tranel, D., & Bechara, A. (2005). The ability to decide advanta-
geously declines in some normal older persons. Neuropsychologia, 43, 1099–1106.
doi:10.1016/j.neuropsychologia.2004.09.012
Diamond, W. D., & Sanyal, A. (1990). The effect of frame on the choice of super-
market coupons. Advances in Consumer Research. Association for Consumer
Research (U.S.), 17, 488–493.
Dickinson, D. L., & McElroy, T. (2010). Rationality around the clock: Sleep and
time-of-day effects on guessing game responses. Economic Letter, 108, 245–248.
Dickinson, D. L., & McElroy, T. (2012). Circadian effects on strategic reasoning.
Experimental Economics, 15, 444–459. doi:10.1007/s10683-011-9307-3
Edwards, W. (1954). The theory of decision making. Psychological Bulletin, 51, 380–
417. doi:10.1037/h0053870
Epstein, S., Lipson, A., Holstein, C., & Huh, E. (1992). Irrational reactions to nega-
tive outcomes: Evidence for two conceptual systems. Journal of Personality and
Social Psychology, 62, 328–339. doi:10.1037/0022-3514.62.2.328
Galvan, A., Hare, T. A., Parra, C. E., Penn, J., Voss, H., Glover, G., & Casey, B. J.
(2006). Earlier development of the accumbens relative to orbitofrontal cortex
might underlie risk taking behavior in adolescents. The Journal of Neuroscience,
26, 6885–6892. doi:10.1523/JNEUROSCI.1062-06.2006
Gamliel, E., & Peer, E. (2006). Positive versus negative framing affects justice judg-
ments. Social Justice Research, 19, 307–322. doi:10.1007/s11211-006-0009-5
Gamliel, E., & Peer, E. (2010). Attribute framing affects the perceived fairness of
allocation principles. Judgment and Decision Making, 5, 11–20.
Gilbert, D. T. (1991). How mental systems believe. American Psychologist, 46, 107–
119. doi:10.1037/0003-066X.46.2.107
Gross, J. J., & Levenson, R. W. (1993). Emotion suppression: Physiology, self-report,
and expressive behavior. Journal of Personality and Social Psychology, 64, 970–
986. doi:10.1037/0022-3514.64.6.970

64       levin et al.

13490-03_Ch02-3rdPgs.indd 64 11/15/13 1:43 PM

 Hare, T. A., & Casey, B. J. (2005). The neurobiology and development of cognitive
and affective control. Cognition, Brain, & Behavior, 3, 273–286.
Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging:
A review and a new view. Psychology of Learning and Motivation, 22, 193–225.
doi:10.1016/S0079-7421(08)60041-9
Hedgcock, W., Denburg, N., Levin, I. P., & Halfmann, K. (2012, November). Why
older adults are impaired on some decision making tasks but not on others—Behavioral
and neuroimaging evidence. Paper presented at the Annual Meeting of the Society
for Judgment and Decision Making, Minneapolis, MN.
Huettel, S. A. (2010). Ten challenges for decision neuroscience. Frontiers in Decision
Neuroscience, 4(171), 1–7. doi:10.3389/fnins.2010.00171
Huettel, S. A., Stowe, C. J., Gordon, E. M., Warner, B. T., & Platt, M. L. (2006).
Neural signatures of economic preferences for risk and ambiguity. Neuron, 49,
765–775. doi:10.1016/j.neuron.2006.01.024
Jasper, J., & Christman, S. (2005). A neuropsychological dimension for anchor-
ing effects. Journal of Behavioral Decision Making, 18, 343–369. doi:10.1002/
bdm.511
Jasper, J. D., Barry, K., & Christman, S. D. (2008). Individual differences in counter­
factual production. Personality and Individual Differences, 45, 488–492. doi:10.1016/
j.paid.2008.05.026
Jasper, J. D., Fournier, C., & Christman, S.D. (2013). Handedness differences in
information framing. Manuscript under review.
Jasper, J. D., Goel, R., Einarson, A., Gallo, M., & Koren, G. (2001). Effects of fram-
ing on teratogenic risk perception in pregnant women. The Lancet, 358, 1237–
1238. doi:10.1016/S0140-6736(01)06353-X
Kahneman, D. (2003). Perspective on judgment and choice: Mapping bounded ratio-
nality. American Psychologist, 58, 697–720. doi:10.1037/0003-066X.58.9.697
Kahneman, D. (2011). Thinking, fast and slow. New York, NY: Farrer, Straus and Giroux.
Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute sub-
stitution in intuitive judgment. In T. Gilovich, D. Griffin, & D. Kahneman
(Eds.), Heuristics & biases: The psychology of intuitive judgment (pp.548–558). New
York, NY: Cambridge University Press. doi:10.1017/CBO9780511808098.004
Kahneman, D., & Tversky, A. (1979). Prospect theory–Analysis of decision under
risk. Econometrica, 47, 263–291. doi:10.2307/1914185
Kovalchik, S., Camerer, C. F., Grether, D. M., Plott, C. R., & Allman, J. M. (2005).
Aging and decision making: A comparison between neurologically healthy
elderly and young individuals. Journal of Economic Behavior & Organization, 58,
79–94. doi:10.1016/j.jebo.2003.12.001
Kühberger, A. (1998). The influence of framing on risky decisions: A meta-analysis.
Organizational Behavior and Human Decision Processes, 75, 23–55. doi:10.1006/
obhd.1998.2781

methods for studying neuroeconomic processes      65

13490-03_Ch02-3rdPgs.indd 65 11/15/13 1:43 PM

 Kühberger, A., & Tanner, C. (2010). Risky choice framing: Task versions and a com-
parison of prospect theory and fuzzy-trace theory. Journal of Behavioral Decision
Making, 23, 314–329. doi:10.1002/bdm.656
Kuhnen, C. M., & Knutson, B. (2005). The neural basis of financial risk taking.
Neuron, 47, 763–770. doi:10.1016/j.neuron.2005.08.008
La Fleur, S. E. (2003). Daily rhythms in glucose metabolism. Suprachiasmatic
nucleus output to peripheral tissue. Journal of Neuroendocrinology, 15, 315–322.
doi:10.1046/j.1365-2826.2003.01019.x
Lauriola, M., & Levin, I. P. (2001). Personality traits and risky decision making in
a controlled experimental task: An exploratory study. Personality and Individual
Differences, 31, 215–226. doi:10.1016/S0191-8869(00)00130-6
LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience,
2, 155–184. doi:10.1146/annurev.neuro.23.1.155
Levin, I. P. (1987). Associative effects of information framing. Bulletin of the Psycho-
nomic Society, 25, 85–86.
Levin, I. P. (1999). Why do you and I make difference decisions? Tracking individual dif-
ferences in decision making. Los Angeles, CA: Presidential Address for Society for
Judgment and Decision Making. doi:10.1037/e683312011-044
Levin, I. P., & Gaeth, G. J. (1988). How consumers are affected by the framing
of attribute information before and after consuming the product. Journal of
Consumer Research, 15, 374–378. doi:10.1086/209174
Levin, I. P., Gaeth, G. J., Schreiber, J., & Lauriola, M. (2002). A new look at fram-
ing effects: Distribution of effect sizes, individual differences, and independence
of types of effects. Organizational Behavior and Human Decision Processes, 88,
411–429. doi:10.1006/obhd.2001.2983
Levin, I. P., Gui, X., Weller, J. A., Reimann, M., Lauriola, M., & Bechara, A.
(2012). A neuropsychological approach to understanding risk-taking for poten-
tial gains and losses. Frontiers in Decision Neuroscience, 6(15). doi:10.3389/
fnins.2012.00015
Levin, I. P., & Hart, S. S. (2003). Risk preferences in young children: Early evidence
of individual differences in reaction to potential gains and losses. Journal of
Behavioral Decision Making, 16, 397–413. doi:10.1002/bdm.453
Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames are not created equal:
A typology and critical analysis of framing effects. Organizational Behavior and
Human Decision Processes, 76, 149–188. doi:10.1006/obhd.1998.2804
Mahoney, K. T., Buboltz, W., Levin, I. P., Doverspike, D., & Svyantek, D. J. (2011).
Individual differences in a within-subjects risky-choice framing study. Personal-
ity and Individual Differences, 51, 248–257. doi:10.1016/j.paid.2010.03.035
Manly, T., Lewis, G. H., Robertson, I. H., Watson, P. C., & Datta, A. K. (2002). Cof-
fee in the cornflakes: Time of day as a modulator of executive response control.
Neuropsychologia, 40, 1–6. doi:10.1016/S0028-3932(01)00086-0

66       levin et al.

13490-03_Ch02-3rdPgs.indd 66 11/15/13 1:43 PM

 Masicampo, E. J., & Baumeister, R. F. (2008). Toward a physiology of dual-process
reasoning and judgment: Lemonade, willpower, and expensive rule-based analy-
sis. Psychological Science, 19, 255–260. doi:10.1111/j.1467-9280.2008.02077.x
McElroy, T., Dickinson, C., Corbin, J., & Beck, H. (2013). Tracking risky decisions:
Comparing fuzzy-trace theory and prospect theory through eye-tracking. Manu-
script submitted for publication.
McElroy, T., & Dickinson, D. L. (2010). Thoughtful days and valenced nights:
How much will you think about the problem? Judgment and Decision Making,
5, 516–523.
McElroy, T., & Seta, J. (2003). Framing effect: An analytic-holistic perspective.
Journal of Experimental Social Psychology, 39, 610–617. doi:10.1016/S0022-1031
(03)00036-2
McElroy, T., & Seta, J. (2004). On the other hand am I rational? Hemispheric activa-
tion and the framing effect. Brain and Cognition, 55, 572–580. doi:10.1016/
j.bandc.2004.04.002
McMahon, A. J., & Scheel, M. H. (2010). Glucose promotes controlled process-
ing: Matching, maximizing, and root beer. Judgment and Decision Making, 5,
450–457.
McNeil, B. J., Pauker, S. G., Sox, H. C., & Tversky, A. (1982). On the elicitation of
preferences for alternative therapies. The New England Journal of Medicine, 306,
1259–1262. doi:10.1056/NEJM198205273062103
Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends
in Cognitive Sciences, 8, 102–107. doi:10.1016/j.tics.2004.01.004
Ornstein, R. E. (1972). The psychology of consciousness. San Francisco, CA: Freeman.
Pardo, J. V., Lee, J. T., Sheikh, S. A., Surerus-Johnson, C., Shah, H., Munch,
K. R., . . . Dysken, M. W. (2007). Where the brain grows old: Decline in ante-
rior cingulate and medial prefrontal function with normal aging. NeuroImage,
35, 1231–1237. doi:10.1016/j.neuroimage.2006.12.044
Parker, A. M., & Fischhoff, B. (2005). Decision-making competence: External vali-
dation through an individual-differences approach. Journal of Behavioral Deci-
sion Making, 18, 1–27. doi:10.1002/bdm.481
Payne, J. W., Sagara, N., Shu, S. B., Appelt, K. C., & Johnson, E. J. (2012). Life
expectancy as a constructed belief: Evidence of a live-to or die-by framing
effect. Columbia Business School Research Paper No. 12-10l. doi:10.2139/
ssrn.1987618
Peters, E., & Levin, I. P. (2008). Dissecting the risky-choice framing effect: Numer-
acy as an individual difference factor in weighting risky and riskless options.
Judgment and Decision Making, 3, 435–448.
Phelps, E. A. (2006). Emotion and cognition: Insights from studies of the human
amygdala. Annual Review of Psychology, 57, 27–53. doi:10.1146/annurev.
psych.56.091103.070234

methods for studying neuroeconomic processes      67

13490-03_Ch02-3rdPgs.indd 67 11/15/13 1:43 PM

 Ralph, M. R., Foster, R. G., Davis, F. C., & Menaker, M. (1990). Transplanted
suprachiasmatic nucleus determines circadian period. Science, 247, 975–978.
doi:10.1126/science.2305266
Resnick, S. M., Pham, D. L., Kraut, M. A., Zonderman, A. B., & Davatzikos, C.
(2003). Longitudinal magnetic resonance imaging studies of older adults: A
shrinking brain. The Journal of Neuroscience, 23, 3295–3301.
Reyna, V. F. (2012). A new intuitionism: Meaning, memory, and development in
fuzzy-trace theory. Judgment and Decision Making, 7, 332–359.
Reyna, V. F., & Brainerd, C. J. (1991). Fuzzy-trace theory and framing effects in
choice: Gist extraction, truncation, and conversion. Journal of Behavioral Deci-
sion Making, 4, 249–262. doi:10.1002/bdm.3960040403
Reyna, V. F., & Brainerd, C. J. (2011). Dual processes in decision making and
developmental neuroscience: A fuzzy-trace model. Developmental Review, 31,
180–206.
Sanfey, A. G., Rilling, J. K., Aronson, J. A., Nystrom, L. E., & Cohen, J. D. (2003).
The neural basis of economic decision-making in the Ultimatum Game. Science,
300, 1755–1758. doi:10.1126/science.1082976
Schmidt, C., Collette, F., Cajochen, C., & Peigneux, P. (2007). A time to think: Cir-
cadian rhythms in human cognition. Cognitive Neuropsychology, 24, 755–789.
doi:10.1080/02643290701754158
Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic information
processing: II. Perceptual learning, automatic attending, and a general theory.
Psychological Review, 84, 127–190. doi:10.1037/0033-295X.84.2.127
Simon, A. F., Fagley, N. S., & Halleran, J. G. (2004). Decision framing: Moderating
effects of individual differences and cognitive processing. Journal of Behavioral
Decision Making, 17, 77–93. doi:10.1002/bdm.463
Sloman, S. A. (1996). The empirical case for two systems of reasoning. Psychological
Bulletin, 119, 3–22. doi:10.1037/0033-2909.119.1.3
Smith, S. M., & Levin, I. P. (1996). Need for cognition and choice framing effects.
Journal of Behavioral Decision Making, 9, 283–290. doi:10.1002/(SICI)
1099-0771(199612)9:4<283::AID-BDM241>3.0.CO;2-7
Stanovich, K. E. (1999). Who is rational? Studies of individual differences in reasoning.
Mahwah, NJ: Erlbaum.
Stanovich, K. E., & West, R. F. (1998a). Cognitive ability and variation in selec-
tion task performance. Thinking & Reasoning, 4, 193–230. doi:10.1080/1354678
98394139
Stanovich, K. E., & West, R. F. (1998b). Individual differences in rational thought.
Journal of Experimental Psychology: General, 127, 161–188. doi:10.1037/0096-
3445.127.2.161
Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Impli-
cations for the rationality debate? Behavioral and Brain Sciences, 23, 645–665.
doi:10.1017/S0140525X00003435

68       levin et al.

13490-03_Ch02-3rdPgs.indd 68 11/15/13 1:43 PM

 Tom, S. M., Fox, C. R., Trepel, C., & Poldrack, R. A. (2007). The neural basis of loss
aversion in decision-making under risk. Science, 315, 515–518. doi:10.1126/
science.1134239
Trepel, C., Fox, C. R., & Poldrack, R. A. (2005). Prospect theory on the brain? Toward
a cognitive neuroscience of decisions under risk. Cognitive Brain Research, 23,
34–50. doi:10.1016/j.cogbrainres.2005.01.016
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology
of choice. Science, 211, 453–458. doi:10.1126/science.7455683
Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative
representation of uncertainty. Journal of Risk and Uncertainty, 5, 297–323.
doi:10.1007/BF00122574
Wang, X. T., & Dvorak, R. D. (2010). Sweet future: Fluctuating blood glucose lev-
els affect future discounting. Psychological Science, 21, 183–188. doi:10.1177/
0956797609358096
Weller, J. A., Levin, I. P., & Denburg, N. L. (2011). Trajectory of risky decision mak-
ing for potential gains and losses from ages 5 to 85. Journal of Behavioral Decision
Making, 24, 331–344. doi:10.1002/bdm.690
Weller, J. A., Levin, I. P., Rose, J., & Bossard, E. (2012). Assessment of decision-
making competence in preadolescence. Journal of Behavioral Decision Making,
25, 414–426. doi:10.1002/bdm.744
Weller, J. A., Levin, I. P., Shiv, B., & Bechara, A. (2007). Neural correlates of adap-
tive decision making in risky gains and losses. Psychological Science, 18, 958–
964. doi:10.1111/j.1467-9280.2007.02009.x
Weller, J. A., Levin, I. P., Shiv, B., & Bechara, A. (2009). The effects of insula dam-
age on decision-making for risky gains and losses. Social Neuroscience, 4, 347–
358. doi:10.1080/17470910902934400
West, R. (2000). In defense of the frontal lobe hypothesis of cognitive aging. Jour-
nal of the International Neuropsychological Society, 6, 727–729. doi:10.1017/
S1355617700666109
West, R., Murphy, K. J., Armillo, M. L., Craik, F. I. M., & Stuss, D. (2002). Lapses
of intention and performance variability reveal age-related increases in fluc-
tuations of executive control. Brain and Cognition, 49, 402–419. doi:10.1006/
brcg.2001.1507
West, R. L. (1996). An application of prefrontal cortex function theory to cognitive
aging. Psychological Bulletin, 120, 272–292. doi:10.1037/0033-2909.120.2.272
Wood, S., Busemeyer, J., Kolings, A., Cox, C. R., & Davis, H. (2005). Older adults
as adaptive decision makers: Evidence from the Iowa Gambling Task. Psychology
and Aging, 20, 220–225. doi:10.1037/0882-7974.20.2.220

methods for studying neuroeconomic processes      69

13490-03_Ch02-3rdPgs.indd 69 11/15/13 1:43 PM

13490-03_Ch02-3rdPgs.indd 70 11/15/13 1:43 PM
 II
Neurodevelopment

13490-04_PT2_Ch03-3rdPgs.indd 71 11/15/13 1:43 PM

13490-04_PT2_Ch03-3rdPgs.indd 72 11/15/13 1:43 PM
 3
Risks, Rewards, and
the Developing Brain in
Childhood and Adolescence
Barbara R. Braams, Linda van Leijenhorst,
and Eveline A. Crone

Childhood and adolescence are a time of changes in the physical, cognitive, and
social–emotional domains. Behaviorally, one of the prominent features of ado-
lescence is an increase in risk taking. In this chapter, we review current theories
and research to explain risk-taking behavior from a neural perspective. After
a general introduction, we lay out behavioral findings focusing on risk taking,
then describe current models of adolescent brain development that provide pos-
sible explanations for the observed risk taking. Next, we describe neuroimaging
research and how the findings map to the models. Finally, we propose new
directions for future research.
Adolescence, defined as the transition phase between childhood and adult-
hood, is a time of many physical, cognitive, and social–emotional changes. It
is a natural time of exploring, thrill seeking, and eventually setting long-term
goals and aspirations (Dahl, 2004; Steinberg, 2008b). The first phase of ado-
lescence (also defined as early to middle adolescence) starts with the onset

http://dx.doi.org/10.1037/14322-004
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

73

13490-04_PT2_Ch03-3rdPgs.indd 73 11/15/13 1:43 PM

 of pubertal maturation around the age of 10 to 11 years (but approximately
1.5 years earlier for girls than for boys) and lasts until approximately ages 15
to 16 (Shirtcliff, Dahl, & Pollak, 2009). At the onset of puberty, a dramatic
increase in the secretion of adrenal androgens, gonadal steroids, and growth
hormone causes many changes in physical appearance (e.g., facial and physi-
cal changes) and in brain regions with high receptor density for gonadal hor-
mones such as testosterone and estradiol (Scherf, Berman, & Dahl, 2012).
Following pubertal maturation (Op de Macks et al., 2011), the second phase
of adolescence (also defined as middle to late adolescence) lasts from approxi-
mately ages 15–16 to 21–22 years, during which adolescents gradually reach
independence from parents and obtain mature social goals (Steinberg, 2008b;
Steinberg & Morris, 2001).

Why Do Adolescents Take Risks?

One of the most prominent findings in observational studies (i.e., cor-

relational studies) is that adolescents take more risks than children or adults
(Beyth-Marom & Fischhoff, 1997; Boyer, 2006). Risks are defined as engage-
ment in behaviors that are associated with potentially negative outcomes. For
example, adolescents are more likely to have casual sexual partners, engage
in binge drinking, and get into car accidents, and they seem to act without
thinking about the long-term consequences of their actions (Eaton et al.,
2008; Steinberg, 2008b). In a striking juxtaposition, although adolescents are
physically in the best condition of their lives, mortality rates go up by 200%
to 300% compared with children (Dahl, 2004). This mortality is primar-
ily due to preventable causes, such as getting involved in accidents; driving
under the influence; and engaging in high-risk behaviors under pressure from
peers or when with friends, such as driving with friends encourages reckless
driving (presumably to show off and impress peers, but the mechanisms are
not fully understood; see Steinberg, 2008b).
Yet, as we describe in the next section of this chapter, the behavior
observed in the laboratory does not always support the observation of increased
risk taking in adolescents’ daily lives. Given this, it is perhaps not surpris-
ing that programs that teach adolescents to engage in less risky behaviors or
teach them the consequences of their actions have doubtful efficacy in the
reduction of risk taking (but see Fischhoff, Bruine de Bruin, Parker, Millstein,
& Halpern-Felsher, 2010).
We propose that the deficits in risky decision making observed dur-
ing adolescence are most likely related not only to cognitive immaturities
but also to a combination of cognitive, emotional, and social factors (Crone
& Dahl, 2012; Steinberg, 2008b). Understanding these factors is of great

74       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 74 11/15/13 1:43 PM

 importance in eventually changing behavior and preventing adverse conse-
quences of risk taking in the future. There is a great need for an integrative
approach focusing on contextual influences on behavior, brain maturation,
hormonal investigation, and genetic susceptibility to understand this crucial
phase in life.
In the present chapter, we argue for such an integrative approach. We
start by reviewing the empirical evidence, both from daily, naturalistic set-
tings and controlled laboratory settings, regarding whether adolescents show
greater risky behaviors. We next show that approaching the problem from
a cognitive neuroscience perspective is a promising way to advance under-
standing of when adolescents take risks, why they take risks, and who is at
risk. We then focus specifically on the neural responses to rewards, covering
both reward anticipation and reward processing. These processes seem to
show differential developmental trajectories in terms of neural activity, illu-
minating that brain imaging informs risk-taking understanding in a way that
has additive value above behavioral observations only. Finally, we lay out
future directions and conclude with a new working model for studying the
dynamics of adolescent risk taking.

Evidence for Increases in Risk Taking in Adolescence

The ability to identify and avoid immediate and long-term undesirable

consequences of actions and avoid excessive risk is one of the key aspects
of mature decision making, and its development has been studied by devel-
opmental psychologists and developmental cognitive neuroscientists using
various approaches. Developmental neuroscience research is influenced most
by studies using a cognitive perspective, which tries to explain risk-taking
behavior in adolescence by examining the development of decision-making
skills. In these studies, risk-taking behavior is defined as the consequence of
immature decision-making abilities. Although the epidemiological support
for a peak in risky behavior in adolescence (Dahl, 2004; Eaton et al., 2008;
Steinberg, 2008a) seems clear, it has been difficult to capture in laboratory
research, which poses a challenge for neuroscience studies that take place in
a laboratory context.
In most cognitive theories of decision making, decisions are seen as the
result of an evaluative process in which both the value of an outcome and
the probability with which it will be attained are considered. These concepts
are described in terms of the expected value (EV) of choice alternatives. The
EV is defined as the multiplication of the value of a possible outcome and
the likelihood that that outcome will be obtained. Mature, rational deci-
sions should favor the choice options with the higher EV, and decisions can

risks, rewards, and the developing brain      75

13490-04_PT2_Ch03-3rdPgs.indd 75 11/15/13 1:43 PM

 be defined as risky if these rules are not followed. Adolescents’ risky behav-
ior suggests that the evaluative process underlying decisions is immature in
this developmental period, and the curvilinear pattern with a peak in risky
behavior specific to adolescence suggests that decision-making skills do not
improve linearly with development.
However, numerous behavioral experimental studies have found lin-
ear patterns of change with development. That is, in several studies chil-
dren’s decisions were more risky than were adolescents’ and adults’ decisions.
Similarly, the ability to take the long-term consequences of choices into
account improves from childhood until late adolescence. In addition, several
studies suggest that key components of decision making are relatively mature
in adolescence. For example, in our lab we used the cake gambling task, in
which the demands on learning and working memory are minimized by mak-
ing outcome values and associated probabilities explicit (van Leijenhorst,
Westenberg, & Crone, 2008). Decision-making behavior did not differ from
age 8 to 30 on this task, which suggests that from late childhood on, partici-
pants are able to take both reward and probability information into account
when making decisions. Other developmental studies have shown that chil-
dren as young as 4 years already show these kinds of decision-making skills
(Anderson, 1991, 1996; Schlottmann, 2001; Schlottmann & Anderson,
1994). This suggests that decision-making skills improve with age but that
the foundation is laid relatively early in childhood. These findings lead
to the question of why the developmental trajectory of risky behaviors as
identified in epidemiological work is curvilinear, whereas the developmen-
tal trajectory as identified with many lab-based measures suggests that ado-
lescents’ decision making is relatively mature and comparable with those
of adults. How can adolescent-specific risk-taking behaviors be understood
using laboratory tasks?
One approach has been to connect laboratory-based findings to real-life
risk taking or personality constructs linked to real-life risk taking by assess-
ing both in the same participants. In the previously mentioned study by van
Leijenhorst et al. (2008), decision making did not differ between age groups.
Yet, there were consistent differences within groups; participants scoring high
on a self-report measure of sensation seeking, which is associated with a vul-
nerability for problematic behaviors (e.g., Horvath & Zuckerman, 1993),
showed more risky behavior in the experiment. Similarly, a recent study by
Reyna et al. (2011) found individual differences; a laboratory measure of
(gist-based) reasoning and reward sensitivity (e.g., sensation seeking) was
found to independently predict real-life risky sexual behavior in adolescent
participants. These findings support the ecological validity of theory-driven
laboratory-based measures that assess causal mechanisms underlying risky
decision making.

76       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 76 11/15/13 1:43 PM

 A second direction of research that assists in resolving the apparent
discrepancy between the epidemiological findings and the lab-based findings
has been to develop lab-based measures of adolescents’ decision-making skills
that more closely mimic the complex context in which decisions are made
outside of the laboratory. One aspect of context that is gaining attention is
whether the choice context is affective (“hot”) or neutral (“cool”; Metcalfe
& Mischel, 1999). Recent studies have taken this dual process (hot vs. cool)
approach to studying adolescents’ decision-making skills and have found
interesting results. For example, Figner, Mackinlay, Wilkening, and Weber
(2009) examined the effects of choice context on differences in risk-taking
behavior between adolescents age 13 to 16 years and adults in terms of risk
preferences and the amount of information that is taken into account when
decisions are made. This study used two versions of a card gambling paradigm.
One version was emotionally arousing and the other was emotionally neutral.
The arousing version was hypothesized to make more demands on cognitive
resources and as a consequence would result in an immature and risky deci-
sion strategy. In contrast, the neutral version was hypothesized to result in a
deliberative, reasoning-based strategy. Indeed, adolescents were found to take
more risks and base their decisions on less information in the emotionally
arousing condition compared with the neutral condition, in which their per-
formance was similar to that of adults. Comparable results were found by van
Duijvenvoorde, Jansen, Visser, and Huizenga (2010), who contrasted 13- to
15-year-olds’ behavior in an affective decision-making task, the hungry don-
key task, a child-adapted version of the Iowa gambling task (Crone & van der
Molen, 2004) in which participants receive reward and loss feedback on each
trial, to a neutral decision-making task, the gambling machine task, in which
no performance feedback is given. The two tasks involve a similar compari-
son of choice dimensions and require equally complex reasoning capacities.
Adolescents made suboptimal EV decisions in the affective compared with
the neutral task. The authors explain this in terms of the complexity of the
reasoning that adolescents used, which was reduced in the affective context
in which adolescents focused on only one of three relevant choice dimen-
sions. Finally, Burnett, Bault, Coricelli, and Blakemore (2010) examined risk
preferences in 9- to 35-year-old males using a paired gambles task in which
choice alternatives differed in the degree of risk involved. While the ability
to make optimal decisions in terms of EV showed a linear increase with age,
the number of risky gambles peaked around age 14. The findings from these
studies suggest that an affective or hot context more closely mimics the com-
plex context in which decisions are made outside of the laboratory and that
adolescents are particularly influenced by reward and loss information.
Taken together, a body of research supports the ecological validity
of lab-based measures of risk taking. This work has shown increases in

risks, rewards, and the developing brain      77

13490-04_PT2_Ch03-3rdPgs.indd 77 11/15/13 1:43 PM

 decision-making abilities with development, and importantly, individual
differences in performance on these lab-based measures of decision-making
ability are related to individual differences in risk-taking behavior outside the
lab. Moreover, a second line of research that has focused on decision making
in affective contexts has shown the curvilinear trend that is characteristic of
the development trajectory of risk taking identified in epidemiological work.
Collectively, these results suggest that the difference between adolescents
and adults in decision making is relatively weak or absent when assessed with
neutral tasks, consistent with the idea that decision-making skills in neutral
contexts are relatively intact in adolescence. However, individuals do vary
in reasoning abilities, and these neutral tasks appear to capture such within-
group individual differences. The latter findings indicate that age differences
are more likely to be observed in hot, affectively charged decision-making
situations. These results highlight the need for a more integrative approach,
which takes cognitive, emotional, and social context factors into account,
as well as the need for a more mechanistic understanding of risk taking in
adolescence. A better understanding of the mechanisms underlying decision
making and their change with development can help explain the interactions
of the development of cognitive abilities and contextual factors, which will
help understand risky behavior in adolescence. A promising approach in this
regard is to study components of risk taking (e.g., reward sensitivity, inhibi-
tion) from a cognitive neuroscience perspective, an approach to which we
turn in the next section.

Neurobiological Models of Adolescent Risk Taking:

Promises and Challenges

From a neuroscientific perspective, several partly overlapping and partly

complementary theoretical models have been proposed about how trajec-
tories of brain maturation may explain behavioral changes in adolescence.
Even though the models have broader goals than solely understanding risk
taking in adolescence, they provide intriguing hypotheses with respect to
the question of why adolescence may be a vulnerable period for risk taking.
The model that is probably most explicit in hypothesizing about risk
taking in adolescence is the dual processing model postulated by Somerville,
Jones, and Casey (2010). According to this model, there is an imbalance
between the development of subcortical brain regions, such as the ventral
striatum and the amygdala, which are thought to precede the development
of the prefrontal cortex. Among the subcortical areas, the ventral striatum
is involved with reward processing and decision making based on incentives
(Delgado, 2007), whereas the amygdala is involved with processing highly

78       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 78 11/15/13 1:43 PM

 salient and motivational emotional stimuli (Cunningham, Arbuckle, Jahn,
Mowrer, & Abduljalil, 2010), such as potential threats and emotional states
of others. Also, the role of the amygdala in fear and anxiety has been well
established (LeDoux, 2007). The prefrontal cortex is involved with tasks
such as planning, inhibition, and cognitive control more generally (Miller
& Cohen, 2001). The interplay between a not fully developed prefrontal
cortex, which is therefore not fully capable of executing control-related func-
tions, and further developed emotional areas such as the amygdala and the
ventral striatum causes an imbalance resulting in a primarily emotion-driven
approach to rewards and risks in middle adolescence.
A comparable model is the triadic model proposed by Ernst and colleagues
(Ernst, Pine, & Hardin, 2006; Ernst & Fudge, 2009). The triadic model is based
on the interplay between three systems. The first is the approach system, tai-
lored to rewards; the second the avoidance system, tailored to prevent harm;
and the last is the regulatory system. The neural correlates underlying these sys-
tems are the ventral striatum, the amygdala, and the prefrontal cortex, respec-
tively. These three systems interact with each other and, when balanced, work
together in a beneficial way facilitating learning and harm avoidance. However,
during adolescence the balance often tips toward the approach system, result-
ing in reward-driven behavior. The model is comparable to the Somerville et al.
(2010) dual processing model, although it differs with respect to the assump-
tions about speed of maturation. Where the dual processing approach predicts
that the subcortical areas develop earlier than the prefrontal areas, the triadic
model does not make assumptions about earlier or later maturation. Instead,
this model suggests that there is a fragile balance, which may result in tipping
to approach or avoidance more easily in adolescence.
Finally, the third model that is relevant for our discussion is the social
information processing model proposed by Nelson, Leibenluft, McClure, and
Pine (2005). During adolescence, social reorientation takes place. This model
provides a broad explanation for changes in social processing (social percep-
tion, social emotion, and social cognition) in adolescence but also makes pre-
dictions about brain regions involved in risk taking. The social information
processing network distinguishes three nodes: a detection node, an affective
node, and a cognitive-regulation node (Nelson et al., 2005). Although differ-
ent information is being processed in the different nodes, they are connected
and influence each other. In this model, the amygdala and ventral striatum are
considered to be part of the affective node, and the prefrontal cortex is part
of the cognitive-regulatory node, mirroring the earlier described models. The
detection node is involved in basic processing of social stimuli and includes
the fusiform face area, superior temporal sulcus, and anterior temporal cor-
tex. The social information processing model differs from the earlier described
models in that the developmental pattern seen in risk taking is ascribed to

risks, rewards, and the developing brain      79

13490-04_PT2_Ch03-3rdPgs.indd 79 11/15/13 1:43 PM

 hormonally induced changes in the affective node. Specifically, Nelson et al.
(2005) propose that activation in the ventral striatum is elevated due to
the rise in gonadal hormones at the onset of puberty, which causes a fragile
balance with the slowly developing prefrontal cortex. Therefore, this model
does not necessarily predict that the structural development of the ventral
striatum precedes structural development of the prefrontal cortex but instead
that the ventral striatum is more sensitive to specific pubertal changes and
therefore becomes more active (see also Steinberg et al., 2008, for a similar
dual processing hypothesis based on behavioral empirical evidence).
Together, these models provide excellent starting points for a more in-
depth analysis of neural patterns related to risk taking in adolescence, and as
we will see below, they each have received empirical support. Yet, as we will
also see in the next section, the models are not very specific with respect to
the subprocesses that are involved in risk taking and how the striatum and
prefrontal cortex are expected to respond to risks versus rewards.

Putting the Models to the Test: How Does Brain

Imaging Inform Risk-Taking Research in Adolescence?

The empirical studies available to date have mostly used functional

magnetic resonance imaging (fMRI), a safe and noninvasive technique that
allows for the study of neural activation during specific phases of a task. These
studies have focused on two task-related processes that are important for risk-
taking investigation: reward anticipation and responses to receiving a reward.
In such neuroimaging studies, responses to anticipated and received
rewards are usually investigated using tasks in which participants can receive
an outcome that is favorable to them. Outcomes can be points, money, or
primary rewards such as juice (Delgado, 2007). A consistent finding in these
studies is involvement of the ventral striatum/nucleus accumbens in response
to receipt of rewards. In addition, the amygdala, anterior cingulate cortex,
dorsolateral prefrontal cortex, ventromedial prefrontal cortex, and orbito-
frontal cortex are found to be active, although these findings are not consis-
tent across studies and possibly depend on specific task demands. Here we will
focus on the developmental pattern in the ventral striatum specifically, as this
region is most consistently activated in response to reward anticipation and
reward processing.

Ventral Striatum Responses to Reward

In reward studies both under- and overrecruitment of the ventral stria-

tum have been reported in adolescents compared with children and adults.

80       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 80 11/15/13 1:43 PM

 The most prominent finding is an overrecruitment of the ventral striatum in
response to reward receipt for adolescents. These results have been found with
different tasks and in a wide age range that included participants between 7
and 40 years of age, which strengthened the credibility of the results. Four of
these studies used tasks that involved a passive or active aspect of gambling.
With passive gambling, we refer to paradigms in which the participant can-
not influence the outcome, whereas an active task means that the participant
can choose to take a risk or not. Two studies with passive gambling tasks
support the hypothesis of overrecruitment of the ventral striatum to reward
in adolescence. Van Leijenhorst, Zanolie, et al. (2010) used a task in which
participants are shown a slot machine that they can start with a button press.
The slots can fill with three types of fruit. They win only when all three slots
show the same fruit. Winning was associated with increased activation in the
ventral striatum, and more so for middle adolescents. Similarly, Galvan et al.
(2006) used a delayed response two-choice task. In this task, participants
are presented with a cue, after which they need to respond by indicating
the location of the cue when prompted. Correct responses within the time
interval set for the response are rewarded. Each cue is paired with a distinct
reward amount. Again, winning was associated with increased activation in
the ventral striatum, and more so for middle adolescents.
Tasks with an active aspect of gambling, such as the wheel of for-
tune task (Ernst et al., 2005) and the cake gambling task (van Leijenhorst,
Gunther Moor, et al., 2010), similarly show greater activation of the ventral
striatum in response to reward receipt. In the wheel of fortune task, partici-
pants are shown a circle divided into two colors. Only trials were analyzed in
which both colors covered 50% of the circle in this case, but divisions and
therefore probabilities of winning can differ. Participants choose one of
the colors, and if the computer randomly picks the same color, they win. The
cake gambling task is a slightly modified version of the wheel of fortune task.
To make the task more suitable for children, the wheel is explained as a cake
with different flavors. Participants can choose which flavor they would like
to bet on, and if the computer picks the same flavor, they win. Participants
can choose between a low-risk gamble with a 66% chance of winning 1 euro
and a high-risk gamble with a 33% chance of winning 2, 4, 6, or 8 euros. In
these tasks, winning was associated with increased activation in the ventral
striatum, and more so in middle adolescence. This pattern of elevated ven-
tral striatum response in middle adolescence was further confirmed in other
paradigms that have used reward conditions, such as a temporal discounting
task (Christakou, Brammer, & Rubia, 2011) and an antisaccade task (Geier,
Terwilliger, Teslovich, Velanova, & Luna, 2010; Padmanabhan, Geier, Ordaz,
Teslovich, & Luna, 2011). Smith, Halari, Giampetro, Brammer, and Rubia
(2011) used a sustained attention task in which adolescents showed elevated

risks, rewards, and the developing brain      81

13490-04_PT2_Ch03-3rdPgs.indd 81 11/15/13 1:43 PM

 ventral striatum responses to rewarded trials compared with adults, but not
compared with the youngest group. The results from this set of studies seem to
provide consistent evidence for enhanced activation in the ventral striatum
in middle adolescence.
However, in work focusing on the anticipation of rewards, the results
have in some studies been opposite to the pattern described above. Two stud-
ies have found underrecruitment of the striatum in middle adolescence. Both
of these studies used the monetary incentive delay (MID) task (Bjork et al.,
2004; Bjork, Smith, Chen, & Hommer, 2010). In the MID task, participants
can win (or avoid losing) different amounts of money by pressing a button
within a short interval after a target is presented. Failure to press within the
response interval results in omission of gain (or loss), whereas pressing the
button within the response interval results in winning (or avoidance of loss).
Participants win in 66% of the trials. Individual response intervals are deter-
mined on the basis of a response time task, assessed before the start of the MID
task. The MID task has two distinguished phases, separated in time to allow
for fMRI analyses for both phases. As such, it has been specifically designed
to enable testing for differences in brain activation during anticipation, as
well as receipt, of a reward. Both studies (Bjork et al., 2004, 2010) found
underrecruitment of the right ventral striatum for adolescents, compared with
an adult group, during gain versus nongain anticipation. Activation during
reward receipt, in contrast, did not yield any significant differences between
the groups. The MID task allows for the possibility that groups engage in
differential strategies. For example, whether the participant will receive the
outcome is dependent on behavior, pressing a button as fast as possible, and
there is a 66% reward scheme. For some participants, this may lead to greater
certainty about the anticipated reward (knowing that they can influence the
outcome by behavior), whereas for others, this may make the task more uncer-
tain (learning that the reward is only given on 66% of the trials). Whether and
how these tasks differ should be an avenue for future research, but it is clear
that intriguing shifts in response patterns may occur depending on how the
task is framed (Bjork, Lynne-Landsman, Sirocco, & Boyce, 2012).
Based on the studies described above, two general phases can be
distinguished—the anticipation of reward and the receipt of the reward
(feedback although others have also distinguished between other phases,
such as cue-related response and anticipation of outcome, see Geier et al.,
2010). The apparent inconsistency of results might stem from confounding
the anticipation of reward and receipt of reward. Taken together, and con-
sistent with the dual processing model (Somerville et al., 2010) as well as
the triadic model (Ernst & Fudge, 2009), the studies presented show that
the ventral striatum is most likely differentially involved in adolescents,
and most studies point toward enhanced activity, at least for reward receipt

82       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 82 11/15/13 1:43 PM

 (see Color Plate 1 for a meta-analysis; Christakou et al., 2011; Ernst et al.,
2005; Galvan et al., 2006; van Leijenhorst, Gunther Moor, et al., 2010; van
Leijenhorst, Zanolie, et al., 2010). No studies found an underrecruitment of
the ventral striatum during reward receipt. However, during anticipation,
findings are more mixed. Studies find overrecruitment (Galvan et al., 2006;
van Leijenhorst, Gunther Moor, et al., 2010) as well as underrecruitment
(Bjork et al., 2004, 2010). Although the models proposed by Somerville et al.
(2010) and Ernst et al. (2006) provide a framework by which some of the
results can be explained, there may be other factors mediating the effects.
Since adolescents go through major changes in hormonal levels as well as in
the social domain, this leads to the hypothesis that emotional and social con-
textual factors may influence these developmental sensitivities, as proposed
by the social information processing model (Nelson et al., 2005).

Peers and Pubertal Hormones

The social influences on neural correlates of risk taking have hardly

been studied, even though it is well known that peers strongly influence the
behavior of adolescents. As already pointed out by the social information
processing model (Nelson et al., 2005), adolescence is a time when peers
become very important (Dahl, 2004). Also, much risk-taking behavior takes
place in groups (Steinberg, 2004). Adolescents can assess risks and can rea-
son logically when they are asked about risks or consequences (Boyer, 2006).
When adolescents are asked to perform a task in the lab, they are usually
alone or in the presence of an experimenter, not among their peers. The
risk-taking behavior that is observed in a natural setting may therefore not
be captured in the standard laboratory environment.
One intriguing new fMRI study investigated risk taking in a setting with
peers for three age groups: adolescents, ages 14 to 18; young adults ages 19 to
22, and adults ages 24 to 29. In this study by Chein, Albert, O’Brien, Uckert,
and Steinberg (2011; see also Gardner & Steinberg, 2005), participants were
asked to play a game in which they were driving a simulated car. Their goal
was to drive from A to B as fast as possible. On the way, they encountered
several crossings. On some of these crossings a stoplight turned orange just
when they approached. On each crossing they could decide to (continue to)
go and take a risk, or stop. When they decided to take a risk, this could result
in a crash, setting them back further than deciding to wait would have. In one
condition, participants were told that they were performing this task alone
and that no one would see what they did. In the other condition, however,
their peers were physically present and watching them. All groups showed
elevated risk-taking in the peer-present condition, but adolescents took even
more risks than young adults and adults. Intriguingly, adolescents also showed

risks, rewards, and the developing brain      83

13490-04_PT2_Ch03-3rdPgs.indd 83 11/15/13 1:43 PM

 elevated ventral striatum responses in the peer-present condition compared
with the other age groups (see Color Plate 2).
Peers are assumed to have a large influence on the behavior of adoles-
cents in real life. One mediating factor, which may provide a mechanistic
explanation for peer sensitivity, is the role of specific pubertal hormones.
These hormones may not only trigger reproductive development but also
influence social behaviors and interest in peers, such as sensation seeking,
seeking of social status, fear of social rejection, and a drive toward social
acceptance (Dahl, 2004; Nelson et al., 2005). Preliminary evidence for this
assumed role of pubertal hormones in adolescent decision making and reward
processing comes from two studies, which investigated the role of testoster-
one on reward processing and risk anticipation (Hermans et al., 2010; van
Honk et al., 2004).
The influence of testosterone on risk taking has been studied mainly
through the measurement of naturally fluctuating hormone levels, but recently,
exciting new advances in testosterone research have focused on the administra-
tion of testosterone in healthy (mostly female) adult participants. Administering
small doses of testosterone resulted in increased risk taking on the Iowa gam-
bling task (van Honk et al., 2004). In addition, administration of the same
dose of testosterone in females results in enhanced activation in the nucleus
accumbens during reward anticipation in the MID task (Hermans, et al., 2010).
It is well known that in puberty the increase in testosterone is orders of mag-
nitude larger than the increase due to administration of testosterone in the
examples described above. Therefore, an intriguing question is whether hor-
mone level increases are predictive of neural responses to risk and reward in
pubertal adolescents.
Two studies used this approach. Op de Macks et al. (2011) found that
boys and girls with higher natural testosterone levels also showed higher ven-
tral striatum responses to reward receipt, thereby confirming prior studies
that showed that testosterone administration in adults also leads to enhanced
ventral striatal responses to reward (Hermans et al., 2010; see Color Plate 3).
However, as described before, responses to reward anticipation and reward
receipt may rely on different mechanisms. This is highlighted in a study by
Forbes et al. (2010) in which testosterone levels correlated positively with
striatum responses in boys during anticipation but correlated negatively
with striatum responses to reward receipt in both boys and girls. Taken
together, these findings shed light on the link between social influence,
hormones, and decision making as described in the social information pro-
cessing model (Nelson et al., 2005). However, future research is necessary
to unravel these apparently inconsistent findings and integrate different
approached to understand the interaction between peer sensitivity and
pubertal hormones.

84       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 84 11/15/13 1:43 PM

 Future Directions

Major changes take place in the body and brain of adolescents. An

increase in risk-taking behavior is evident in daily life. Findings from the
laboratory show that greater risk taking among adolescents, compared with
other age groups, is only evident under specific contextually arousing con-
ditions (high rewards, peer presence; Burnett et al., 2010; Figner et al.,
2009; Gardner & Steinberg, 2005; van Duijvenvoorde et al., 2010). Brain
research investigating the underlying neural correlates of risk-taking behav-
ior in adolescence has revealed partly consistent (elevated neural response
to reward outcome) and partly inconsistent results (reduced and elevated
neural responses to reward anticipation). To understand risk taking more
fully, future researchers should continue to disentangle neural responses to
anticipation and receipt of reward and to focus on sensitivities to social and
emotional contexts (see also Reyna & Farley, 2006; Rivers, Reyna, & Mills,
2008). We propose that the new advances in brain imaging research, inte-
grated with different approaches, take us to another level of explanation of
neural responses to risks and rewards in adolescence. Three of these advances
are highlighted here.
First, very few studies until now have investigated testosterone levels
and the influence on risk taking and reward processing. Bodily changes during
puberty are primarily driven by hormonal changes, and therefore the influ-
ence of hormones is an important area to explore in future work. Testosterone
administration studies in adult women show a relationship between nucleus
accumbens activity as well as increased risk taking (Hermans et al., 2010;
van Honk et al., 2004). Individual measures of hormone levels as assessed
by, for instance, morning saliva samples provide valuable information above
and beyond the effects of puberty (Op de Macks et al., 2011). Reliably inves-
tigating individual differences calls for large sample sizes, preferably spread
over a wide age range. Participants who have not yet reached puberty, as well
as midpubertal and postpubertal participants, should be included in order
to unravel puberty effects independent of age changes (Quevedo, Benning,
Gunnar, & Dahl, 2009). To track the developmental pattern, a longitudinal
design would be optimal in the investigation of the combined influence of
early or late puberty onset and how this interacts with social environmental
influences.
Second, it is well known that most risk-taking behavior takes place
when adolescents are in the presence of peers, yet social factors influencing
neural responses to risks and rewards have hardly been studied. Only one
study investigated the effects of peer presence on brain activity (Chein et al.,
2011). Especially for adolescents, the presence of peers was found to have a
significantly elevating effect on striatum responses. Due to the confined space

risks, rewards, and the developing brain      85

13490-04_PT2_Ch03-3rdPgs.indd 85 11/15/13 1:43 PM

 and limitations provided by the MRI scanner, it is challenging to manipulate
peer presence within an MRI setting. Two possibilities, other than directly
manipulating actual peer presence, are to investigate the influence of social
stimuli on risk taking and the relationship between real-life risk taking and
neural responses. Social stimuli could, for instance, be status within the peer
group or the influence of emotional faces on risk taking (Casey et al., 2011).
Similar to the approach described earlier in behavioral research, real-life
risk taking could be measured by the use of illicit substances, such as alcohol
or marijuana, or engagement in sexual behavior. Yet another possibility
would be to investigate attitudes toward risk taking, for instance, how likely
someone thinks it is that a certain risk will yield a positive outcome. For
example, Galvan, Hare, Voss, Glover, and Casey (2007) reported a signifi-
cant positive correlation between the self-reported likelihood of engaging
in risky behavior and nucleus accumbens reactivity in response to winning.
Similar correlations can possibly be found between social relations in real
life (e.g., resistance to peer influence) and neural responses to risks and
rewards in the laboratory (see also Paus, Keshavan, & Giedd, 2008, for
a similar approach), which provide us with more information about the
underlying mechanism of risk taking at the neural level and disentangles
who is at risk of showing exaggerated risk taking in real life.
Third, integrating different theoretical approaches as described in the
literature so far, such as affective and social neuroscience, can help unravel risk-
taking behavior. For example, besides numerous behavioral models, only the
neuroscientific model by Nelson et al. (2005) explicitly takes social informa-
tion processing into account, but the model is not yet specific with respect to
how developmental and individual differences in theory of mind or resistance
to peer influence may interact with individual tendencies toward risk taking.
Recently, Blakemore and colleagues (Blakemore, 2008; Burnett, Thompson,
Bird, & Blakemore, 2011) have demonstrated that adolescence is character-
ized by large changes in the brain networks important for mentalizing about
intentions of others. Especially regions in the social brain network, such as
the medial prefrontal cortex, precuneus, temporal parietal junction, and
insula, were shown to develop both functionally and structurally during ado-
lescence (Blakemore, 2008; Gogtay et al., 2004). Thus, it is of great importance
to examine neural responses to rewards not in isolation, but in an integrative
approach with interaction between affective neuroscience (neural responses to
risks and rewards) and social neuroscience (ability to take perspective, under-
stand intentions, and resist peer influence).
The approach for the future is therefore to study risk taking and neural
structure and function in relation to pubertal development, sensitivity ver-
sus resistance to peer influences, and mentalizing about intentions of oth-
ers. Ultimately, we would like to be able to assess individual differences to

86       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 86 11/15/13 1:43 PM

 enable identification of adolescents who are most likely to exhibit excessive
risk-taking behavior. This knowledge has far-reaching implications for policy
questions and for developing interventions that are specific for adolescents
who need them, focusing on developmentally specific sensitivities.

References

Anderson, N. H. (1991). Probability development. In N. H. Anderson (Ed.), Con-

tributions to information integration theory: Vol. III. Developmental (pp. 83–108).
Hillsdale, NJ: Erlbaum.
Anderson, N. H. (1996). A functional theory of cognition. Mahway, NJ: Erlbaum.
Beyth-Marom, R., & Fischhoff, B. (1997). Adolescents’ decisions about risks: A cog-
nitive perspective. In J. Schulenberg & J. L. Maggs (Eds.), Health risks and devel-
opmental transitions in adolescence (pp. 110–135). New York, NY: Cambridge
University Press.
Bjork, J. M., Knutson, B., Fong, G. W., Caggiano, D. M., Bennett, S. M., & Hommer,
D. W. (2004). Incentive-elicited brain activation in adolescents: Similarities
and differences from young adults. The Journal of Neuroscience, 24, 1793–1802.
doi:10.1523/JNEUROSCI.4862-03.2004
Bjork, J. M., Lynne-Landsman, S. D., Sirocco, K., & Boyce, C. A. (2012). Brain matu-
ration and risky behavior: The promise and the challenges of neuroimaging-based
accounts. Child Development Perspectives, 6, 385–391. doi:10.1111/cdep.12001
Bjork, J. M., Smith, A. R., Chen, G., & Hommer, D. W. (2010). Adolescents, adults
and rewards: Comparing motivational neurocircuitry recruitment using fMRI.
PLoS ONE, 5(7), e11440. doi:10.1371/journal.pone.0011440
Blakemore, S. J. (2008). The social brain in adolescence. Nature Reviews Neuro­
science, 9, 267–277. doi:10.1038/nrn2353
Boyer, T. W. (2006). The development of risk-taking: A multi-perspective review.
Developmental Review, 26, 291–345. doi:10.1016/j.dr.2006.05.002
Burnett, S., Bault, N., Coricelli, G., & Blakemore, S. J. (2010). Adolescents’ height-
ened risk-seeking in a probabilistic gambling task. Cognitive Development, 25,
183–196. doi:10.1016/j.cogdev.2009.11.003
Burnett, S., Thompson, S., Bird, G., & Blakemore, S. J. (2011). Pubertal develop-
ment of the understanding of social emotions: Implications for education. Learn-
ing and Individual Differences, 21, 681–689. doi:10.1016/j.lindif.2010.05.007
Casey, B. J., Somerville, L. H., Gotlib, I. H., Ayduk, O., Franklin, N. T., Askren,
M. K., . . . Shoda, Y. (2011). Behavioral and neural correlates of delay of gratifi-
cation 40 years later. Proceedings of the National Academy of Sciences of the United
States of America, 108, 14998–15003. doi:10.1073/pnas.1108561108
Chein, J., Albert, D., O’Brien, L., Uckert, K., & Steinberg, L. (2011). Peers increase
adolescent risk taking by enhancing activity in the brain’s reward circuitry.
Developmental Science, 14, F1–F10. doi:10.1111/j.1467-7687.2010.01035.x

risks, rewards, and the developing brain      87

13490-04_PT2_Ch03-3rdPgs.indd 87 11/15/13 1:43 PM

 Christakou, A., Brammer, M., & Rubia, K. (2011). Maturation of limbic corticos-
triatal activation and connectivity associated with developmental changes in
temporal discounting. NeuroImage, 54, 1344–1354. doi:10.1016/j.neuroimage.
2010.08.067
Crone, E. A., & Dahl, R. E. (2012). Understanding adolescence as a period of social-
affective engagement and goal flexibility. Nature Reviews Neuroscience, 13, 636–
650. doi:10.1038/nrn3313
Crone, E. A., & van der Molen, M. W. (2004). Developmental changes in real life
decision making: Performance on a gambling task previously shown to depend on
the ventromedial prefrontal cortex. Developmental Neuropsychology, 25, 251–279.
doi:10.1207/s15326942dn2503_2
Cunningham, W. A., Arbuckle, N. L., Jahn, A., Mowrer, S. M., & Abduljalil, A. M.
(2010). Aspects of neuroticism and the amygdala: Chronic tuning from motiva-
tional styles. Neuropsychologia, 48, 3399–3404. doi:10.1016/j.neuropsychologia.
2010.06.026
Dahl, R. E. (2004). Adolescent brain development: A period of vulnerabilities
and opportunities. Annals of the New York Academy of Sciences, 1021, 1–22.
doi:10.1196/annals.1308.001
Delgado, M. R. (2007). Reward-related responses in the human striatum. Annals of
the New York Academy of Sciences, 1104, 70–88. doi:10.1196/annals.1390.002
Eaton, D. K., Kann, L., Kinchen, S., Shanklin, S., Ross, J., Hawkins, J., . . . Wechsler,
H. (2008). Youth risk behavior surveillance—United States, 2007. MMWR.
Surveillance summaries: Morbidity and mortality weekly report. CDC Surveillance
summaries, 57(4), 1–131. Available at http://www.cdc.gov/mmwr/preview/
mmwrhtml/ss5704a1.htm
Ernst, M., & Fudge, J. L. (2009). A developmental neurobiological model of moti-
vated behavior: Anatomy, connectivity and ontogeny of the triadic nodes.
Neuroscience and Biobehavioral Reviews, 33, 367–382. doi:10.1016/j.neubiorev.
2008.10.009
Ernst, M., Nelson, E. E., Jazbec, S., McClure, E. B., Monk, C. S., Leibenluft, E., . . .
Pine, D. S. (2005). Amygdala and nucleus accumbens in responses to receipt
and omission of gains in adults and adolescents. NeuroImage, 25, 1279–1291.
doi:10.1016/j.neuroimage.2004.12.038
Ernst, M., Pine, D. S., & Hardin, M. (2006). Triadic model of the neurobiol-
ogy of motivated behavior in adolescence. Psychological Medicine, 36, 299–312.
doi:10.1017/S0033291705005891
Figner, B., Mackinlay, R. J., Wilkening, F., & Weber, E. U. (2009). Affective and delib-
erative processes in risky choice: Age differences in risk taking in the Columbia
Card Task. Journal of Experimental Psychology: Learning, Memory, and Cognition,
35, 709–730. doi:10.1037/a0014983
Fischhoff, B., Bruine de Bruin, W., Parker, A. M., Millstein, S. G., & Halpern-
Felsher, B. L. (2010). Adolescents’ perceived risk of dying. Journal of Adolescent
Health, 46, 265–269. doi:10.1016/j.jadohealth.2009.06.026

88       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 88 11/15/13 1:43 PM

 Forbes, E. E., Ryan, N. D., Phillips, M. L., Manuck, S. B., Worthman, C. M., Moyles,
D. L., . . . Dahl, R. E. (2010). Healthy adolescents’ neural response to reward:
Associations with puberty, positive affect, and depressive symptoms. Journal of
the American Academy of Child and Adolescent Psychiatry, 49, 162–172. Retrieved
from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2837556/
Galvan, A., Hare, T., Voss, H., Glover, G., & Casey, B. J. (2007). Risk-taking
and the adolescent brain: Who is at risk? Developmental Science, 10, F8–F14.
doi:10.1111/j.1467-7687.2006.00579.x
Galvan, A., Hare, T. A., Parra, C. E., Penn, J., Voss, H., Glover, G., & Casey, B. J.
(2006). Earlier development of the accumbens relative to orbitofrontal cortex
might underlie risk-taking behavior in adolescents. The Journal of Neuroscience,
26, 6885–6892. doi:10.1523/JNEUROSCI.1062-06.2006
Gardner, M., & Steinberg, L. (2005). Peer influence on risk taking, risk preference,
and risky decision making in adolescence and adulthood: An experimental
study. Developmental Psychology, 41, 625–635. doi:10.1037/0012-1649.41.4.625
Geier, C. F., Terwilliger, R., Teslovich, T., Velanova, K., & Luna, B. (2010). Imma-
turities in reward processing and its influence on inhibitory control in adoles-
cence. Cerebral Cortex, 20, 1613–1629. doi:10.1093/cercor/bhp225
Gogtay, N., Giedd, J. N., Lusk, L., Hayashi, K. M., Greenstein, D., Vaituzis, A. C., . . .
Thompson, P. M. (2004). Dynamic mapping of human cortical development
during childhood through early adulthood. Proceedings of the National Acad-
emy of Sciences of the United States of America, 101, 8174–8179. doi:10.1073/
pnas.0402680101
Hermans, E. J., Bos, P. A., Ossewaarde, L., Ramsey, N. F., Fernandez, G., & van
Honk, J. (2010). Effects of exogenous testosterone on the ventral striatal BOLD
response during reward anticipation in healthy women. NeuroImage, 52(1),
277–283. doi:10.1016/j.neuroimage.2010.04.019
Horvath, P., & Zuckerman, M. (1993). Sensation seeking, risk appraisal, and risky
behavior. Personality and Individual Differences, 14(1), 41–52. doi:10.1016/
0191-8869(93)90173-Z
LeDoux, J. (2007). The amygdala. Current Biology, 17, R868–R874.
Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of delay of gratification:
Dynamics of willpower. Psychological Review, 106(1), 3–19. doi:10.1037/0033-
295X.106.1.3
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex
function. Annual Review of Neuroscience, 24, 167–202. doi:10.1146/annurev.
neuro.24.1.167
Nelson, E. E., Leibenluft, E., McClure, E. B., & Pine, D. S. (2005). The social re-
orientation of adolescence: A neuroscience perspective on the process and its
relation to psychopathology. Psychological Medicine, 35, 163–174. doi:10.1017/
S0033291704003915
Op de Macks, Z. A., Gunther Moor, B., Overgaauw, S., Guroglu, B., Dahl, R. E.,
& Crone, E. A. (2011). Testosterone levels correspond with increased ventral

risks, rewards, and the developing brain      89

13490-04_PT2_Ch03-3rdPgs.indd 89 11/15/13 1:43 PM

 striatum activation in response to monetary rewards in adolescents. Develop-
mental Cognitive Neuroscience, 1, 506–516. doi:10.1016/j.dcn.2011.06.003
Padmanabhan, A., Geier, C. F., Ordaz, S. J., Teslovich, T., & Luna, B. (2011). Devel-
opmental changes in brain function underlying the influence of reward process-
ing on inhibitory control. Developmental Cognitive Neuroscience, 1, 517–529.
doi:10.1016/j.dcn.2011.06.004
Paus, T., Keshavan, M., & Giedd, J. N. (2008). Why do many psychiatric disor-
ders emerge during adolescence? Nature Reviews Neuroscience, 9, 947–957.
doi:10.1038/nrn2513
Quevedo, K. M., Benning, S. D., Gunnar, M. R., & Dahl, R. E. (2009). The
onset of puberty: Effects on the psychophysiology of defensive and appeti-
tive motivation. Development and Psychopathology, 21(1), 27–45. doi:10.1017/
S0954579409000030
Reyna, V. F., Estrada, S. M., DeMarinis, J. A., Myers, R. M., Stanisz, J. M., & Mills,
B. A. (2011). Neurobiological and memory models of risky decision making in
adolescents versus young adults. Journal of Experimental Psychology, 37, 1125–
1142. doi:10.1037/a0023943
Reyna, V. F., & Farley, F. (2006). Risk and rationality in adolescent decision making:
Implications for theory, practice, and public policy. Psychological Science in the
Public Interest, 7(1), 1–44. doi:10.1111/j.1529-1006.2006.00026.x
Rivers, S. E., Reyna, V. F., & Mills, B. (2008). Risk taking under the influence: A
fuzzy-trace theory of emotion in adolescence. Developmental Review, 28(1),
107–144. doi:10.1016/j.dr.2007.11.002
Scherf, K. S., Berman, M., & Dahl, R. E. (2012). Facing changes and changing faces
in adolescence: A new model for investigating adolescent-specific interactions
between pubertal, brain, and behavioral development. Developmental Cognitive
Neuroscience, 2, 199–219. doi:10.1016/j.dcn.2011.07.016
Schlottmann, A. (2001). Children’s probability intuitions: Understanding
the expected value of complex gambles. Child Development, 72, 103–122.
doi:10.1111/1467-8624.00268
Schlottmann, A., & Anderson, N. H. (1994). Children’s judgments of expected
value. Developmental Psychology, 30, 56–66. doi:10.1037/0012-1649.30.1.56
Shirtcliff, E. A., Dahl, R. E., & Pollak, S. D. (2009). Pubertal development: Cor-
respondence between hormonal and physical development. Child Development,
80, 327–337. doi:10.1111/j.1467-8624.2009.01263.x
Smith, A. B., Halari, R., Giampetro, V., Brammer, M., & Rubia, K. (2011). Devel-
opmental effects of reward on sustained attention networks. NeuroImage, 56,
1693–1704. doi:10.1016/j.neuroimage.2011.01.072
Somerville, L. H., Jones, R. M., & Casey, B. J. (2010). A time of change: Behav-
ioral and neural correlates of adolescent sensitivity to appetitive and aver-
sive environmental cues. Brain and Cognition, 72(1), 124–133. doi:10.1016/
j.bandc.2009.07.003

90       braams, van leijenhorst, and crone

13490-04_PT2_Ch03-3rdPgs.indd 90 11/15/13 1:43 PM

 Steinberg, L. (2004). Risk taking in adolescence: What changes, and why? Annals of
the New York Academy of Sciences, 1021, 51–58. doi:10.1196/annals.1308.005
Steinberg, L. (2008a). Adolescence. New York, NY: McGraw-Hill.
Steinberg, L. (2008b). A social neuroscience perspective on adolescent risk-taking.
Developmental Review, 28(1), 78–106. doi:10.1016/j.dr.2007.08.002
Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., & Woolard, J.
(2008). Age differences in sensation seeking and impulsivity as indexed by
behavior and self-report: Evidence for a dual systems model. Developmental Psy-
chology, 44, 1764–1778. doi:10.1037/a0012955
Steinberg, L., & Morris, A. S. (2001). Adolescent development. Annual Review of
Psychology, 52, 83–110. doi:10.1146/annurev.psych.52.1.83
van Duijvenvoorde, A. C. K., Jansen, B. R. J., Visser, I., & Huizenga, H. M. (2010).
Affective and cognitive decision-making in adolescents. Developmental Neuro-
psychology, 35, 539–554. doi:10.1080/87565641.2010.494749
van Honk, J., Schutter, D. J., Hermans, E. J., Putman, P., Tuiten, A., & Koppeschaar,
H. (2004). Testosterone shifts the balance between sensitivity for punishment
and reward in healthy young women. Psychoneuroendocrinology, 29, 937–943.
doi:10.1016/j.psyneuen.2003.08.007
van Leijenhorst, L., Gunther Moor, B., Op de Macks, Z. A., Rombouts, S. A.,
Westenberg, P. M., & Crone, E. A. (2010). Adolescent risky decision-making:
Neurocognitive development of reward and control regions. NeuroImage, 51(1),
345–355. doi:10.1016/j.neuroimage.2010.02.038
van Leijenhorst, L., Westenberg, P. M., & Crone, E. A. (2008). A developmental
study of risky decisions on the cake gambling task: Age and gender analyses of
probability estimation and reward evaluation. Developmental Neuropsychology,
33, 179–196. doi:10.1080/87565640701884287
van Leijenhorst, L., Zanolie, K., Van Meel, C. S., Westenberg, P. M., Rombouts,
S. A., & Crone, E. A. (2010). What motivates the adolescent? Brain regions
mediating reward sensitivity across adolescence. Cerebral Cortex, 20(1), 61–69.
doi:10.1093/cercor/bhp078

risks, rewards, and the developing brain      91

13490-04_PT2_Ch03-3rdPgs.indd 91 11/15/13 1:43 PM

13490-04_PT2_Ch03-3rdPgs.indd 92 11/15/13 1:43 PM
 4
the Adolescent
Sensation-Seeking Period:
Development of Reward
Processing and Its Effects
on Cognitive Control
Beatriz Luna, Aarthi Padmanabhan, and Charles Geier

In this chapter, we review existing animal models and human literature on the
adolescent brain in the context of risk-taking behaviors. Specific changes in brain
structure and neurochemistry as well as a review of the functional neuroimaging
literature reflecting the development of cognitive control and incentive process-
ing will be discussed. We extend previous reviews (e.g., Geier & Luna, 2009;
Wahlstrom, White, & Luciana, 2010) by proposing that relative immaturities
in the adolescent brain, specific to networks supporting motivated behaviors,
may underlie increases in sensation-seeking behaviors in adolescence. Finally,
the implications for juvenile law and education are discussed. Which aspects can
and cannot be informed by this research are carefully delineated, especially with
regard to ecological validity.
Adolescence is typically demarcated by the onset of puberty, early in the
second decade of life in humans, and the acquisition of adult psycho­social roles
(approximately early 20s). This period is widely recognized across cultures
and species as demonstrating a peak in sensation seeking—the propensity

http://dx.doi.org/10.1037/14322-005
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

93

13490-05_Ch04-4thPgs.indd 93 11/15/13 1:43 PM

 to approach novel and varied experiences (Spear, 2010). Sensation seeking
during adolescence is increasingly theorized to be an adaptive mechanism
leading to the acquisition of skills that can be implemented independently
of adult supervision supporting lifelong survival (although there are, to our
knowledge, no studies that have tested this directly). Increased sensation
seeking can also lead to risk taking (engaging in behaviors with a high per-
ceived reward value but known potential negative consequences), which
can undermine survival. Increased risk taking during adolescence plays a pri-
mary role in increased mortality rates relative to childhood (Heron, 2012).
As discussed below, the unique neurobiological changes that occur during
this time may play a crucial role in supporting a rise in sensation seeking
and risk taking. On the one hand, brain systems supporting executive func-
tion are reaching adult levels during adolescence, supporting their ability to
make goal-directed decisions in a manner similar to that of adults. However,
important limitations in the ability to engage these executive processes in a
controlled and flexible manner may enhance their vulnerability to errors in
executive function, such as failing to suppress an automatic response. Neuro­
biological changes, in conjunction with still-maturing executive processes,
affect motivated behaviors (behaviors that are influenced by reward and
punishment) and increase adolescent sensitivity to reward incentives. Taken
together, the maturation of executive control of behavior, coupled with
immaturities in assessing rewards and punishment, may lead to a system that
is optimally tuned for exploration and sensation seeking.

Adolescent Risk Taking

Adolescence is well recognized to be a period of peak risk taking

(Steinberg et al., 2008), which can undermine survival. Accidental deaths
and unintended injuries peak during the adolescent period (see Figure 4.1)
due in great part to risky and reckless behaviors, including substance abuse,
criminal behavior, and unprotected sex (Chambers, Taylor, & Pentenza, 2003).
The decision-making process involved in risk taking is affected by many com-
ponent processes and includes the following: the ability to assess the value
of a potential reward relative to potential consequences, arousal processes in
response to potential rewards, and the planning of behaviors that increase
the probability of obtaining the reward. Investigating the maturational integ-
rity of brain processes that support components that affect the risk-taking
decision-making process can inform us with regard to the nature of an ado-
lescent predisposition for engaging in risk taking.
To better conceptualize this multiprocess perspective on decision mak-
ing, it is useful to more carefully consider what goes into a “risky” decision.

94       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 94 11/15/13 1:43 PM

 60

Percent of Total Accidental Deaths (2008)

50

40

30

20

10

0
1-4y 5-9y 10-14y 15-19y 20-24y 25-34y 35-44y 45-54y 55-64y 65-74y 75-84y All
Ages in Years Ages

Figure 4.1. Percentage of total accidental deaths by age. From “Deaths: Leading
Causes for 2008,” by M. Heron, 2012, National Vital Statistics Reports, 60, 17–19.
In the public domain.

For example, envision a scenario in which an adolescent is faced with the

choice to share his or her friend’s lit cigarette. The adolescent faces at least
two distinct behavioral choices here: take a drag or refuse. Initially, the ado-
lescent must first assess and generate representations of the benefits and costs
of each action, supported in large measure by reward system processing. For
example, “If I take a puff, my friend will think I’m cool [social reward], but I
might get sick and get in trouble” versus “If I don’t take a puff, I won’t get sick,
but I might lose ‘face’ [social punishment].” The costs and benefits of each
action must then be evaluated against each other, supported in part by execu-
tive processes (i.e., working memory) and processing of potential reward-
ing outcomes. Next, behavioral action toward one option is planned and
executed while action toward the alternative is inhibited. Finally, the reward/
loss outcome associated with the behavioral choice is noted and stored to
help guide future choices.
Although there are important environmental factors, such as social
demands and learned responses, that influence decision making, the mecha-
nisms that support the basic aspects of reward processing and goal-directed
behaviors are supported by brain processes that are still immature in ado-
lescence. As we discuss next, the adolescent brain systems that coordinate
and integrate reward and cognitive processes may be distinct from those of
adults, biasing toward impulsive behaviors driven by a greater influence of
reward anticipation. Hence, adolescents may have the ability to assess risk,

the adolescent sensation-seeking period      95

13490-05_Ch04-4thPgs.indd 95 11/15/13 1:43 PM

 but the value of rewards may have a larger impact on the behaviors that are
ultimately engaged.

Brain Maturation in Adolescence

Brain Structure

The gross morphology of the brain is well formed even in childhood

(Caviness, Kennedy, & Bates, 1996). Through adolescence, however, progres-
sive and regressive brain maturation processes, which are believed to optimize the
brain to its environmental demands, continue to affect brain function. Synaptic
pruning, the loss of underused synaptic connections, reaches adult levels in the
visual cortex and temporal language regions in childhood, while the prefrontal
cortex (PFC) reaches adult levels in middle adolescence (Huttenlocher, 1990;
Huttenlocher & Dabholkar, 1997). Magnetic resonance imaging (MRI) studies
of gray matter thickness indicate that gray matter specialization proceeds in an
inverted U fashion with increased thickening into adolescence and thinning
proceeding into adulthood (Giedd et al., 1999; Gogtay et al., 2004; see also
Chapter 3, this volume).
Importantly, the regions in PFC and basal ganglia, which support execu-
tive function and reward processing, show continued thinning through ado-
lescence (Gogtay et al., 2004; Sowell, Thompson, Holmes, Jernigan, & Toga,
1999). Pruning of excess synaptic connections in the PFC, which contributes
to gray matter thinning, provides more streamlined and direct computations
(decreasing signal-to-noise), which can enhance more complex information
processing, such as working memory and the ability to voluntarily inhibit
responses (Fuster, 2002). As a result, the adolescent period is characterized
by a generalized streamlining of localized circuitry including the addition of
specialized inputs (see Neurochemical Maturation below) supporting more
complex, integrated function.
In addition to gray matter remodeling, processes influencing white mat-
ter integrity continue to develop through adolescence. The myelin sheath
surrounding axons continues to thicken across the brain through myelina-
tion, speeding neuronal transmission and enhancing neuronal connectiv-
ity, which contributes to the establishment of specialized neural networks.
Histological studies indicate earlier myelination in the visual cortex, with
frontal, temporal, and parietal myelination continuing through adolescence
and early adulthood (Yakovlev & Lecours, 1967). Diffusion tensor imaging
studies indicate that white matter integrity in pathways such as the cingu-
lum, the superior longitudinal fasciculus, and uncinate fasciculus continues
to mature through childhood and adolescence and even into early adulthood

96       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 96 11/15/13 1:43 PM

 (Asato, Terwilliger, Woo, & Luna, 2010; Eluvathingal, Hasan, Kramer,
Fletcher, & Ewing-Cobbs, 2007; Lebel, Walker, Leemans, Phillips, & Beaulieu,
2008; Giorgio et al., 2008). These increases in white matter integrity support
speeding of neuronal transmission of cortico-cortical and cortico-subcortical
integration. Effective cortico-subcortical connectivity is essential for top-
down modulation of behavior where executive cognitive systems can exert
control over more reflexive/impulsive systems (Hwang, Velanova, & Luna,
2010). Taken together, these results indicate that the structural maturation
of the adolescent prefrontal system is newly online while white matter path-
ways are beginning to strengthen. These milestones may support the abil-
ity of adolescents to have access to executive processes as adults, but their
reliable implementation may still be suboptimal, possibly undermining the
engagement of this system when there are competing processes such as those
of the reward system, which also have a unique status in adolescence, as
discussed below.

Neurochemical Maturation

Underlying behavioral and systems-level maturation, a number of changes

occur at a more micro level over childhood and adolescence, including an
overexpression of receptors for serotonin, dopamine, adrenergic, and endo-
cannabinoids (Lidow & Rakic, 1992); a peak in the density of interneurons
with a subsequent decline into adulthood (Anderson, Classey, Conde, Lund,
& Lewis, 1995; Erickson & Lewis, 2002; Lewis, 1997); an increase in levels of
the inhibitory neurotransmitter gamma-aminobutyric acid (GABA; Hedner,
Iverson, & Lundberg, 1984), and a change in the expression of glutamate
activating receptors in the PFC. These changes are believed to refine neuronal
signaling and connectivity, which enhance efficient cellular signaling into
adulthood.
The reward system is highly dependent on the function of the neuro­
transmitter dopamine (DA). Lesion, single-cell recording, neuroimaging,
electrical self-stimulation, and drug studies have collectively demonstrated
that DA neurons respond to reward prediction, expected reward value, salience,
punishment, and valence (for reviews, see Cools, 2008; Schultz et al., 2008;
Wise, 2004). DA, which primarily modulates fast-acting synapses (i.e., glu-
tamate and GABA), extensively innervates brain circuitry and modulates
a strong reciprocal relationship between the striatum and the PFC (Cools,
2008). The prevalence and function of DA as well as its influence on modulat-
ing neuronal processes demonstrate protracted change over adolescence, which
we briefly summarize below (for more extensive reviews, see Chambers et al.,
2003; Spear, 2000; Wahlstrom, Collins, White, & Luciana, 2010; Wahlstrom,
White, & Luciana, 2010).

the adolescent sensation-seeking period      97

13490-05_Ch04-4thPgs.indd 97 11/15/13 1:43 PM

 In nonhuman primates, there is evidence of pubertal “peaks” in DA
neuron activity in the midbrain (McCutcheon, White, & Marinelli, 2009),
increased DA tissue concentration in the PFC in adolescence (Goldman-Rakic
& Brown, 1982), a peak in DA tone in the striatum (Andersen, Thompson,
Krenzel, & Teicher, 2002), and a peak in activity in D1-expressing neu-
rons (Brenhouse, Sonntag, & Andersen, 2008). Postmortem human studies
demonstrate that DA innervations to the PFC peak in adolescence (Benes,
Taylor, & Cunningham, 2000; Lambe, Krimer, & Goldman-Rakic, 2000;
Rosenberg & Lewis, 1994, 1995), with the largest increase in PFC cortical
layer III, where neural processing supporting cognitive behaviors (i.e., work-
ing memory) has been characterized (Lewis, Melchitzky, & Burgos, 2002).
The major DA receptors (D1 and D2) show increased density in the PFC
and the striatum in adolescence, with peaks in late childhood and subsequent
declines through adolescence (Andersen et al., 2002; Lidow & Rakic, 1992;
Montague, Lawler, Mailman, & Gilmore, 1999; Seeman et al., 1987). The
changes in D1- and D2-receptor function in the PFC serve to balance excit-
atory and inhibitory responses in the brain, fine-tuning neuronal responses
and supporting more efficient processing in the PFC (Tseng & O’Donnell,
2005). The striatum has relatively more D2 receptors than the PFC, while
the opposite is the case for D1 receptors. It is not yet clear how changes in the
density of the DA receptors would specifically underlie adolescent behavior.
The peak in DA signaling in the midbrain in adolescence has also been cor-
roborated by evidence that adolescent animals exhibit increased reinforc-
ing effects of stimulant drugs (Adriani, Chiarotti, & Laviola, 1998; Adriani
& Laviola, 2000; Badanich, Adler, & Kirstein, 2006; Laviola, Adriani,
Terranova, & Gerra, 1999; Mathews & McCormick, 2007) and demonstrate
increased sensitivity to neuroleptics that are DA receptor antagonists (Spear
& Brake, 1983; Spear, Shalaby, & Brick, 1980).
Overall, changes in maturational rates in D1 and D2 receptors, metabo-
lite function, and midbrain DA neuron function may underlie changes seen
in motivated behaviors in adolescence as well as confer risk for addiction and
other pathologies related to DA function that emerge in adolescence. Tonic
levels of DA are elevated in adolescence, which may result in inefficient
regulation of phasic signaling in response to rewards (for review, see Luciana,
Wahlstrom, Porter, & Collins, 2012). Taken together, this literature indicates
an overall peak in DA availability, which may enhance reward reactivity
in adolescence, coupled with a short-acting response bias toward immedi-
ate rewards and resulting in overexcitation to perceived reward incentives
and enhanced sensation seeking (Chambers et al., 2003). In addition to
reward assessment, risk taking involves engaging in an action that results in
reward receipt. Next, we discuss the state of the cognitive executive system
in adolescence.

98       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 98 11/15/13 1:43 PM

 Cognitive Development in Adolescence

Crucial to decision making is the ability to engage systems that support

immediate goal-directed responses (vs. reflexive responses or habits) to inter-
act optimally with the environment. Motivation is a key component, which
can be guided by a potential reward with high perceived value. Executive
control of behavior is available early in development, as demonstrated in
tasks that assess the ability to update goal-driven responses (Diamond &
Goldman-Rakic, 1989; Johnson, 1995). However, executive control becomes
increasingly flexible and stable with age as prefrontal networks become more
readily engaged, as evident in working memory tasks and tasks where rules
are unexpectedly changed (Bjorklund & Harnishfeger, 1995; Hwang & Luna,
2011; Luciana & Nelson, 1998).
Developmental functional MRI (fMRI) studies of working memory,
where subjects’ performance is guided by information retained online, show
that prefrontal systems play a primary role (Goldman-Rakic, 1987) and the
core processes to engage working memory are available starting in childhood
(Nelson et al., 2000; Thomas et al., 1999). When more complex computa­
tions are required by the experimental task (i.e., manipulating informa-
tion held in working memory), developmental differences are often more
apparent. These increases in cognitive abilities from early childhood to ado-
lescence include increases in the ability to engage optimally the prefron-
tal systems in generating an accurate response and to suppress distracters
(Ciesielski, Lesnik, Savoy, Grant, & Ahlfors, 2006; Crone, Wendelken,
Donohue, van Leijenhorst, & Bunge, 2006; Geier, Garver, Terwilliger, &
Luna, 2009; Klingberg, Forssberg, & Westerberg, 2002; Olesen, Macoveanu,
Tegner, & Klingberg, 2007; Scherf, Sweeney, & Luna, 2006 ), suggesting
that refinements in executive function are relatively late maturing. The
fMRI studies on the development of inhibitory control, that is, the ability
to suppress reflexive goal-incompatible responses, also demonstrate imma-
turities that persist through adolescence in engaging optimal brain circuit-
ries. Immaturities are reflected in both increases and decreases in engaging
the PFC as well as the concomitant cortical and subcortical regions (Bunge,
Dudukovic, Thomason, Vaidya, & Gabrieli, 2002; Crone et al., 2006; Luna
et al., 2001; Rubia et al., 2006).
Although the main circuitry supporting cognitive control is online
by adolescence, specific processes underlying the ability to effectively
engage cognitive control continue to show limitations. The anterior cin-
gulate cortex, which is robustly engaged in adults when inhibitory control
errors are committed, is not reliably engaged in adolescence, contributing
to limitations in error monitoring and performance monitoring (Velanova,
Wheeler, & Luna, 2008; see Color Plate 4a). In addition, the circuitry,

the adolescent sensation-seeking period      99

13490-05_Ch04-4thPgs.indd 99 11/15/13 1:43 PM

 including prefrontal and posterior cortical regions (e.g., inferior parietal
sulcus), that supports the ability to sustain attention is also engaged to
a lesser degree in adolescence than in adulthood and limits the ability
for adolescents to establish a controlled cognitive state and reach adult
rates of correct performance (Velanova, Wheeler, & Luna, 2009; see
Color Plate 4b). These limitations may be due to immaturities in func-
tional connectivity of top-down control processes that support the ability
to exert executive control (Hwang et al., 2010; Stevens, Kiehl, Pearlson,
& Calhoun, 2007; see Color Plate 4c).
Prior research has suggested that from childhood to adolescence the
reliance on posterior parietal systems shifts to prefrontally guided circuitry
(Hwang et al., 2010; Luna et al., 2001). By adolescence, the foundational
architecture of a prefrontally guided circuitry is evident. From adolescence
to adulthood there are increases in the number and strength of prefron-
tally guided connections supporting a ready system to control responses in
adulthood (Hwang, Hallquist, & Luna, 2012; see Color Plate 4d). These
developmental advances in readily recruiting integrated networks are
likely supported by age-related weakening of short-range functional con-
nections that give way to long-distance connections in adulthood (Fair
et al., 2007). Effective executive control is supported by the ability of
prefrontal executive regions to readily integrate attention and sensory
information in order to exert control over subcortical regions to execute
goal-directed behavior.
Taken together, these studies provide evidence that by adolescence
the brain systems supporting cognitive control are indeed available, sup-
porting the ability to plan and execute decisions. However, ready access to
networks that support controlled use of cognitive processes continues to
strengthen through adulthood, undermining the ability to sustain activity
and monitor performance. As such, on the one hand, access to prefrontal
networks may provide adolescents with the ability to start making complex
executive decisions without assistance from adults as had been needed in
childhood and to engage in new experiences. On the other hand, limita-
tions in the flexible, reliable, and prompt engagement of optimal execu-
tive networks can increase the occurrence of failures of exerting executive
control in favor of more impulsive responses. As described below, ado-
lescents may have a greater demand on inhibitory control as motivation
may play a larger role in affecting behavior and further undermining the
effective engagement of cognitive control. Hence, although the ability to
mathematically assess risk has been shown to be available in adolescence
(Reyna & Brainerd, 2011), the ability to readily integrate this information
in situations with competing processes, such as in adolescent risk taking,
may be limited.

100       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 100 11/15/13 1:43 PM

 Neuroimaging Evidence of Unique
Reward Processing in Adolescence

Extensive work in animal models (Apicella, Ljungberg, Scarnati, &

Schultz, 1991; Hikosaka, Nakumura, & Nakahara, 2006; Roesch & Olson,
2003, 2004; Schultz, 2000; Wise, 2002) and with human adults using func-
tional neuroimaging (Breiter, Aharon, Kahneman, Dale, & Shizgal, 2001;
Delgado, Locke, Stenger, & Fiez, 2003; Delgado, Nystrom, Fissell, Noll, &
Fiez, 2000; Elliott, Newman, Longe, & Deakin, 2003; Knutson, Westdorp,
Kaiser, & Hommer, 2000; McClure, York, & Montague, 2004; O’Doherty,
2004; Thut et al., 1997; ) have carefully delineated reward-related neuro-
circuitry, providing the essential groundwork for developmental investiga-
tions. Indeed, a growing number of developmental neuroimaging studies
have focused on characterizing developmental differences in how incentives
(rewards, losses) are processed during adolescence (for reviews, see Geier &
Luna, 2009; Somerville & Casey, 2010). Studies consistently have demon-
strated central roles for the ventral striatum (VS) and orbitofrontal cortex
(OFC) in multiple aspects of reward processing; as such, these regions will be
the primary focus below.
The VS, including the nucleus accumbens, is rich in DA receptors
and has extensive connections throughout cortical and subcortical struc-
tures (Chikama, McFarland, Amaral, & Haber, 1997; Di Martino et al.,
2008; Fudge, Kinishio, Walsh, Richard, & Haber, 2002; Haber, Kunishio,
Mizobuchi, & Lynd-Balta, 1995; Selemon & Goldman-Rakic, 1985). This
key node in the reward circuitry is involved in various aspects of motivated
behaviors, including the detection of incentives, the processing of value,
and reward prediction (Knutson & Cooper, 2005). The OFC, a functionally
heterogeneous and complex prefrontal executive region, has important con-
nections to the basal ganglia (Fuster, 1989; Rushworth, Noonan, Boorman,
Walton, & Behrens, 2011), positioning it to support varied functions related
to the executive assessment of rewards, including representations of subjec-
tive preference, valence and value estimation, and, importantly, the regu-
lation of planned behavior associated with incentives (Bechara, Damasio,
Damasio, & Anderson, 1994; Hare, O’Doherty, Camerer, Schultz, & Rangel,
2008; Kringelbach, 2005). Interestingly, the VS and OFC may be recruited
differentially during the adolescent period, with adolescent behavior being
relatively more driven by stimulus or bottom-up (e.g., stronger VS influence)
than planned or regulated (e.g., a weaker OFC influence) processes, relative
to adults. Below, we review evidence for both immature recruitment of the
VS and decreased engagement of the OFC during reward processing.
Increased engagement of the VS in adolescence has been found across
different task contexts, including those in which subjects had to guess which

the adolescent sensation-seeking period      101

13490-05_Ch04-4thPgs.indd 101 11/15/13 1:43 PM

 response would generate a monetary reward (e.g., Ernst et al., 2005), when
monetary incentives were dependent on simple working memory responses
(e.g., Galvan et al., 2006), during passive viewing of different reward prob-
abilities (van Leijenhorst, Gunther Moor, et al., 2010), and when reward was
dependent on cognitive control performance (Geier, Terwilliger, Teslovich,
Velanova, & Luna, 2010; Padmanabhan, Geier, Ordaz, Teslovich, & Luna,
2011; van Leijenhorst, Gunther Moor, et al., 2010; see Figure 4.2). However,
when participants are prompted by an abstract cue associated with rewards
to make a quick button press to earn the reward (using the monetary incen-
tive delay task), adolescents show decreased VS activity compared to adults
during initial stages of reward processing (Bjork et al., 2004; Bjork, Smith,
Chen, & Hommer, 2010). These results may be due in part to the use of an

Neutral Trials
A. Whole Trial (Ventral Striatum) Reward Trials

Children Adolescents Adults

0.2 0.2 0.2

% signal change

% signal change

0.1 0.1 % signal change 0.1

0.0 0.0 0.0

0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
18
.5
22 1
.5
24

0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
19 8
.5
21
.5
24

0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
19 8
.5
22 1
.5
24
1.

4.

7.

1.

4.

7.

1.

4.

7.

2
13

19

13

22

13
-0.1 -0.1 -0.1

-0.2 -0.2 -0.2

Time (seconds) Time (seconds) Time (seconds)

Adult Neutral Trials

Adult Reward Trials
Adolescent Neutral Trials
B. Stages of Reward Processing Adolescent Reward Trials

Cue Preparation Response

0.15 0.5 0.3
(11, 8, -7)
0.1
VS 0.4 VS 0.2 (17, 8, 2)
% MR Signal Change

% MR Signal Change

% MR Signal Change

0.05 0.3
0.1
0 0.2
0
0.1
-0.05 -0.1
0
-0.1 -0.2
-0.1
-0.15
-0.2
(14, 2, -7) -0.2 -0.3 Putamen
-0.3 -0.4
0
5
3
5
6
5
10 9
.5
13 2
.5
16 5
.5
18

16 5
.5
18

.5
13 2
.5
16 5
.5
18
0
5
3
5
6
5
10 9
.5
12
.5

0
5
3
5
6
5
10 9
1.

4.

7.

4.

7.

1
1.

4.

7.

1.
13

TIME (sec) TIME (sec) TIME (sec)

Figure 4.2. (A). Adolescents show greater activity in ventral striatum (VS) when con-
sidering a whole reward trial versus neutral trial. From “Developmental Changes in
Brain Function Underlying the Influence of Reward Processing on Inhibitory Control,”
by A. Padmanabhan, C. F. Geier, S. J. Ordaz, T. Teslovich, and B. Luna, 2011, Devel-
opmental Cognitive Neuroscience, 1, p. 525. Copyright 2011 by Elsevier. Adapted
with permission. (B). When considering stages of reward processing it is evident that
this increase occurs during the response preparation period. Adults show increased
VS activity compared with adolescents in the initial period when the reward cue is
being assessed. From “Immaturities in Reward Processing and Its Influence on Inhibi-
tory Control in Adolescence,” by C. F. Geier, R. Terwilliger, T. Teslovich, K. Velanova,
and B. Luna, 2010, Cerebral Cortex, 20, pp. 1620–1623. Copyright 2010 by Oxford
University Press. Adapted with permission.

102       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 102 12/9/13 11:55 AM

 abstract incentive cue, which may undermine motivation or may point to
unique phases of reward processing that may have different developmental
trajectories (Geier et al., 2010). These results suggest that adolescents have
differential engagement of the VS, which typically expresses as an increased
engagement of the VS, reflecting sensitivity to detection, valuation, and pre-
diction of rewards but can also result in suboptimal recruitment of reward
circuitry in specific reward contingencies or during specific phases of reward
processing.
In our own studies (Geier et al., 2010; Padmanabhan et al., 2011), we
have used a rewarded antisaccade (AS) task to assess developmental change
in reward sensitivity and its interaction with a basic component of cognitive
control, response inhibition. At the start of each rewarded AS trial, partici-
pants are presented with cues that signal whether corrected performance on
the forthcoming trial will result in the accrual, loss, or no gain of points lead-
ing to a monetary reward. This stage is when cue information is processed
regarding potential incentives. Next, the incentive cue disappears and is
replaced by a central red fixation. Participants are told beforehand that they
should prepare to inhibit a response at this time. This second stage serves as
an instruction cue that is known from single-cell nonhuman primate stud-
ies to engage preparatory oculomotor control regions, including the frontal
eye field (FEF), supplementary eye field, and parietal eye field, that deter-
mine a correct inhibitory response (Everling & Munoz, 2000). Importantly,
this stage is also one when anticipation of an upcoming reward takes place.
Finally, the fixation cue disappears and a small filled circle appears at an
unpredictable peripheral location. This is the motor response stage, when
participants must not look at the stimulus when it appears, instead generating
an eye movement to the exact opposite location. In this task, reward accrual
is contingent on the ability to inhibit a prepotent saccadic response to the
suddenly appearing visual target (Hallett, 1978). Developmental differences
were investigated during these separate stages of reward processing: reward
assessment (processing of the rewarded/neutral cue), response preparation/
reward anticipation (presentation of fixation/instruction cue), and response
(correct inhibitory response; Geier et al., 2010).
Behaviorally, both reaction time and performance (correct inhibitory
response rates) improved during rewarded versus neutral trials (Geier et al.,
2010; Geier & Luna, 2012). The fMRI results indicated that adults, but
not adolescents, showed prompt recruitment of the VS during the initial
stage of reward cue assessment (see Figure 4.2b). By the response prepara-
tion period, however, adolescents but not adults showed robust engagement
of the VS, reaching higher levels of activation than adults. Importantly, in
parallel to relative greater recruitment of the VS, adolescents also showed
greater engagement of oculomotor control regions including the FEF and

the adolescent sensation-seeking period      103

13490-05_Ch04-4thPgs.indd 103 11/15/13 1:43 PM

 intraparietal sulcus than did adults (see Figure 4.3). In single-cell nonhuman
primate studies, these regions have been found to increase activity during
the preparatory period determining a successful inhibitory eye movement
response (Everling & Munoz, 2000). Given that adolescents performed at
adult levels only during incentive trials, when they recruit oculomotor con-
trol regions to a greater level than adults, and not in neutral trials, reflects
the greater effort that adolescents exert to engage inhibitory control. The
parallel increased recruitment of VS and oculomotor regions during reward
trials suggests that VS activity may support enhanced recruitment of cir-
cuitry that enables reward receipt. This extends prior interpretation of the
relationship between VS function and control processes, which is most often
Neutral Trials
A. Whole Trial (Intraparietal Sulcus) Reward Trials

Children Adolescents Adults

0.4 0.4 0.4

0.3 0.3 0.3
% signal change

% signal change

% signal change
0.2 0.2 0.2
0.1 0.1 0.1
0.0 0 .00
0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
19 8
.5
22 1
.5
24

0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
19 8
.5
22 1
.5
24

0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
19 8
.5
21
.5
24
-0.1 -0.1 -0.1
1.

4.

7.

1.

4.

7.

1.

4.

7.

1
13

13

13

22
-0.2 -0.2 -0.2
Time (seconds) Time (seconds) Time (seconds)

Adult Neutral Trials

Adult Reward Trials
Adolescent Neutral Trials
B. Stages of Reward Processing (Frontal Eye Fields) Adolescent Reward Trials

Cue Preparation Response

0.15
(26, -13, 53) 0.25 0.25
(-31, -10, 44) 0.2 (26, -13, 50)
% MR Signal Change

0.1 0.2
% MR Signal Change
% MR Signal Change

0.15
0.15 0.1
0.05
0.1 0.05
0 0.05 0
-0.05
-0.05 0
-0.1
-0.05 -0.15
-0.1
-0.1 -0.2
0
5
3
5
6
5
10 9
.5
12
.5
16 5
.5
18
1.

4.

7.

1
13

0
5
3
5
6
5
10 9
.5
13 2
.5
16 5
.5
18

0
5
3
5
6
5
10 9
.5
13 2
.5
16 5
.5
18
1.

4.

7.

1.

4.

7.

TIME (sec)
TIME (sec) TIME (sec)

Figure 4.3. Adolescents show greater activity in regions that support cognitive
control of eye movements during a rewarded antisaccade task. (A). When considering
the whole reward trial, adolescents compared with adults show increased recruitment
of the intraparietal sulcus, known to support voluntary saccade inhibition in reward
versus neutral trials. From “Developmental Changes in Brain Function Underlying the
Influence of Reward Processing on Inhibitory Control,” by A. Padmanabhan,
C. F. Geier, S. J. Ordaz, T. Teslovich, and B. Luna, 2011, Developmental Cognitive
Neuroscience, 1, p. 525. Copyright 2011 by Elsevier. Adapted with permission.
(B). More specifically, adolescents compared with adults show greater engagement of
the frontal eye fields, which are core to the ability to suppress an impending saccade
during preparation to make a correct inhibitory response. From “Immaturities in Reward
Processing and Its Influence on Inhibitory Control in Adolescence,” by C. F. Geier,
R. Terwilliger, T. Teslovich, K. Velanova, and B. Luna, 2010, Cerebral Cortex, 20,
pp. 1620–1623. Copyright 2010 by Oxford University Press. Adapted with permission.

104       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 104 12/9/13 11:56 AM

 characterized in a top-down manner with the VS (and reward processes more
generally) being regulated by prefrontal-mediated executive systems. The VS
may play a role in integrating reward information to affect response behavior.
As such, adolescents’ heightened reward responses may be viewed as “adap-
tive” in that reward approach is enhanced. However, adolescents’ engage-
ment of reward and motor systems closer to the response stage (preparation
vs. cue, as in adults) may lead to a more “impulsive” system in the context of
reward motivation.
Although there is general consensus that adolescents often show height-
ened responses in the VS, specific elements of reward-related decision making
contributing to VS activation may be developing along different timelines
in the adolescent period. In an elegant study, Cohen et al. (2010) found that
when decision value (the value assigned to a potential choice) and predic-
tion error (the difference between the expected value and actual outcome
of a choice or behavior) are considered independently, there is evidence
that the peak in adolescent hyperactivity in the VS may be due to the pro-
cessing of underlying prediction errors separate from the decision-making
process per se. That is, the magnitude of the reward itself does not affect
adolescent behavior as much as the anticipation of a reward in relation to
reward receipt. In addition, heightened responses in the VS in adolescence
to rewarding stimuli may in part be linked to hormonal changes associated
with puberty, as early puberty has been found to be associated with increased
striatal activity (subsequently declining with more advanced puberty) and
has been linked with testosterone levels (Forbes & Dahl, 2010; Op de Macks
et al., 2011).
In contrast to the directionality of immature VS responding, several
studies have also shown decreased engagement of the OFC (Galvan et al.,
2006; Geier et al., 2010; van Leijenhorst, Zamoli, et al., 2010) in adoles-
cents relative to adults, suggesting more limited executive assessment, regu-
lation of incentives, or both. In addition, adolescents with higher reported
risk taking show less recruitment of the OFC, as well as other prefrontal
regions (Galvan, Hare, Voss, Glover, & Casey, 2007; Shad, Bidesi, Chen,
Ernst, & Rao, 2011; van Leijenhorst, Crone, & Bunge, 2006), suggesting a
mechanism by which limitations in engaging executive control may lead to
increased reward-driven behavior. Supporting the findings in OFC, consider-
able evidence suggests increased engagement of more distributed prefrontal
regions (dorsolateral prefrontal cortex, dlPFC; ventromedial prefrontal cortex,
vmPFC) concurrent with the transition from adolescence to adulthood (Hwang
et al., 2010). Choosing high-risk options to obtain low-magnitude rewards has
been found to decrease from adolescence to adulthood and is mediated by pre-
frontal activity (van Leijenhorst, Gunther Moor, et al., 2010). From childhood
to later adulthood, the engagement of sustained attention networks that

the adolescent sensation-seeking period      105

13490-05_Ch04-4thPgs.indd 105 11/15/13 1:43 PM

 overlap with reward regions increases with reward contingencies (e.g., dlPFC
and vmPFC, dorsal striatal, and temporal-parietal regions; Smith, Halari,
Giampetro, Brammer, & Rubia, 2011; van Leijenhorst, Gunther Moor, et al.,
2010), while stimulus saliency networks, including posterior cingulate and
inferior temporal regions, decreased. In one imaging study using a decision-
making task where subjects could choose a safe bet or take a gamble, adoles-
cents and adults demonstrated similar behavior (Paulsen, Platt, Huettel, &
Brannon, 2011). However, brain regions in the decision-making network
that support processes such as integrating contextual information and pre-
vious outcomes, including the insula, the hippocampus, and the amygdala,
show increased activation through adolescence into adulthood (Paulsen et
al., 2011). These results suggest that with increasing age, the influence of
reward enhances top-down executive processing circuits, supporting mature,
adult levels of goal-directed behavior. Effective decision making in the con-
text of motivational cues necessitates successful integration of frontal execu-
tive systems and their influence on striatal reward systems. Indeed, increased
connectivity between the vmPFC and striatum, as well as frontoparietal and
insular brain regions, has been found to be associated with decreased impul-
sive choices on a temporal discounting paradigm (Christakou, Brammer, &
Rubia, 2011). Prefrontally guided functional connections supporting cogni-
tive control have been found to increase in number and in strength (Hwang
et al., 2010; Stevens et al., 2007) through adolescence as has the integrity of
white matter frontally guided connections (Asato et al., 2010). The effective
modulation of subcortical processes may support adult-level decision making
and, ultimately, decreases in risk-taking behavior (but also see Berns, Moore,
& Capra, 2009).
Taken together, these studies suggest two key components that are unique
in the adolescent period. First, adolescents have a predisposition for greater
sensitivity to reward contingencies, engaging areas of the reward circuitry
(VS) that heighten motivation and engaging motor action systems that may
result in more impulsive responses in the context of rewards. Second, pre-
frontal systems that support the ability to generate executive decisions are
available, permitting the generation of voluntary decision making. However,
though available, executive processes are still sluggish, possibly undermining
the assessment of reward contingencies in the context of planned behavior
and the ability to readily provide control of reflexive responses, undermining
controlled, reward-related decision making.
One limitation of developmental fMRI studies of reward processing
is that they typically use monetary incentives. However, monetary incen-
tives may be valued differently across development (e.g., $25 may be val-
ued more highly by an adolescent than an adult who has more ready access
to money). This could potentially confound interpretation of the effects of

106       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 106 11/15/13 1:43 PM

 rewards on behavior, as a more valued reward may induce greater motivation
and improved performance. Geier and Luna (2012) explored this notion in a
behavioral study aimed at minimizing reward value differences between ado-
lescents and adults. In this study, participants chose their own reward prior to
testing (maximizing individual-level subjective value), and instead of mon-
etary rewards, points toward a prize were offered, setting up a “fixed economy”
where the range of points was standardized across all subjects. Even in this
context, although adolescents committed more cognitive control errors than
adults, their error rates improved with increasing rewards, suggesting greater
sensitivity to reward. This was in contrast to the adults, whose performance
did not change with different reward amounts.
As a final point, in addition to considering age-related differences in
how adolescents value the same reward relative to adults, it is important
to consider age-related differences in how types of rewards are valued. For
instance, Chein, Albert, O’Brien, Uckert, and Steinberg (2011) found that
the presence of peers engaged the reward circuitry in adolescents to a greater
extent than in adults, which was associated with increased risk taking in a
driving task. These results suggest that not only do adolescents show greater
reward sensitivity but also certain types of incentives (e.g., social reward) may
be particularly salient motivators during this period. Adolescence is a period
of social reorientation during which there is a predilection for social experi-
ences that support the development of adult independent skills, including
establishing peer groups and increasing interests in forming romantic rela-
tionships (Blakemore, 2008; Forbes & Dahl, 2010).
In sum, the studies reviewed above demonstrate that adolescents exhibit
heightened responses in core regions of the reward system, likely reflecting
sensitivity to incentives, which may vary according to value magnitude and
type. Recruitment of the reward system may engage action processes that
facilitate reward receipt but may also contribute to impulsive decisions and
actions in the context of rewards. Cognitive systems are available supporting
the production of executive decisions. The cognitive systems are still limited
in important ways, such as in performance monitoring and error processing,
which in the presence of reward-driven action may undermine the system’s
ability to process context to inform optimal decision making.

Relevance to the Juvenile Justice System

In the past few years, there has been attention on how developmen-
tal neuroscience findings can and has informed the juvenile justice system,
particularly as it relates to sentencing such as the death penalty and life
without parole for juveniles (Scott & Steinberg, 2008). Although these are

the adolescent sensation-seeking period      107

13490-05_Ch04-4thPgs.indd 107 11/15/13 1:43 PM

 inherently ethical questions, the role that neuroscience plays in providing
important biologically based evidence to inform decisions at the level of the
U.S. Supreme Court has been increasingly considered (Graham v. Florida,
2010; Miller v. Alabama, 2012; Roper v. Simmons, 2005). First, it is important
to emphasize that developmental neuroscience findings describe the adoles-
cent period when considering adolescents as a group; the findings do not apply
at the individual level. There is great variability in both behavioral and fMRI
findings undermining the ability to inform individual legal cases in the deter-
mination of guilt or the ability to be rehabilitated. Even if in the future fMRI
evidence can inform propensities at the individual level, the risk of a false
positive, incorrectly identifying an individual as a life criminal, would still
limit its applicability.
Developmental neuroscience evidence of specific immaturities in brain
processes supporting decision making provides a biological model support-
ing what psychology has already revealed about limitations in adolescent
behavior and what has been surmised intuitively in society. As such, this
biologically based information does not provide novel insights so much as
it strengthens the impact of the information. The fMRI evidence depict-
ing increased activity in the VS during reward processing and limitations in
engaging executive circuitry in adolescents, combined with immaturities in
brain structure and neurotransmitter function, substantiates the behavioral
profile of increased motivational reactivity coupled with limitations in exec-
utive decision making during this period of development. The implication is
that age may be a mitigating circumstance when considering the culpability
(Graham v. Florida, 2010; Miller v. Alabama, 2012; Roper v. Simmons, 2005).
This evidence does not speak to guilt or exonerate responsibility but provides
insight into an important factor that underlies a risk-taking act that resulted
in a crime. As such, neuroscience evidence of immaturities in the adolescent
brain has been found to be particularly relevant in informing the extent of
sentencing. The fact that a crime had been committed during a period of
development when there are biologically based vulnerabilities to risk taking
implies that a behavior may not have occurred if the individual was mature
and that given time to develop, this propensity may be resolved. Sentences
such as the death penalty or life without parole assume that the individual
will not change his or her propensity for crime. In the majority of individuals
antisocial behavior may be limited only to the adolescent period (Moffitt,
1993). In addition, the processes that underlie increased sensation seeking
in adolescence also support enhanced learning. DA supports not only reward
processing but also learning contextual associations (Cools, 2011; Wise,
2004). Although an enhanced ability to learn contextual associations can
lead to establishing bad habits, such as substance abuse in adolescence, it

108       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 108 11/15/13 1:43 PM

 can also support a period when rehabilitation may be particularly effective.
Hence, juveniles may be more amenable to rehabilitation and have a greater
ability to be reintegrated in society.
The neuroscience evidence of immaturities in the adolescent brain sup-
porting decision making does not imply that adolescents are unable to make
responsible decisions. In fact, most adolescents make responsible decisions
and do not engage in risk-taking behavior. Adult-level executive systems
are starting to reach adult-like levels of function, although limitations still
remain in terms of the efficiency of engaging distributed networks that sup-
port top-down control of behavior. Adolescents are slower to engage and
maintain an executive mode of control and may be prone to errors. When
sensitivity to motivational cues is increased, the limitations in accessing the
control network may be further compromised. Hence, adolescent immatu-
rities are more pertinent to impulsive acts that occur without the ability to
plan and consider consequences. In the context of being able to deliberate a
decision, especially with guidance from a mature individual, the time to exert
the available adult-level executive systems allows the adolescent to guide a
responsible decision and consider long-term consequences.

The Adaptive Adolescent Period

Although the term immaturity makes reference to the adolescent brain

as a suboptimal adult brain, this period can also be viewed as adaptive and,
as such, the adolescent brain can be seen as operating in an optimal manner
for the demands of such a developmental stage. As adolescents gain access
to mature-level cognitive processes, they are able to make decisions that
are independent of the guidance of adults. Increases in motivational drives
encourage adolescents to make independent decisions that allow them to
explore new contexts and learn new abilities that are of particular fit to
them individually. One area of growth is in the social domain as adoles-
cents establish social circles and romantic relationships, which are also
influenced by hormonal changes during puberty (Forbes & Dahl, 2010).
These decisions are essential in establishing a foundation for moving away
from the proverbial “nest” and being able to establish adult roles. The abil-
ity to access cognitive processes that support executive decision making
(despite increased sensitivity to rewarded stimuli) provides an important
platform to encourage the acquisition of independent skills. Viewed in this
manner, the adolescent brain is in fact optimal for the critical task of this
stage of development—beginning to establish independence for a success-
ful adulthood.

the adolescent sensation-seeking period      109

13490-05_Ch04-4thPgs.indd 109 11/15/13 1:43 PM

 Conclusion

Neuroimaging evidence indicates that the adolescent period is unique

in its hyperreactivity to reward incentives in the context of having access to
adult-level cognitive tools that are still strengthening and not yet optimally
reliable. Gray matter is still thinning throughout the cortex, supporting more
efficient processing needed for complex neuronal computations that support
processes underlying executive control. White matter tracts are still myelina-
ting, speeding neuronal transmission needed for the engagement of distrib-
uted networks that support integrated information processing and ready
top-down control of behavior. Importantly, the reward system is hyperactive
as there is a peak of DA availability. Taken together, the adolescent period is
one in which independent executive decisions can be made, but the period
may be suboptimal in the ability to maintain a controlled executive state and
integrate errors, especially when reward contingencies are in place. Thus, the
adolescent period is one of heightened risk taking. The transitional nature of
this reward-driven and reactive period of development is particularly infor-
mative to the juvenile justice system when considering lifelong sentences
that assume an unchanging profile. On the other hand, adolescent access
to adult-level executive tools allows for responsible decisions to be made
in optimal circumstances. Finally, the adolescent brain is not to be viewed
as a suboptimal adult brain but rather as a stage in brain maturation that is
optimal for fostering the acquisition of independent skills and social networks
essential for adult survival.
Sensation/novelty-seeking behaviors peak during adolescence and likely
contribute to well-documented increases in risk-taking behaviors during this
period of development. Concurrent with these behavioral changes are a number
of neurobiological changes influencing both affective and cognitive processes,
resulting in an increased sensitivity to reward contingencies in the context
of maturing executive control systems. Brain maturational changes in gray
and white matter during adolescence enhance neuronal processing support-
ing complex cognitive computations. Connectivity between distal brain areas
strengthens, allowing integration of networks supporting top-down control
of behavior. Furthermore, neurochemical changes occur, including a peak in
dopamine availability, which may critically underlie peaks in sensation-seeking
behavior. Functional neuroimaging studies, showing increased recruitment of
reward processing regions and immature engagement of prefrontal regions
in adolescence compared with adults, are in agreement with brain structural
imaging studies showing continued maturation of fronto-striatal systems sup-
porting reward processing. Taken together, the literature suggests that ado-
lescents are especially influenced by reward contingencies, which can lead to
vulnerability to biased and, at times, suboptimal decision making. A different

110       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 110 11/15/13 1:43 PM

 yet complementary perspective suggests that an adolescent sensation-seeking
period may also be adaptive as it encourages exploration of one’s environment,
novel social interactions, and the acquisition of skills supporting independence
into adulthood. The juvenile justice system is increasingly informed by devel-
opmental neuroscience, which provides a model for adolescent vulnerabilities
for risk taking that can result in crime. This research underscores the nature of
this developmental stage, which can be pertinent to questions of culpability,
sentencing, and rehabilitation.

References

Adriani, W., Chiarotti, F., & Laviola, G. (1998). Elevated novelty seeking and
peculiar d-amphetamine sensitization in periadolescent mice compared with
adult mice. Behavioral Neuroscience, 112, 1152–1166. doi:10.1037/0735-7044.
112.5.1152
Adriani, W., & Laviola, G. (2000). A unique hormonal and behavioral hypo­
responsivity to both forced novelty and d-amphetamine in periadolescent mice.
Neuro­pharmacology, 39, 334–346. doi:10.1016/S0028-3908(99)00115-X
Andersen, S. L., Thompson, A. P., Krenzel, E., & Teicher, M. H. (2002). Pubertal
changes in gonadal hormones do not underlie adolescent dopamine receptor
overproduction. Psychoneuroendocrinology, 27, 683–691. doi:10.1016/S0306-
4530(01)00069-5
Anderson, S. A., Classey, J. D., Conde, F., Lund, J. S., & Lewis, D. A. (1995). Syn-
chronous development of pyramidal neuron dendritic spines and parvalbumin-
immunoreactive chandelier neuron axon terminals in layer III of monkey
prefrontal cortex. Neuroscience, 67, 7–22. doi:10.1016/0306-4522(95)00051-J
Apicella, P., Ljungberg, T., Scarnati, E., & Schultz, W. (1991). Responses to reward in
monkey dorsal and ventral striatum. Experimental Brain Research, 85, 491–500.
doi:10.1007/BF00231732
Asato, M. R., Terwilliger, R., Woo, J., & Luna, B. (2010). White matter development
in adolescents: A DTI study. Cerebral Cortex, 20, 2122–2131. doi:10.1093/cercor/
bhp282
Badanich, K. A., Adler, K. J., & Kirstein, C. L. (2006). Adolescents differ from
adults in cocaine conditioned place preference and cocaine-induced dopamine
in the nucleus accumbens septi. European Journal of Pharmacology, 550, 95–106.
doi:10.1016/j.ejphar.2006.08.034
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994). Insensitivity
to future consequences following damage to human prefrontal cortex. Cognition,
50, 7–15. doi:10.1016/0010-0277(94)90018-3
Benes, F. M., Taylor, J. B., & Cunningham, M. C. (2000). Convergence and plasticity
of monoaminergic systems in the medial prefrontal cortex during the postnatal

the adolescent sensation-seeking period      111

13490-05_Ch04-4thPgs.indd 111 11/15/13 1:43 PM

 period: Implications for the development of psychopathology. Cerebral Cortex,
10, 1014–1027. doi:10.1093/cercor/10.10.1014
Berns, G. S., Moore, S., & Capra, C. M. (2009). Adolescent engagement in dan-
gerous behaviors is associated with increased white matter maturity of frontal
cortex. PLoS ONE, 4, e6773. doi:10.1371/journal.pone.0006773
Bjork, J. M., Knutson, B., Fong, G. W., Caggiano, D. M., Bennett, S. M., & Hommer,
D. W. (2004). Incentive-elicited brain activation in adolescents: Similarities
and differences from young adults. The Journal of Neuroscience, 24, 1793–1802.
doi:10.1523/JNEUROSCI.4862-03.2004
Bjork, J. M., Lynne-Landsman, S. D., Sirocco, K., & Boyce, C. A. (2012). Brain
maturation and risky behavior: The promise and the challenges of neuro­
imaging-based accounts. Child Development Perspectives, 6, 385–391. doi:10.1111/
cdep.12001
Bjork, J. M., Smith, A. R., Chen, G., & Hommer, D. W. (2010). Adolescents, adults
and rewards: Comparing motivational neurocircuitry recruitment using fMRI.
PLoS ONE, 5, e11440. doi:10.1371/journal.pone.0011440
Bjorklund, D. F., & Harnishfeger, K. K. (1995). The evolution of inhibition mecha-
nisms and their role in human cognition and behavior. In F. N. Dempster &
C. J. Brainerd (Eds.), Interference & inhibition in cognition (pp. 141–173). San
Diego, CA: Academic Press. doi:10.1016/B978-012208930-5/50006-4
Blakemore, S. J. (2008). The social brain in adolescence. Nature Reviews Neuroscience,
9, 267–277. doi:10.1038/nrn2353
Breiter, H. C., Aharon, I., Kahneman, D., Dale, A., & Shizgal, P. (2001). Functional
imaging of neural responses to expectancy and experience of monetary gains
and losses. Neuron, 30, 619–639. doi:10.1016/S0896-6273(01)00303-8
Brenhouse, H. C., Sonntag, K. C., & Andersen, S. L. (2008). Transient D1 dopa-
mine receptor expression on prefrontal cortex projection neurons: Relationship
to enhanced motivational salience of drug cues in adolescence. The Journal of
Neuroscience, 28, 2375–2382. doi:10.1523/JNEUROSCI.5064-07.2008
Brown, M. R., Vilis, T., & Everling, S. (2007). Frontoparietal activation with prepa-
ration for antisaccades. Journal of Neurophysiology, 98, 1751–1762. doi:10.1152/
jn.00460.2007
Bunge, S. A., Dudukovic, N. M., Thomason, M. E., Vaidya, C. J., & Gabrieli,
J. D. (2002). Immature frontal lobe contributions to cognitive control in
children: Evidence from fMRI. Neuron, 33, 301–311. doi:10.1016/S0896-
6273(01)00583-9
Caviness, V. S., Kennedy, D. N., Bates, J. F., & Makris, N. (1996). The develop-
ing human brain: A morphometric profile. In R. W. Thatcher, G. R. Lyon,
J. Rumsey, & N. A. Krasnegor (Eds.), Developmental neuroimaging: Mapping the
development of brain and behavior (pp. 3–14). New York, NY: Academic Press.
Chambers, R. A., Taylor, J. R., & Petenza, M. N. (2003). Developmental neuro-
circuitry of motivation in adolescence: A critical period of addiction vulner-

112       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 112 11/15/13 1:43 PM

 ability. The American Journal of Psychiatry, 160, 1041–1052. doi:10.1176/appi.
ajp.160.6.1041
Chein, J., Albert, D., O’Brien, L., Uckert, K., & Steinberg, L. (2011). Peers increase
adolescent risk taking by enhancing activity in the brain’s reward circuitry.
Developmental Science, 14, F1–F10. doi:10.1111/j.1467-7687.2010.01035.x
Chikama, M., McFarland, N. R., Amaral, D. G., & Haber, S. N. (1997). Insular cor-
tical projections to functional regions of the striatum correlate with cortical
cytoarchitectonic organization in the primate. Journal of Neuroscience, 17(24),
9686–9705.
Christakou, A., Brammer, M., & Rubia, K. (2011). Maturation of limbic cortico-
striatal activation and connectivity associated with developmental changes
in temporal discounting. NeuroImage, 54, 1344–1354. doi:10.1016/j.neuro­
image.2010.08.067
Ciesielski, K. T., Lesnik, P. G., Savoy, R. L., Grant, E. P., & Ahlfors, S. P. (2006).
Developmental neural networks in children performing a Categorical N-Back
Task. NeuroImage, 33, 980–990. doi:10.1016/j.neuroimage.2006.07.028
Cohen, J. R., Asarnow, R. F., Sabb, F. W., Bilder, R. M., Bookheimer, S. Y., Knowlton,
B. J., & Poldrack, R. A. (2010). A unique adolescent response to reward predic-
tion errors. Nature Neuroscience, 13, 669–671. doi:10.1038/nn.2558
Cools, R. (2008). Role of dopamine in the motivational and cognitive control of
behavior. The Neuroscientist, 14, 381–395. doi:10.1177/1073858408317009
Cools, R. (2011). Dopaminergic control of the striatum for high-level cognition.
Current Opinion in Neurobiology, 21, 402–407. doi:10.1016/j.conb.2011.
04.002
Crone, E. A., Wendelken, C., Donohue, S., van Leijenhorst, L., & Bunge, S. A.
(2006). Neurocognitive development of the ability to manipulate information
in working memory. Proceedings of the National Academy of Sciences of the United
States of America, 103, 9315–9320. doi:10.1073/pnas.0510088103
Delgado, M. R., Locke, H. M., Stenger, V. A., & Fiez, J. A. (2003). Dorsal striatum
responses to reward and punishment: Effects of valence and magnitude manipu-
lations. Cognitive, Affective & Behavioral Neuroscience, 3, 27–38. doi:10.3758/
CABN.3.1.27
Delgado, M. R., Nystrom, L. E., Fissell, C., Noll, D. C., & Fiez, J. A. (2000). Tracking
the hemodynamic responses to reward and punishment in the striatum. Journal
of Neurophysiology, 84, 3072–3077.
Diamond, A., & Goldman-Rakic, P. S. (1989). Comparison of human infants and
rhesus monkeys on Piaget’s AB task: Evidence for dependence on dorsolat-
eral prefrontal cortex. Experimental Brain Research, 74, 24–40. doi:10.1007/
BF00248277
Di Martino, A., Scheres, A., Margulies, D. S., Kelly, A. M., Uddin, L. Q., Shehzad,
Z., . . . Milham, M. P. (2008). Functional connectivity of human striatum: A
resting state FMRI study. Cerebral Cortex, 18(12), 2735–2747.

the adolescent sensation-seeking period      113

13490-05_Ch04-4thPgs.indd 113 11/15/13 1:43 PM

 Elliott, R., Newman, J. L., Longe, O. A., & Deakin, J. F. (2003). Differential response
patterns in the striatum and orbitofrontal cortex to financial reward in humans:
A parametric functional magnetic resonance imaging study. The Journal of Neuro­
science, 23, 303–307.
Eluvathingal, T. J., Hasan, K. M., Kramer, L., Fletcher, J. M., & Ewing-Cobbs, L.
(2007). Quantitative diffusion tensor tractography of association and projection
fibers in normally developing children and adolescents. Cerebral Cortex, 17,
2760–2768. doi:10.1093/cercor/bhm003
Erickson, S. L., & Lewis, D. A. (2002). Postnatal development of parvalbumin- and
GABA transporter-immunoreactive axon terminals in monkey prefrontal cor-
tex. The Journal of Comparative Neurology, 448, 186–202. doi:10.1002/cne.10249
Ernst, M., Nelson, E. E., Jazbec, S., McClure, E. B., Monk, C. S., Leibenluft, E., . . . Pine,
D. S. (2005). Amygdala and nucleus accumbens in responses to receipt and omis-
sion of gains in adults and adolescents. NeuroImage, 25, 1279–1291. doi:10.1016/
j.neuroimage.2004.12.038
Everling, S., & Munoz, D. P. (2000). Neuronal correlates for preparatory set associ-
ated with pro-saccades and anti-saccades in the primate frontal eye field. The
Journal of Neuroscience, 20, 387–400.
Fair, D. A., Dosenbach, N. U., Church, J. A., Cohen, A. L., Brahmbhatt, S., Miezin,
F. M., . . . Schlaggar, B. L. (2007). Development of distinct control networks
through segregation and integration. Proceedings of the National Academy
of Sciences of the United States of America, 104, 13507–13512. doi:10.1073/
pnas.0705843104
Forbes, E. E., & Dahl, R. E. (2010). Pubertal development and behavior: Hormonal
activation of social and motivational tendencies. Brain and Cognition, 72,
66–72. doi:10.1016/j.bandc.2009.10.007
Fudge, J. L., Kinishio, K., Walsh, P., Richard, C., & Haber, S. N. (2002). Amygdaloid
projections to ventromedial striatal subterritories in the primate. Neuroscience,
110 (2), 257–275.
Fuster, J. M. (1989). The Prefrontal Cortex (Vol. 2). New York: Raven Press.
Fuster, J. M. (2002). Frontal lobe and cognitive development. Journal of Neurocytology,
31, 373–385. doi:10.1023/A:1024190429920
Galvan, A., Hare, T., Voss, H., Glover, G., & Casey, B. J. (2007). Risk-taking
and the adolescent brain: Who is at risk? Developmental Science, 10, F8–F14.
doi:10.1111/j.1467-7687.2006.00579.x
Galvan, A., Hare, T. A., Parra, C. E., Penn, J., Voss, H., Glover, G., & Casey, B. J.
(2006). Earlier development of the accumbens relative to orbitofrontal cortex
might underlie risk-taking behavior in adolescents. The Journal of Neuroscience,
26, 6885–6892. doi:10.1523/JNEUROSCI.1062-06.2006
Geier, C. F., Garver, K., Terwilliger, R., & Luna, B. (2009). Development of work-
ing memory maintenance. Journal of Neurophysiology, 101, 84–99. doi:10.1152/
jn.90562.2008

114       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 114 11/15/13 1:43 PM

 Geier, C. F., & Luna, B. (2009). The maturation of incentive processing and cogni-
tive control. Pharmacology, Biochemistry and Behavior, 93, 212–221. doi:10.1016/
j.pbb.2009.01.021
Geier, C. F., & Luna, B. (2012). Developmental effects of incentives on response inhibi-
tion. Child Development, 83, 1262–1274. doi:10.1111/j.1467-8624.2012.01771.x
Geier, C. F., Terwilliger, R., Teslovich, T., Velanova, K., & Luna, B. (2010). Imma-
turities in reward processing and its influence on inhibitory control in adoles-
cence. Cerebral Cortex, 20, 1613–1629. doi:10.1093/cercor/bhp225
Giedd, J. N., Blumenthal, J., Jeffries, N. O., Castellanos, F. X., Liu, H., Zijdenbos,
A., . . . Rappaport, J. L. (1999). Brain development during childhood and
adolescence: A longitudinal MRI study. Nature Neuroscience, 2, 861–863.
doi:10.1038/13158
Giorgio, A., Watkins, K. E., Douaud, G., James, A. C., James, S., De Stefano,
N., . . . Johansen-Berg, H. (2008). Changes in white matter microstructure dur-
ing adolescence. NeuroImage, 39, 52–61. doi:10.1016/j.neuroimage.2007.07.043
Gogtay, N., Giedd, J. N., Lusk, L., Hayashi, K. M., Greenstein, D., Vaituzis,
A. C., . . . Thompson, P. M. (2004). Dynamic mapping of human cortical
development during childhood through early adulthood. Proceedings of the
National Academy of Sciences of the United States of America, 101, 8174–8179.
doi:10.1073/pnas.0402680101
Goldman-Rakic, P. S. (1987). Circuitry of primate prefrontal cortex and regulation
of behavior by representational memory. In V. B. Mountcastle (Ed.), Handbook
of physiology, section 1. The nervous system, vol. v. Higher functions of the brain,
part 1 (pp. 373–417). Bethesda, MD: American Physiology Society.
Goldman-Rakic, P. S., & Brown, R. M. (1982). Postnatal development of mono-
amine content and synthesis in the cerebral cortex of rhesus monkeys. Brain
Research, 254, 339–349.
Graham v. Florida, 560 U.S. _____ (2010).
Haber, S. N., Kunishio, K., Mizobuchi, M., & Lynd-Balta, E. (1995). The orbital
and medial prefrontal circuit through the primate basal ganglia. The Journal of
Neuroscience, 15 (7 Pt 1), 4851–4867.
Hallett, P. E. (1978). Primary and secondary saccades to goals defined by instructions.
Vision Research, 18, 1279–1296. doi:10.1016/0042-6989(78)90218-3
Hare, T. A., O’Doherty, J., Camerer, C. F., Schultz, W., & Rangel, A. (2008). Disso-
ciating the role of the orbitofrontal cortex and the striatum in the computation
of goal values and prediction errors. The Journal of Neuroscience, 28, 5623–5630.
doi:10.1523/JNEUROSCI.1309-08.2008
Hedner, T., Iversen, K., & Lundborg, P. (1984). Central GABA mechanisms dur-
ing postnatal development in the rat: Neurochemical characteristics. Journal of
Neural Transmission, 59, 105–118. doi:10.1007/BF01255409
Heron, M. (2012). Deaths: Leading causes for 2008. National Vital Statistics Report,
6, 1–94.

the adolescent sensation-seeking period      115

13490-05_Ch04-4thPgs.indd 115 11/15/13 1:43 PM

 Hikosaka, O., Nakumura, K., & Nakahara, H. (2006). Basal ganglia orient eyes to
reward. Journal of Neurophysiology, 95, 567–584. doi:10.1152/jn.00458.2005
Huttenlocher, P. R. (1990). Morphometric study of human cerebral cortex develop-
ment. Neuropsychologia, 28, 517–527. doi:10.1016/0028-3932(90)90031-I
Huttenlocher, P. R., & Dabholkar, A. S. (1997). Regional differences in synapto-
genesis in human cerebral cortex. The Journal of Comparative Neurology, 387,
167–178. doi:10.1002/(SICI)1096-9861(19971020)387:2<167::AID-CNE1
>3.0.CO;2-Z
Hwang, K., Hallquist, M. N., & Luna, B. (2012). The development of hub architec-
ture in the human functional brain network. Cerebral Cortex. Advance online
publication. doi:10.1093/cercor/bhs227
Hwang, K., & Luna, B. (2011). The development of brain connectivity supporting
prefrontal cortical functions. In D. T. Stuss & R. T. Knight (Eds.), Principle of
frontal lobe functions (2nd ed.). New York, NY: Oxford University Press.
Hwang, K., Velanova, K., & Luna, B. (2010). Strengthening of top-down frontal cog-
nitive control networks underlying the development of inhibitory control: An
fMRI effective connectivity study. The Journal of Neuroscience, 30, 15535–15545.
doi:10.1523/JNEUROSCI.2825-10.2010
Johnson, M. H. (1995). The inhibition of automatic saccades in early infancy. Devel-
opmental Psychobiology, 28, 281–291. doi:10.1002/dev.420280504
Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Increased brain activity in
frontal and parietal cortex underlies the development of visuospatial working
memory capacity during childhood. Journal of Cognitive Neuroscience, 14, 1–10.
doi:10.1162/089892902317205276
Knutson, B., & Cooper, J. C. (2005). Functional magnetic resonance imaging of
reward prediction. Current Opinion in Neurology, 18, 411–417. doi:10.1097/01.
wco.0000173463.24758.f6
Knutson, B., Westdorp, A., Kaiser, E., & Hommer, D. (2000). FMRI visualization of
brain activity during a monetary incentive delay task. NeuroImage, 12, 20–27.
doi:10.1006/nimg.2000.0593
Kringelbach, M. L. (2005). The human orbitofrontal cortex: Linking reward to hedonic
experience. Nature Reviews Neuroscience, 6, 691–702. doi:10.1038/nrn1747
Lambe, E. K., Krimer, L. S., & Goldman-Rakic, P. S. (2000). Differential postnatal
development of catecholamine and serotonin inputs to identified neurons in
prefrontal cortex of rhesus monkey. The Journal of Neuroscience, 20, 8780–8787.
Laviola, G., Adriani, W., Terranova, M. L., & Gerra, G. (1999). Psychobiological
risk factors for vulnerability to psychostimulants in human adolescents and ani-
mal models. Neuroscience and Biobehavioral Reviews, 23, 993–1010. doi:10.1016/
S0149-7634(99)00032-9
Lebel, C., Walker, L., Leemans, A., Phillips, L., & Beaulieu, C. (2008). Microstruc-
tural maturation of the human brain from childhood to adulthood. NeuroImage,
40, 1044–1055. doi:10.1016/j.neuroimage.2007.12.053

116       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 116 11/15/13 1:43 PM

 Lewis, D. A. (1997). Development of the prefrontal cortex during adolescence:
Insights into vulnerable neural circuits in schizophrenia. Neuropsychopharma-
cology, 16, 385–398. doi:10.1016/S0893-133X(96)00277-1
Lewis, D. A., Melchitzky, D. S., & Burgos, G. G. (2002). Specificity in the functional
architecture of primate prefrontal cortex. Journal of Neurocytology, 31, 265–276.
doi:10.1023/A:1024174026286
Lidow, M. S., & Rakic, P. (1992). Scheduling of monoaminergic neurotransmitter
receptor expression in the primate neocortex during postnatal development.
Cerebral Cortex, 2, 401–416. doi:10.1093/cercor/2.5.401
Luciana, M., & Nelson, C. A. (1998). The functional emergence of prefrontally-
guided working memory systems in four- to eight-year-old children. Neuro­
psychologia, 36, 273–293. doi:10.1016/S0028-3932(97)00109-7
Luciana, M., Wahlstrom, D., Porter, J. N., & Collins, P. F. (2012). Dopaminergic
modulation of incentive motivation in adolescence: Age-related changes in
signaling, individual differences, and implications for the development of self-
regulation. Developmental Psychology, 48, 844–861. doi:10.1037/a0027432
Luna, B., Thulborn, K. R., Munoz, D. P., Merriam, E. P., Garver, K. E., Minshew,
N. J., . . . Sweeney, J. A. (2001). Maturation of widely distributed brain func-
tion subserves cognitive development. NeuroImage, 13, 786–793. doi:10.1006/
nimg.2000.0743
Mathews, I. Z., & McCormick, C. M. (2007). Female and male rats in late adoles-
cence differ from adults in amphetamine-induced locomotor activity, but not in
conditioned place preference for amphetamine. Behavioural Pharmacology, 18,
641–650. doi:10.1097/FBP.0b013e3282effbf5
McClure, S. M., York, M. K., & Montague, P. R. (2004). The neural substrates of
reward processing in humans: The modern role of fMRI. The Neuroscientist, 10,
260–268. doi:10.1177/1073858404263526
McCutcheon, J. E., White, F. J., & Marinelli, M. (2009). Individual differences in
dopamine cell neuroadaptations following cocaine self-administration. Biologi-
cal Psychiatry, 66, 801–803. doi:10.1016/j.biopsych.2009.04.018
Miller v. Alabama, 567 U.S. ­____ (2012).
Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial
behavior: A developmental taxonomy. Psychological Review, 100, 674–701.
doi:10.1037/0033-295X.100.4.674
Montague, D. M., Lawler, C. P., Mailman, R. B., & Gilmore, J. H. (1999). Devel-
opmental regulation of the dopamine D1 receptor in human caudate and
putamen. Neuropsychopharmacology, 21, 641–649. doi:10.1016/S0893-
133X(99)00062-7
Nelson, C. A., Monk, C. S., Lin, J., Carver, L. J., Thomas, K. M., & Truwitt, C. L.
(2000). Functional neuroanatomy of spatial working memory in children.
Developmental Psychology, 36, 109–116. doi:10.1037/0012-1649.36.1.109

the adolescent sensation-seeking period      117

13490-05_Ch04-4thPgs.indd 117 11/15/13 1:43 PM

 O’Doherty, J. P. (2004). Reward representations and reward-related learning in the
human brain: Insights from neuroimaging. Current Opinion in Neurobiology, 14,
769–776. doi:10.1016/j.conb.2004.10.016
Olesen, P. J., Macoveanu, J., Tegner, J., & Klingberg, T. (2007). Brain activity related
to working memory and distraction in children and adults. Cerebral Cortex, 17,
1047–1054. doi:10.1093/cercor/bhl014
Op de Macks, Z. A., Gunther Moor, B.. Overgaauw, S., Guroglu, B., Dahl, R. E., &
Crone, E. A. (2011). Testosterone levels correspond with increased ventral stri-
atum activation in response to monetary rewards in adolescents. Developmental
Cognitive Neuroscience, 1, 517–529. doi:10.1016/j.dcn.2011.06.003
Padmanabhan, A., Geier, C. F., Ordaz, S. J., Teslovich, T., & Luna, B. (2011). Devel-
opmental changes in brain function underlying the influence of reward process-
ing on inhibitory control. Developmental Cognitive Neuroscience, 1, 517–529.
doi:10.1016/j.dcn.2011.06.004
Paulsen, D. J., Platt, M. L., Huettel, S. A., & Brannon, E. M. (2011). Decision-making
under risk in children, adolescents, and young adults. Frontiers in Psychology, 2(72).
doi:10.3389/fpsyg.2011.00072
Reyna, V. F., & Brainerd, C. J. (2011). Dual processes in decision making and develop-
mental neuroscience: A fuzzy-trace model. Developmental Review, 31, 180–206.
doi:10.1016/j.dr.2011.07.004
Roesch, M. R., & Olson, C. R. (2003). Impact of expected reward on neuronal activ-
ity in prefrontal cortex, frontal and supplementary eye fields and premotor cor-
tex. Journal of Neurophysiology, 90, 1766–1789. doi:10.1152/jn.00019.2003
Roesch, M. R., & Olson, C. R. (2004). Neuronal activity related to reward value
and motivation in primate frontal cortex. Science, 304, 307–310. doi:10.1126/
science.1093223
Roper v. Simmons, 543 U.S. 551 (2005).
Rosenberg, D. R., & Lewis, D. A. (1994). Changes in the dopaminergic innerva-
tion of monkey prefrontal cortex during late postnatal development: A tyrosine
hydroxylase immunohistochemical study. Biological Psychiatry, 36, 272–277.
doi:10.1016/0006-3223(94)90610-6
Rosenberg, D. R., & Lewis, D. A. (1995). Postnatal maturation of the dopaminergic
innervation of monkey prefrontal and motor cortices: A tyrosine hydroxylase
immunohistochemical analysis. The Journal of Comparative Neurology, 358,
383–400. doi:10.1002/cne.903580306
Rubia, K., Smith, A. B., Woolley, J., Nosarti, C., Heyman, I., Taylor, E., & Brammer,
M. (2006). Progressive increase of frontostriatal brain activation from child-
hood to adulthood during event-related tasks of cognitive control. Human Brain
Mapping, 27, 973–993. doi:10.1002/hbm.20237
Rushworth, M. F., Noonan, M. P., Boorman, E. D., Walton, M. E., & Behrens, T. E.
(2011). Frontal cortex and reward-guided learning and decision-making. Neu-

118       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 118 11/15/13 1:43 PM

 ron, 70(6), 1054–1069. doi: S0896-6273(11)00395-3[pii]10.1016/j.neuron.
2011.05.014
Scherf, K. S., Sweeney, J. A., & Luna, B. (2006). Brain basis of developmental
change in visuospatial working memory. Journal of Cognitive Neuroscience, 18,
1045–1058. doi:10.1162/jocn.2006.18.7.1045
Schultz, W. (2000). Multiple reward signals in the brain. Nature Reviews Neuroscience,
1, 199–207. doi:10.1038/35044563
Schultz, W., Preuschoff, K., Camerer, C., Hsu, M., Fiorillo, C. D., Tobler, P. N., &
Bossaerts, T. (2008). Explicit neural signals reflecting reward uncertainty. Philo-
sophical Transactions of the Royal Society of London. Series B, Biological Sciences,
363, 3801–3811. doi:10.1098/rstb.2008.0152
Scott, E. S., & Steinberg, L. (2008). Rethinking juvenile justice. Boston, MA: Harvard
University Press.
Seeman, P., Bzowej, N. H., Guan, H. C., Bergeron, C., Becker, L. E., Reynolds,
G. P., . . . Tourtellotte, W. W. (1987). Human brain dopamine receptors in
children and aging adults. Synapse, 1(5), 399–404. doi:10.1002/syn.890010503
Selemon, L. D., & Goldman-Rakic, P. S. (1985). Longitudinal topography and inter-
digitation of corticostriatal projections in the Rhesus monkey. Journal of Neuro­
science, 5, 776–794.
Shad, M. U., Bidesi, A. P., Chen, L. A., Ernst, M., & Rao, U. (2011). Neurobiology
of decision making in depressed adolescents: A functional magnetic resonance
imaging study. Journal of the American Academy of Child & Adolescent Psychiatry,
50, 612–621. doi:10.1016/j.jaac.2011.03.011
Smith, A. B., Halari, R., Giampetro, V., Brammer, M., & Rubia, K. (2011). Devel-
opmental effects of reward on sustained attention networks. NeuroImage, 56,
1693–1704. doi:10.1016/j.neuroimage.2011.01.072
Somerville, L. H., & Casey, B. (2010). Developmental neurobiology of cognitive con-
trol and motivational systems. Current Opinion in Neurobiology, 20, 236–241.
doi:10.1016/j.conb.2010.01.006
Sowell, E. R., Thompson, P. M., Holmes, C. J., Jernigan, T. L., & Toga, A. W. (1999).
In vivo evidence for post-adolescent brain maturation in frontal and striatal
regions. Nature Neuroscience, 2, 859–861. doi:10.1038/13154
Spear, L. P. (2000). The adolescent brain and age-related behavioral manifesta-
tions. Neuroscience & Biobehavioral Reviews, 24, 417–463. doi:10.1016/S0149-
7634(00)00014-2
Spear, L. P. (2010). The behavioral neuroscience of adolescence. New York, NY: Norton.
Spear, L. P., & Brake, S. C. (1983). Periadolescence: Age-dependent behavior and
psychopharmacological responsivity in rats. Developmental Psychobiology, 16,
83–109. doi:10.1002/dev.420160203
Spear, L. P., Shalaby, I. A., & Brick, J. (1980). Chronic administration of haloperidol
during development: Behavioral and psychopharmacological effects. Psycho­
pharmacology, 70, 47–58. doi:10.1007/BF00432369

the adolescent sensation-seeking period      119

13490-05_Ch04-4thPgs.indd 119 11/15/13 1:43 PM

 Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., & Woodlard, J.
(2008). Age differences in sensation seeking and impulsivity as indexed by
behavior and self-report: Evidence for a dual systems model. Developmental Psy-
chology, 44, 1764–1778. doi:10.1037/a0012955
Stevens, M. C., Kiehl, K. A., Pearlson, G. D., & Calhoun, V. D. (2007). Functional
neural networks underlying response inhibition in adolescents and adults.
Behavioural Brain Research, 181, 12–22. doi:10.1016/j.bbr.2007.03.023
Thomas, K. M., King, S. W., Franzen, P. L., Welsh, T. F., Berkowitz, A. L., Noll,
D. C., . . . Casey, B. J. (1999). A developmental functional MRI study of
spatial working memory. NeuroImage, 10, 327–338. doi:10.1006/nimg.1999.
0466
Thut, G., Schultz, W., Roelcke, U., Nienhusmeier, M., Missimer, J., Maguire,
R. P., . . . Leenders, K. L. (1997). Activation of the human brain by mon-
etary reward. Neuroreport, 8, 1225–1228. doi:10.1097/00001756-199703240-
00033
Tseng, K. Y., & O’Donnell, P. (2005). Post-pubertal emergence of prefrontal corti-
cal up states induced by D1-NMDA co-activation. Cerebral Cortex, 15, 49–57.
doi:10.1093/cercor/bhh107
van Leijenhorst, L., Crone, E. A., & Bunge, S. A. (2006). Neural correlates of devel-
opmental differences in risk estimation and feedback processing. Neuropsycholo-
gia, 44, 2158–2170. doi:10.1016/j.neuropsychologia.2006.02.002
van Leijenhorst, L., Gunther Moor, B., Op de Macks, Z. A., Rombouts, S. A.,
Westenberg, P. M., & Crone, E. A. (2010). Adolescent risky decision-making:
Neurocognitive development of reward and control regions. NeuroImage, 51,
345–355. doi:10.1016/j.neuroimage.2010.02.038
van Leijenhorst, L., Zanolie, K., Van Meel, C. S., Westenberg, P. M., Rombouts,
S. A., & Crone, E. A. (2010). What motivates the adolescent? Brain regions
mediating reward sensitivity across adolescence. Cerebral Cortex, 20(1), 61—69.
doi:10.1093/cercor/bhp078
Velanova, K., Wheeler, M. E., & Luna, B. (2008). Maturational changes in anterior
cingulate and frontoparietal recruitment support the development of error process-
ing and inhibitory control. Cerebral Cortex, 18, 2505–2522. doi:10.1093/cercor/
bhn012
Velanova, K., Wheeler, M. E., & Luna, B. (2009). The maturation of task set–related
activation supports late developmental improvements in inhibitory control.
The Journal of Neuroscience, 29, 12558–12567. doi:10.1523/JNEUROSCI.
1579-09.2009
Wahlstrom, D., Collins, P., White, T., & Luciana, M. (2010). Developmental
changes in dopamine neurotransmission in adolescence: Behavioral implica-
tions and issues in assessment. Brain and Cognition, 72, 146–159. doi:10.1016/
j.bandc.2009.10.013

120       luna, padmanabhan, and geier

13490-05_Ch04-4thPgs.indd 120 11/15/13 1:43 PM

 Wahlstrom, D., White, T., & Luciana, M. (2010). Neurobehavioral evidence for
changes in dopamine system activity during adolescence. Neuroscience and
Biobehavioral Reviews, 34, 631–648. doi:10.1016/j.neubiorev.2009.12.007
Wise, R. A. (2002). Brain reward circuitry: Insights from unsensed incentives. Neu-
ron, 36, 229–240. doi:10.1016/S0896-6273(02)00965-0
Wise, R. A. (2004). Dopamine, learning and motivation. Nature Reviews Neuro­
science, 5, 483–494. doi:10.1038/nrn1406
Yakovlev, P. I., & Lecours, A. R. (1967). The myelogenetic cycles of regional matura-
tion of the brain. In A. Minkowski (Ed.), Regional development of the brain in early
life (pp. 3–70). Oxford, England: Blackwell Scientific.

the adolescent sensation-seeking period      121

13490-05_Ch04-4thPgs.indd 121 11/15/13 1:43 PM

13490-05_Ch04-4thPgs.indd 122 11/15/13 1:43 PM
 5
Reward Processing
and Risky Decision Making
in the Aging Brain
Gregory R. Samanez-Larkin and Brian Knutson

This chapter reviews studies that have examined how age influences psycho-
logical and neural responses to financial incentives and risk. The findings
suggest that although processing of basic rewards may be maintained across
the adult life span, learning about new rewards may decline as a function of
age. These behavioral changes can be linked to relative preservation of stria-
tal function in the face of age-related declines in the connectivity of the pre-
frontal cortex to the striatum. This frontostriatal disconnection may impair
risky decision making, both in the laboratory and in the real world. These
novel findings inform theory about how affect and cognition combine to guide

G.R.S.L. is currently supported by a Pathway to Independence Award (K99-AG042596) from the

National Institute on Aging. Much of our own work that is discussed in this chapter was supported
by grants from the National Institute on Aging (R21-AG030778 to B.K.; F31-AG032804 and
F32-AG039131 to G.R.S.L.) and the FINRA Investor Education Foundation. A portion of this chapter
is from Incentive Processing in the Aging Brain: Individual Differences in Value-Based Learning and Decision
Making Across the Adult Life Span (Doctoral Dissertation, Stanford University). Retrieved from http://
purl.stanford.edu/vy834vv5149. Adapted with permission.
http://dx.doi.org/10.1037/14322-006
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

123

13490-06_Ch05-3rdPgs.indd 123 11/15/13 1:43 PM

 choice, and they imply that a deeper understanding of how the aging brain
processes incentives may eventually inform the design of more targeted and
effective decision aids for individuals of all ages.
As global demographics shift toward an increase in the proportion of
older adults in the population, older adults will control correspondingly more
global resources. Will they allocate those resources in the same way they
might have earlier in life, or will the very process of aging influence the man-
ner in which financial decisions are approached and negotiated? Although
stereotypes about the effects of aging on financial risk taking abound (such
as the notion that risk aversion globally increases with age), relatively little
evidence has been collected either to verify or refute these commonly held
beliefs. Even less is known about what psychological and neural mechanisms
might underlie age differences in risky decision making. In this chapter, we
review the emerging literature on the influence of aging on incentive process-
ing and financial risk taking, correlated neural substrates, and generalizations
to real-world behavior, with a goal of identifying fruitful directions for future
research.
Multiple psychological factors, both cognitive and affective, contrib-
ute to decision making. An extensive body of research suggests that fluid
cognitive capacities such as attention, working memory, and executive con-
trol decline with age (Salthouse, 2004), while leaving crystallized cognitive
capacities relatively spared (Park et al., 2002). Neuroscience research has
linked declines in fluid cognitive capacity to changes in medial temporal and
lateral prefrontal cortical function (Hedden & Gabrieli, 2004). In contrast to
these fluid cognitive declines, affective processing remains intact in old age,
and emerging neuroscience research has begun to link these affective vari-
ables to changes in striatal, insular, and prefrontal function (Samanez-Larkin
& Carstensen, 2011). Yet, few studies have investigated how age-related
changes in cognition, affect, or associated neural circuits influence incentive
processing and risky decision making (Samanez-Larkin, 2010).
Over the past several years, we have conducted a series of studies in an
initial attempt to examine individual differences in reward processing across
the adult life span using an interdisciplinary and translational approach.
This approach combines psychological theory, imaging methods from neuro­
science, and experimental tasks from behavioral economics and finance to
examine decision making in the laboratory and in the real world. Overall,
the results reveal a pattern of age differences in the function of neural sys-
tems that support aspects of incentive processing and risky decision making.
These studies and related work from other laboratories have explored indi-
vidual differences across a range of reward-related tasks ranging from basic
anticipatory and consummatory responses to incentive cues, to probabilistic
reward learning, to risky decision making about investments.

124       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 124 11/15/13 1:43 PM

 Processing of Monetary Gains and Losses

The development of event-related functional magnetic resonance

imaging (fMRI) in the early 1990s allowed investigators to visualize second-
to-second changes in cortical and subcortical brain function, illuminating
neural responses not only in response to events but also during anticipation
of those events. Coupled with calibrated tasks involving monetary incen-
tives, researchers could for the first time visualize subcortical responses dur-
ing anticipation of uncertain gains and losses (Knutson & Cooper, 2005).
These experiments were recently extended to studies of human aging. These
studies suggested that older adults showed similar subcortical ventral stria-
tal responses similar to those of younger adults when anticipating monetary
gains (e.g., in the context of cued reaction time tasks such as the monetary
incentive delay task; Knutson, Adams, Fong, & Hommer, 2001; Knutson &
Greer, 2008; Samanez-Larkin et al., 2007; see Color Plate 5). Younger and
older adults also showed similar neural responses to reward outcomes in the
medial prefrontal cortex and ventral striatum (Cox, Aizenstein, & Fiez, 2008;
Samanez-Larkin et al., 2007; Samanez-Larkin, Kuhnen, Yoo, & Knutson,
2010; Color Plate 5). Together, these findings provided initial evidence that
basic neural responses to the anticipation and receipt of monetary gains are
relatively preserved from young adulthood to old age.
A strikingly different pattern, however, emerged during anticipation
of monetary losses. Specifically, older adults showed reduced reactivity in
the dorsal striatum and anterior insula than younger adults while anticipat-
ing monetary loss (Samanez-Larkin et al., 2007). This age difference was
also reflected in self-reported affect, such that older adults reported lower
levels of anticipatory negative arousal when anticipating losses than did
younger adults. This diminished negative arousal during loss anticipation
for older adults replicated in an independent sample (Nielsen, Knutson, &
Carstensen, 2008). Interestingly, the age differences did not extend to loss
outcomes—when older adults actually lost money, they showed neural and
affective reactions to loss outcomes similar to those of younger adults (see
Color Plate 6).
This age-related asymmetry in gain and loss anticipation coheres with a
large body of behavioral research suggesting an age-related “positivity” effect
(Carstensen & Mikels, 2005; Mather & Carstensen, 2005). Socioemotional
selectivity theory (Carstensen, 2006; Carstensen & Mikels, 2005) posits that
as time horizons shrink across adulthood, humans are increasingly motivated
to optimize well-being. In fact, cross-sectional and longitudinal experience
sampling research has repeatedly demonstrated that emotional experience in
everyday life becomes less negative across adulthood (Carstensen, Pasupathi,
Mayr, & Nesselroade, 2000; Carstensen et al., 2011). A similar age by valence

reward, risk, and aging      125

13490-06_Ch05-3rdPgs.indd 125 11/15/13 1:43 PM

 interaction emerges in studies of attention and memory (Carstensen &
Mikels, 2005; Mather & Carstensen, 2005) and appears to extend to incen-
tive anticipation. By extension, diminished negativity during anticipation
of losses (but not in response to loss outcomes) may result from older adults
avoiding anxiety associated with potential losses unless they actually occur
(however, see Wood, Busemeyer, Koling, Cox, & Davis, 2005, for evidence
that positivity effects may sometimes extend to outcomes). Whether this
suppression of anticipatory anxiety is strategic or more automatic remains
an open question. Alternatively, regulation may grow more automatic and
require less effort with age (Samanez-Larkin & Carstensen, 2011).
Reduced loss anticipation may enhance the well-being of older adults
and may also contribute to age differences in some, but not all, decisions. In
fact, a recent study found similarly reduced insula activation in older com-
pared to younger adults when they were offered unfair divisions of money in
an ultimatum game (Harlé & Sanfey, 2012). Older adults showed reduced
insular activity (previously associated with heightened negative affect) but
still rejected more unfair offers than did younger adults. Thus, it remains
unclear whether reduced insular responsiveness influences affective experi-
ence only or whether it also directly alters learning and decision making. In
the monetary incentive delay paradigms mentioned above, performance is
typically controlled via adaptive target durations, so that all subjects succeed
(or “hit”) on approximately two thirds of trials. Thus changes in loss antici-
pation have no consequence for task performance. It is possible that reduced
neural anticipation of loss in old age only emerges when loss anticipation is
not required for performance—a possibility we return to later.
In contrast to research suggesting that neural responses to monetary
gains are relatively preserved over adulthood, another early study of adult age
differences in reward processing drew the opposite conclusion: older adults
show reduced neural anticipation of monetary gains and so are less motivated
than younger adults (Schott et al., 2007). A number of important differences
between these studies may account for their differential findings. First, Schott
et al. (2007) used very small magnitude rewards (70.50) on each trial. They
reported an age difference such that ventral striatal signals in older adults
did not distinguish between 70.50 and 70.00 during anticipation. In fact, a
similar effect is observed in Samanez-Larkin et al., 2007 (see the $0.50 and
$0.00 lines in Color Plate 5.). However, in the Samanez-Larkin et al. (2007)
study, the $5.00 cues elicited high levels of neural activity in both age groups.
Thus, it is possible that low-magnitude rewards were not large enough to
elicit strong neural responses. In other studies, low-magnitude rewards pro-
duce striatal signals with much lower levels of test–retest reliability than elic-
ited by higher-magnitude rewards (Wu, Samanez-Larkin, & Knutson, 2012).
Second, Schott et al. used symbolic cues that did not explicitly communicate

126       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 126 11/15/13 1:43 PM

 reward magnitude on each trial during anticipation. Associations between
the cue shapes and reward value were verbally communicated to subjects
during the task instructions, but, at some level, the association between these
symbolic cues and their value had to be learned. Older adults may not have
encoded these cue–reward associations as strongly as younger adults. Such
an alternative account predicts that the older adults should have had lower
levels of striatal signal during the cue phase (anticipation) but higher levels
of striatal signal during the outcome phase (due to a larger reward prediction
error)—which exactly matches the observed pattern of findings. Thus, the
study may reveal more about the influence of aging on reward learning than
about basic incentive processing and motivation. Accordingly, we now turn
to studies that explicitly focus on adult age differences in reward learning.

Gain and Loss Learning

In contrast to the studies suggesting relative preservation of basic reward

processing across adulthood, the literature on reward learning reveals more
consistent age differences in performance and associated neural activity. For
instance, older adults show reduced ventral striatal activation during probabi-
listic reward learning in fMRI studies (Mell et al., 2009). Electrophysiological
studies have similarly shown reduced activity in frontal cortical regions dur-
ing probabilistic reward learning in older adults (Eppinger, Hämmerer, & Li,
2011; Eppinger, Kray, Mock, & Mecklinger, 2008; Hämmerer, Li, Müller, &
Lindenberger, 2011). Some have suggested that these age differences may
be due to older adults’ difficulty in dynamically computing prediction errors
in novel environments (Eppinger et al., 2011), and some evidence exists to
support this account. Specifically, in fMRI studies, striatal activity appears
functionally intact in older adults during simple reward-based tasks that do
not require novel learning, but more variable and less associated with reward
prediction error signals (Samanez-Larkin, 2010) during tasks that require
learning in the same individuals and brain regions. These variable responses
appear to generalize to some risk-taking tasks, as discussed in the next section
on risky decisions.
Recently, theorists have begun to debate whether learning about gains
versus losses differentially changes with age. Some evidence suggests that older
adults are more sensitive to positive than to negative feedback during probabi-
listic learning than are younger adults (Denburg, Recknor, Bechara, & Tranel,
2006; Wood et al., 2005). Others have suggested the opposite: that older adults
are more sensitive to negative than to positive feedback (Hämmerer et al.,
2011). However, one study often cited as providing evidence for increases
in learning from negative feedback with age (Frank & Kong, 2008) did not

reward, risk, and aging      127

13490-06_Ch05-3rdPgs.indd 127 11/15/13 1:43 PM

 include a younger adult group. Therefore, those findings cannot provide evi-
dence for an age by valence interaction across adulthood. Instead, they sug-
gest that a subset of older adults show a stronger bias toward learning from
losses, whereas late middle-aged and young-older adults did not show the
same bias. In contrast, healthy young adults and healthy older adults show
no valence bias, consistent with an absence of an age by valence interaction
in reward learning over much of the adult life span (Frank, Seeberger, &
O’Reilly, 2004; Lighthall, Gorlick, Schoeke, Frank, & Mather, 2013). If a
shift toward negativity occurs, it appears to occur much later in old age (e.g.,
after age 80; Frank & Kong, 2008; J. R. Simon, Howard, & Howard, 2010),
when investigators must be careful to control for potential confounds related
to declining health. A slight nonlinear increase in negativity near the end of
life is in fact consistent with the larger literature on emotional experience in
everyday life (Carstensen et al., 2011).
Overall, across probabilistic reward learning tasks, investigators have
most consistently reported main effects of age without qualification by valence
(Eppinger et al., 2011). By implication, the majority of age differences in
reward learning tasks may relate to more general difficulties with probabilis-
tic learning. In probabilistic learning tasks, the uncertainty of the outcome
on any individual trial may introduce risk. Might these age-related learning
impairments also contribute to adult age differences in risky decision mak-
ing? The tasks described thus far did not systematically vary the level of risk
among options or include certain alternatives, but recent studies have begun
to explore age differences in choice among risky and safe options.

Risky Decision Making

Risky decisions involve choices between options in which at least one

alternative has an uncertain outcome. To test societal stereotypes of older
adults as being more risk averse than younger adults, a long history of behav-
ioral research has focused on aging and risk aversion, beginning with stud-
ies of cautiousness in responding (Botwinick, 1969; Calhoun & Hutchison,
1981; Okun, 1976). A critical review of this work suggests that stereotypes of
age-related risk aversion are not supported by findings from well-controlled
experimental tasks (Mather, 2006). The past several decades of this research
were recently reviewed in a meta-analysis. The meta-analysis, which focused
on gambling tasks and risky investment choices, found no evidence for sys-
tematic adult age differences in risk preferences (Mata, Josef, Samanez-Larkin,
& Hertwig, 2011). Rather, the findings identified a subset of tasks in which
older adults choose to avoid risk, but other tasks in which older adults choose

128       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 128 11/15/13 1:43 PM

 to seek risk more often than younger adults (Mata et al., 2011). Notably, in
many of these tasks, the expression of this “risk preference” is opposite of the
reward maximizing strategy. Thus, in these tasks, apparent age differences in
risk preferences may instead result from cognitive limitations (Henninger,
Madden, & Huettel, 2010). Consistent with this account, tasks that require
subjects to learn from recent experience show larger age differences in per-
formance than do tasks in which performance does not depend on learning
(Mata et al., 2011).
Few neuroimaging studies involving choices between high- and low-
risk options have compared younger and older adults. In an early study,
which included a risky decision making task that does not require learning,
older adults behaviorally chose more low-risk options and showed greater
insula activation during choice (Lee, Leung, Fox, Gao, & Chan, 2008).
However, one important limitation of this study was that the older adult
group included only nine subjects. The small sample size makes it difficult to
evaluate the generalizability of either the behavior or brain activity observed
in the older sample, especially in light of the results of the meta-analysis
described above. Also, insula activation showed high variability between
subjects—particularly in older adults. The unequal variance between groups
implies strong individual differences in the older group, which may violate
assumptions of conventional statistical comparisons. Thus, wide generaliza-
tions based on these findings may be premature.
Another study of adult age differences in risky decision making used a
different task that elicits a mix of both low-risk and high-risk choices within
each individual across the task (Kuhnen & Knutson, 2005). This investment
task (the BIAS Task) was designed to emulate everyday financial decisions
by requiring subjects to make a series of choices between “safe” bonds and
“risky” stocks. Reward maximization requires rapid learning from probabilis-
tic feedback throughout the task. Given the task structure, a “rational actor”
(i.e., Bayesian updating, risk-neutral agent) should begin each task round by
choosing the safe asset (bond) and then should shift over time to a risky asset
(stock) when the expected value of choosing that risky asset exceeds the fixed
expected value of the safe asset. Individual choices that deviate from those
of the rational actor can be classified as risk averse (i.e., the choice of a bond
over a better stock) or risk seeking (i.e., the choice of an uncertain stock over
a bond or alternative stock). In a large community sample of young, middle-
aged, and older adults, we found no age differences in risk aversion but did
find age differences in risk seeking (see Color Plate 7a; Samanez-Larkin
et al., 2010). Age differences specifically surfaced during trials in which
older adults chose risky assets—a pattern that replicated in two other groups
who participated in the task outside the MRI scanner (Samanez-Larkin et

reward, risk, and aging      129

13490-06_Ch05-3rdPgs.indd 129 11/15/13 1:43 PM

 al., 2010; Samanez-Larkin, Wagner, & Knutson, 2011). Consistent with
a neurocomputational account (S.-C. Li, 2005; S.-C. Li, Lindenberger, &
Sikström, 2001), age differences in risky choice were mediated by a neural
measure of functional variability in ventral striatal activity (see Color Plate
7b; Samanez-Larkin et al., 2010). This neural variability1 increased with age
in the midbrain and striatum, and the age-related variability effects replicated
in an entirely independent study which used a different task that did not
involve reward (Garrett, Kovacevic, McIntosh, & Grady, 2010).
The study suggests that variability in neural signals that compute and
represent expected value in a dynamic environment may increase with age.
Consistent with this account, related behavioral evidence suggests that
older adults have more difficulty estimating the specific value of ambiguous
stimuli during reward learning tasks (e.g., whether a nonrewarded outcome
should be weighted positively or negatively following a gain or loss cue;
Eppinger & Kray, 2011). Together, these findings suggest that apparent age
differences in risk preference may instead relate to differences in learning
ability. In support of this conclusion, neuroimaging studies of decisions that
do not require rapid learning from recent experience show similar patterns of
neural activity in striatal (Samanez-Larkin, Mata, et al., 2011) and prefron-
tal (Hosseini et al., 2010) regions of both younger and older adults.

Resolving an Apparent Paradox

The combined findings on basic reward processing, reward learning, and

risky decision making raises a puzzling question. How can the same striatal
regions that appear to be functionally intact in basic reward tasks show func-
tional irregularities in learning and decision tasks? One solution to this appar-
ent contradiction may be that a broader neural network lies at the source of
the age differences in striatal activity during learning and decision making.
For instance, while basic striatal function may remain preserved over the
adult life span, prefrontal input may change. Functionally, such a discon-
nection might not impair basic motivation but may instead misdirect that
motivation away from appropriate but novel goals (Knutson, Fong, Bennett,
Adams, & Hommer, 2003).
The highly interconnected anatomy of reward circuitry has been
extensively characterized in both human and nonhuman primates (Cohen,
Schoene-Bake, Elger, & Weber, 2009; Draganski et al., 2008; Haber, 2003;

1Note that the vast majority of fMRI studies compare mean signal between different task conditions.
Signal variability may be an important, overlooked individual difference measure relevant to under-
standing age differences in brain function.

130       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 130 11/15/13 1:43 PM

 Haber & Knutson, 2010). If connections from the medial prefrontal cor-
tex to the ventral striatum are compromised in aging, they might influence
reward learning and risk taking. The recent development of diffusion ten-
sor imaging tractography allows investigators to visualize the “integrity” (a
combination of fiber density, axonal diameter, and myelination) of even
subcortical tracts such as those implicated in reward processing. In a recent
study, we examined the structural integrity of reward circuitry using diffu-
sion tensor imaging in a group of younger, middle-aged, and older adults
who also completed a probabilistic reward learning task (Samanez-Larkin,
Levens, Perry, Dougherty, & Knutson, 2012). Analyses focused on a ventro-
medial circuit from the ventral tegmental area in the midbrain to the ventral
striatum through the pallidum to the thalamus, from the thalamus to the
medial prefrontal cortex, and from the frontal cortex back into the ventral
striatum. Findings specifically indicated that the integrity of pathways con-
necting the thalamus to the prefrontal cortex and the prefrontal cortex to
the ventral striatum was reduced in older age and was also associated with
reward learning (see Color Plate 8; Samanez-Larkin et al., 2012). These
findings suggest that one source of the age-related striatal functional vari-
ability observed in prior studies may arise from structural degradation of
prefrontal input to the striatum, which might compromise reward learning.
Interestingly, these frontostriatal pathways primarily utilize the neurotrans-
mitter glutamate. While the vast majority of research and theory about age
differences in neurochemical signaling has focused on dopamine (Braver
& Barch, 2002; S.-C. Li, Lindenberger, & Bäckman, 2010), these findings
emphasize the importance of also considering interactions with other neuro-
chemical systems (Mora, Segovia, & Del Arco, 2007; Segovia, Porras, Del
Arco, & Mora, 2001). Notably, age differences in reward learning are dis-
tinct from traditional impairments in explicit memory—in terms of both
associated neural circuits (i.e., ventromedial frontostriatal networks versus
lateral frontal and medial temporal networks) and psychological processes
(e.g., feedback-driven learning versus working memory; Samanez-Larkin,
Wagner, & Knutson, 2011).
In summary, apparently contradictory findings may reflect age differ-
ences in broader prefrontal networks associated with increased task demands
rather than more general deficits in striatal activity associated with moti-
vation. Thus, what appear to be motivational deficits may instead result
from cognitive deficits. Although some have claimed that basic motiva-
tional function declines with age (Eppinger, Nystrom, & Cohen, 2012), this
claim is inconsistent with decades of behavioral research on the psychology
of aging (Carstensen, 2006; Carstensen, Mikels, & Mather, 2005; Charles
& Carstensen, 2010) as well as a growing number of neuroscience studies
(Samanez-Larkin, 2011; Samanez-Larkin & Carstensen, 2011).

reward, risk, and aging      131

13490-06_Ch05-3rdPgs.indd 131 11/15/13 1:43 PM

 Individual Differences in Old Age

Studies of adult age differences in risky decision making often find

that individual differences increase with age (Eppinger & Kray, 2011). Even
in tasks in which decision making appears to decline on average with age,
many older individuals show no deficits whatsoever compared with healthy
young adults. For instance, using the Iowa gambling task (Denburg, Tranel, &
Bechara, 2005), researchers found that although some older individuals took
more risks and made fewer returns than younger adults, others performed as
well as young adults. These findings raise at least two possibilities: (a) either
a subset of older adults retains function from young adulthood to old age, or
(b) changes in motivation and cognition across adulthood alter the specific
processes that guide decision making in some older individuals.
Although no longitudinal studies have definitively tested these alter-
native accounts, some suggestive cross-sectional evidence supports the lat-
ter account. For instance, in the Iowa gambling task, which involves risky
choice, memory, and learning, the mechanisms that determine performance
may change across the life span in the subset of individuals who continue to
perform well. Interestingly, older adults who perform as well as younger adults
seem to be guided by physiological responses (e.g., assessed with skin conduc-
tance) during anticipation of gains. In stark contrast, younger adults in the
same circumstances appear to be guided instead by physiological responses
during the anticipation of loss (Denburg et al., 2006). Thus, these findings
are consistent with an age-related shift in the processing of positive relative
to negative information (Carstensen & Mikels, 2005; Mather & Carstensen,
2005) and further imply that some older adults leverage this appetitive moti-
vation in the service of making successful risky decisions.

Decision Making Does Not Globally Decline With Age

A number of decision-making scenarios show that older adults perform

just as well or even better than younger adults (Castel, 2005; Hosseini et al.,
2010; Kovalchik, Camerer, Grether, Plott, & Allman, 2005; Kühn et al.,
2011; Y. Li, Baldassi, Johnson, & Weber, 2011; Löckenhoff, 2011; Mata &
Nunes, 2010; Mata et al., 2012; Mather, 2006; Mienaltowski, 2011; Mikels
et al., 2010; Nielsen et al., 2008; Reyna & Brainerd, 2011; Roalf, Mitchell,
Harbaugh, & Janowsky, 2012; Roesch, Bryden, Cerri, Haney, & Schoenbaum,
2012; Samanez-Larkin, Mata, et al., 2011; Scheibe, Mata, & Carstensen,
2011; N. W. Simon et al., 2010; Spaniol, Voss, Bowen, & Grady, 2011;
Strough, Karns, & Schlosnagle, 2011; Worthy, Gorlick, Pacheco, Schnyer,

132       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 132 11/15/13 1:43 PM

 & Maddox, 2011). Many of these scenarios involve decisions that do not
require learning in a novel environment (however, see Worthy et al., 2011)
but instead require accumulation of experience, crystallized intelligence, gist
memory, or emotional and motivational processing.

Enhancing Choice Through Decision Aids

In decision scenarios in which older adults do not perform as well as

younger adults, research might ideally identify opportunities for effective
intervention. The research reviewed above suggests that age differences in
risky decision making may be related to difficulty in computing dynamic rep-
resentations of expected value in a novel environment. Thus, in subsequent
work, we sought to determine whether targeted decision aids could improve
the financial risk taking of older adults (Samanez-Larkin, Wagner, & Knutson,
2011). We found that presentation of explicit expected value information
(either in the form of a graphical representation of all trial outcomes or a sum-
mary measure of prior and projected earnings) improved decision making in
both younger and older adults. Remarkably, providing expected value infor-
mation improved the performance of older adults to match that of younger
adults at baseline (see Color Plate 7c). These findings imply that providing
simplified information about expected value may improve financial decision
making across the adult life span. Similar benefits of “informed” choice have
also been reported in children (Van Duijvenvoorde, Jansen, Bredman, &
Huizenga, 2012).
Although these results are encouraging, this research is many steps away
from broad implementation outside the laboratory. Some evidence suggests
that these and related behavioral interventions may have limited effective-
ness in older individuals. For example, describing an expected value compu-
tation strategy and exposing subjects to repeated training with this strategy
evokes less behavioral change in older than in younger adults (Westbrook,
Martins, Yarkoni, & Braver, 2012), since older adults tend to shift away from
the suggested strategy over time. Research has yet to clarify whether this
shift is related to memory decay or a subjective perception that the suggested
strategy lacks efficacy.
New decision strategies may prove more difficult to train after a life-
time of experience with other strategies. In fact, in old age it may be more
adaptive for individuals to stick with simpler strategies (Worthy & Maddox,
2012). Some evidence suggests that in many real-world decision scenarios,
more simple (or “satisficing”) strategies do not diminish decision quality for
older individuals (Mata & Nunes, 2010). Importantly, the extent to which

reward, risk, and aging      133

13490-06_Ch05-3rdPgs.indd 133 11/15/13 1:43 PM

 age differences in decision quality emerge in the real world depends on the
context or environment in which the decisions are made (Mata et al., 2012).
Although decision aids (Samanez-Larkin, Wagner, & Knutson, 2011;
Van Duijvenvoorde et al., 2012) may provide more reliable support than
strategic training, they may be more limited in application. In many real-
world circumstances, prior outcomes and expected values cannot be easily
recorded, computed, or displayed. However, precise computations of value
may not be required in most scenarios. In fact, some have argued that get-
ting the gist rather than exact verbatim details is most critical in real-world
scenarios (e.g., Reyna & Farley, 2006; Reyna & Lloyd, 2006). Overall, ini-
tial findings seem promising enough to imply that experimentally informed
implementation of decision aids at a broader societal level represents a viable
long-term research goal.

Extending Findings to the Real World

An important focus of our research on reward processing, reward learn-

ing, risky decision making, and aging involves extending performance mea-
sures from the laboratory to the real world. Interestingly, individuals who
make fewer “mistakes” in our laboratory financial investment task (or more
choices in line with those of a risk-neutral, Bayesian-updating actor) also
accumulate more assets in the real world (Samanez-Larkin et al., 2010). In
a related probabilistic learning task, we found that individual differences in
learning to acquire gains were associated with accumulation of financial assets,
whereas individual differences in learning to avoid losses were distinctly asso-
ciated with avoidance of financial debt—which was reflected in individuals’
credit scores (Knutson, Samanez-Larkin, & Kuhnen, 2011). Findings from
this study also indicated that overall probabilistic learning ability was associ-
ated with a more general measure of financial well-being—the debt-to-asset
ratio (Knutson et al., 2011), and all of these effects remained significant after
controlling for age and intelligence quotient. These studies not only provide
evidence for the ecological validity of laboratory-based incentive tasks, but
also may help to identify individuals with specific vulnerabilities that com-
promise financial decision making in the real world (Denburg et al., 2007).
Unfortunately, older adults are disproportionately targeted by financial fraud
(SaveAndInvest.org, 2011). In current work, we are studying older individu-
als who are at heightened risk for making financial mistakes based on prior
fraud victimization. We are also examining how potential vulnerability to
financial fraud may relate to affective or cognitive individual differences in
neural activity and behavior across adulthood.

134       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 134 11/15/13 1:43 PM

 Conclusion

Basic motivational processes and associated brain function appear

to remain relatively intact across much of adulthood and into old age
(Samanez-Larkin, 2010; Samanez-Larkin & Carstensen, 2011). In simple
incentive processing tasks with low cognitive demands, younger and older
adults show similar patterns of subcortical activity. In risky decision making
tasks, contrary to societal stereotypes, risk tolerance appears relatively sta-
ble across adulthood (Mata et al., 2011), with some evidence that individ-
ual differences may increase later in life (Eppinger & Kray, 2011; Spaniol &
Wegier, 2012). Rather than systematically increasing with age, suboptimal
risky decisions in older adults seem more linked to cognitive demands of
particular decision scenarios. Subcortical regions like the ventral striatum
that reliably activate in reward tasks with low cognitive demands begin to
show irregular activity in decision tasks with higher cognitive demands.
Related studies of broader network structure suggest that these specific
decision impairments may relate more closely to circuit dysfunction in
general, and to frontostriatal connectivity in particular (Samanez-Larkin
et al., 2012).
Taken together, these and other findings suggest that aging does not uni-
formly degrade decision making. Many studies find little evidence of declining
decision quality in either some or all older adults as compared with younger
adults (Samanez-Larkin, 2011). Future research should explore both contexts
in which older adults may show suboptimal choice as well as situations in
which decision making remains stable or even improves with age (Y. Li et al.,
2011; Löckenhoff, 2011; Strough et al., 2011). Future research should also
strive to establish the ecological validity of laboratory findings. Currently,
few studies have directly examined adult age differences in risky economic
decision making in the real world (Agarwal, Driscoll, Gabaix, & Laibson,
2009; Korniotis & Kumar, 2011; Mata & Nunes, 2010). Researchers can and
should integrate laboratory measures (including brain imaging) with real-
world measures of decision making to more fully characterize how decision
making changes across the life span.
Although the research reviewed here encompasses a broad range of
processes, the field is still young and will surely continue to grow in com-
ing years. An integrative decision neuroscience approach has tremendous
potential to have both scientific and societal impact. Given the current and
unique moment in human history in which demographic changes are drasti-
cally increasing the age of decision makers around the globe, scientists have
the potential to make major contributions to improving the well-being of
humans of all ages.

reward, risk, and aging      135

13490-06_Ch05-3rdPgs.indd 135 11/15/13 1:43 PM

 References

Agarwal, S., Driscoll, J. C., Gabaix, X., & Laibson, D. I. (2009). The age of reason:
Financial decisions over the life-cycle with implications for regulation. Brook-
ings Papers on Economic Activity, 40, 51–117. doi:10.1353/eca.0.0067
Botwinick, J. (1969). Disinclination to venture response versus cautiousness in
responding: Age differences. The Journal of Genetic Psychology, 115, 55–62. doi:
10.1080/00221325.1969.10533870
Braver, T. S., & Barch, D. M. (2002). A theory of cognitive control, aging cognition,
and neuromodulation. Neuroscience and Biobehavioral Reviews, 26, 809–817.
doi:10.1016/S0149-7634(02)00067-2
Calhoun, R. E., & Hutchison, S. L. (1981). Decision-making in old age: Cautious-
ness and rigidity. The International Journal of Aging & Human Development, 13,
89–98. doi:10.2190/LFUA-4KR0-B0QX-M584
Carstensen, L. L. (2006). The influence of a sense of time on human development.
Science, 312, 1913–1915. doi:10.1126/science.1127488
Carstensen, L. L., & Mikels, J. A. (2005). At the intersection of emotion and cogni-
tion: Aging and the positivity effect. Current Directions in Psychological Science,
14, 117–121. doi:10.1111/j.0963-7214.2005.00348.x
Carstensen, L. L., Mikels, J. A., & Mather, M. (2005). Aging and the intersection
of cognition, motivation and emotion. In J. E. Birren & K. W. Schaie (Eds.),
Handbook of the psychology of aging (6th ed.; pp. 343–362.). San Diego, CA:
Academic Press.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). Emotional
experience in everyday life across the adult life span. Journal of Personality and
Social Psychology, 79, 644–655. doi:10.1037/0022-3514.79.4.644
Carstensen, L. L., Turan, B., Scheibe, S., Ram, N., Ersner-Hershfield, H., Samanez-
Larkin, G. R., . . . Nesselroade, J. K.. (2011). Emotional experience improves
with age: Evidence based on over 10 years of experience sampling. Psychology
and Aging, 26, 21–33. doi:10.1037/a0021285
Castel, A. D. (2005). Memory for grocery prices in younger and older adults: The role
of schematic support. Psychology and Aging, 20, 718–721. doi:10.1037/0882-
7974.20.4.718
Charles, S. T., & Carstensen, L. L. (2010). Social and emotional aging. Annual
Review of Psychology, 61, 383–409. doi:10.1146/annurev.psych.093008.100448
Cohen, M. X., Schoene-Bake, J.-C., Elger, C. E., & Weber, B. (2009). Connectivity-
based segregation of the human striatum predicts personality characteristics.
Nature Neuroscience, 12(1), 32–34. doi:10.1038/nn.2228
Cox, K. M., Aizenstein, H. J., & Fiez, J. A. (2008). Striatal outcome processing
in healthy aging. Cognitive, Affective & Behavioral Neuroscience, 8, 304–317.
doi:10.3758/CABN.8.3.304

136       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 136 11/15/13 1:43 PM

 Denburg, N. L., Cole, C. A., Hernandez, M., Yamada, T. H., Tranel, D., Bechara, A.,
& Wallace, R. B. (2007). The orbitofrontal cortex, real-world decision-making,
and normal aging. Annals of the New York Academy of Sciences. doi:10.1196/
annals.1401.031
Denburg, N. L., Recknor, E. C., Bechara, A., & Tranel, D. (2006). Psychophysiologi-
cal anticipation of positive outcomes promotes advantageous decision-making
in normal older persons. International Journal of Psychophysiology, 61(1), 19–25.
doi:10.1016/j.ijpsycho.2005.10.021
Denburg, N. L., Tranel, D., & Bechara, A. (2005). The ability to decide advanta-
geously declines prematurely in some normal older persons. Neuropsychologia,
43, 1099–1106. doi:10.1016/j.neuropsychologia.2004.09.012
Draganski, B., Kherif, F., Kloppel, S., Cook, P. A., Alexander, D. C., Parker,
G. J. M., . . . Frackowiak, R. S. J. (2008). Evidence for segregated and integrative
connectivity patterns in the human basal ganglia. The Journal of Neuroscience,
28, 7143–7152. doi:10.1523/JNEUROSCI.1486-08.2008
Eppinger, B., Hämmerer, D., & Li, S.-C. (2011). Neuromodulation of reward-based
learning and decision making in human aging. Annals of the New York Academy
of Sciences, 1235(1), 1–17. doi:10.1111/j.1749-6632.2011.06230.x
Eppinger, B., & Kray, J. (2011). To choose or to avoid: Age differences in learning
from positive and negative feedback. Journal of Cognitive Neuroscience, 23(1),
41–52. doi:10.1162/jocn.2009.21364
Eppinger, B., Kray, J., Mock, B., & Mecklinger, A. (2008). Better or worse than
expected? Aging, learning, and the ERN. Neuropsychologia, 46, 521–539.
doi:10.1016/j.neuropsychologia.2007.09.001
Eppinger, B., Nystrom, L. E., & Cohen, J. D. (2012). Reduced sensitivity to immedi-
ate reward during decision-making in older than younger adults. PLoS ONE,
7(5), e36953. doi:10.1371/journal.pone.0036953
Frank, M. J., & Kong, L. (2008). Learning to avoid in older age. Psychology and Aging,
23, 392–398. doi:10.1037/0882-7974.23.2.392
Frank, M. J., Seeberger, L. C., & O’Reilly, R. C. (2004). By carrot or by stick:
Cognitive reinforcement learning in parkinsonism. Science, 306, 1940–1943.
doi:10.1126/science.1102941
Garrett, D. D., Kovacevic, N., McIntosh, A. R., & Grady, C. L. (2010). Blood oxy-
gen level–dependent signal variability is more than just noise. The Journal of
Neuroscience, 30, 4914–4921. doi:10.1523/JNEUROSCI.5166-09.2010
Haber, S. N. (2003). The primate basal ganglia: Parallel and integrative networks.
Journal of Chemical Neuroanatomy, 26, 317–330. doi:10.1016/j.jchemneu.
2003.10.003
Haber, S. N., & Knutson, B. (2010). The reward circuit: linking primate anatomy
and human imaging. Neuropsychopharmacology, 35(1), 4–26. doi:10.1038/npp.
2009.129

reward, risk, and aging      137

13490-06_Ch05-3rdPgs.indd 137 11/15/13 1:43 PM

 Hämmerer, D., Li, S.-C., Müller, V., & Lindenberger, U. (2011). Life span differences
in electrophysiological correlates of monitoring gains and losses during proba-
bilistic reinforcement learning. Journal of Cognitive Neuroscience, 23, 579–592.
doi:10.1162/jocn.2010.21475
Harlé, K. M., & Sanfey, A. G. (2012). Social economic decision-making across
the lifespan: An fMRI investigation. Neuropsychologia, 50, 1416–1424.
doi:10.1016/j.neuropsychologia.2012.02.026
Hedden, T., & Gabrieli, J. D. E. (2004). Insights into the ageing mind: A view from
cognitive neuroscience. Nature Reviews Neuroscience, 5, 87–96. doi:10.1038/
nrn1323
Henninger, D. E., Madden, D. J., & Huettel, S. A. (2010). Processing speed and
memory mediate age-related differences in decision making. Psychology and
Aging, 25, 262–270. doi:10.1037/a0019096
Hosseini, S. M. H., Rostami, M., Yomogida, Y., Takahashi, M., Tsukiura, T., &
Kawashima, R. (2010). Aging and decision making under uncertainty: Behav-
ioral and neural evidence for the preservation of decision making in the absence
of learning in old age. NeuroImage, 52, 1514–1520. doi:10.1016/j.neuroimage.
2010.05.008
Knutson, B., Adams, C. M., Fong, G. W., & Hommer, D. (2001). Anticipation of
increasing monetary reward selectively recruits nucleus accumbens. The Journal
of Neuroscience, 21, RC159).
Knutson, B., & Cooper, J. C. (2005). Functional magnetic resonance imaging of
reward prediction. Current Opinion in Neurology, 18, 411–417. doi:10.1097/01.
wco.0000173463.24758.f6
Knutson, B., Fong, G. W., Bennett, S. M., Adams, C. M., & Hommer, D. (2003).
A region of mesial prefrontal cortex tracks monetarily rewarding outcomes:
Characterization with rapid event-related fMRI. NeuroImage, 18, 263–272.
doi:10.1016/S1053-8119(02)00057-5
Knutson, B., & Greer, S. M. (2008). Anticipatory affect: Neural correlates and con-
sequences for choice. Philosophical Transactions of the Royal Society of London.
Series B, Biological Sciences, 363, 3771–3786. doi:10.1098/rstb.2008.0155
Knutson, B., Samanez-Larkin, G. R., & Kuhnen, C. M. (2011). Gain and loss learn-
ing differentially contribute to life financial outcomes. PLoS ONE, 6, e24390.
doi:10.1371/journal.pone.0024390
Korniotis, G. M., & Kumar, A. (2011). Do older investors make better investment
decisions? The Review of Economics and Statistics, 93(1), 244–265. doi:10.1162/
REST_a_00053
Kovalchik, S., Camerer, C. F., Grether, D., Plott, C., & Allman, J. M. (2005). Aging
and decision making: A comparison between neurologically healthy elderly and
young individuals. Journal of Economic Behavior & Organization, 58(1), 79–94.
doi:10.1016/j.jebo.2003.12.001

138       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 138 11/15/13 1:43 PM

 Kühn, S., Schmiedek, F., Schott, B. H., Ratcliff, R., Heinze, H.-J., Düzel,
E., . . . Lövden, M. (2011). Brain areas consistently linked to individual dif-
ferences in perceptual decision-making in younger as well as older adults
before and after training. Journal of Cognitive Neuroscience, 23, 2147–2158.
doi:10.1162/jocn.2010.21564
Kuhnen, C. M., & Knutson, B. (2005). The neural basis of financial risk taking.
Neuron, 47, 763–770. doi:10.1016/j.neuron.2005.08.008
Lee, T. M. C., Leung, A. W. S., Fox, P. T., Gao, J.-H., & Chan, C. C. H. (2008). Age-
related differences in neural activities during risk taking as revealed by func-
tional MRI. Social Cognitive and Affective Neuroscience, 3(1), 7–15. doi:10.1093/
scan/nsm033
Li, S.-C. (2005). Neurocomputational perspectives linking neuromodulation, process-
ing noise, representational distinctiveness, and cognitive aging. In R. Cabeza,
L. Nyberg & D. Park (Eds.), Cognitive neuroscience of aging: Linking cognitive and
cerebral aging (pp. 354–379). New York, NY: Oxford University Press.
Li, S.-C., Lindenberger, U., & Bäckman, L. (2010). Dopaminergic modulation of cog-
nition across the life span. Neuroscience and Biobehavioral Reviews, 34, 625–630.
doi:10.1016/j.neubiorev.2010.02.003
Li, S.-C., Lindenberger, U., & Sikström, S. (2001). Aging cognition: From neuromod-
ulation to representation. Trends in Cognitive Sciences, 5, 479–486. doi: 10.1016/
S1364-6613(00)01769-1
Li, Y., Baldassi, M., Johnson, E. J., & Weber, E. U. (2011). Compensating cognitive
capabilities, decision performance, and aging. Unpublished manuscript. Columbia
University, New York, NY.
Lighthall, N. R., Gorlick, M. A., Schoeke, A., Frank, M. J., & Mather, M. (2013).
Stress modulates reinforcement learning in younger and older adults. Psychology
and Aging. doi:10.1037/a0029823
Löckenhoff, C. E. (2011). Age, time, and decision making: From processing speed
to global time horizons. Annals of the New York Academy of Sciences, 1235(1),
44–56. doi:10.1111/j.1749-6632.2011.06209.x
Mata, R., Josef, A. K., Samanez-Larkin, G. R., & Hertwig, R. (2011). Age differences
in risky choice: A meta-analysis. Annals of the New York Academy of Sciences,
1235(1), 18–29. doi:10.1111/j.1749-6632.2011.06200.x
Mata, R., & Nunes, L. (2010). When less is enough: Cognitive aging, information
search, and decision quality in consumer choice. Psychology and Aging, 25,
289–298. doi:10.1037/a0017927
Mata, R., Pachur, T., von Helversen, B., Hertwig, R., Rieskamp, J., & Schooler,
L. (2012). Ecological rationality: A framework for understanding and aid-
ing the aging decision maker. Frontiers in Neuroscience, 6, 19. doi:10.3389/
fnins.2012.00019
Mather, M. (2006). A review of decision-making processes: Weighing the risks
and benefits of aging. In L. L. Carstensen & C. R. Hartel (Eds.), When I’m 64
(pp. 145–173). Washington, DC: The National Academies Press.

reward, risk, and aging      139

13490-06_Ch05-3rdPgs.indd 139 11/15/13 1:43 PM

 Mather, M., & Carstensen, L. L. (2005). Aging and motivated cognition: The posi-
tivity effect in attention and memory. Trends in Cognitive Sciences, 9, 496–502.
doi:10.1016/j.tics.2005.08.005
Mell, T., Wartenburger, I., Marschner, A., Villringer, A., Reischies, F. M., & Heekeren,
H. R. (2009). Altered function of ventral striatum during reward-based deci-
sion making in old age. Frontiers in Human Neuroscience, 3, 34. doi:10.3389/
neuro.09.034.2009
Mienaltowski, A. (2011). Everyday problem solving across the adult life span: Solu-
tion diversity and efficacy. Annals of the New York Academy of Sciences, 1235(1),
75–85. doi:10.1111/j.1749-6632.2011.06207.x
Mikels, J. A., Löckenhoff, C. E., Maglio, S. J., Goldstein, M. K., Garber, A., &
Carstensen, L. L. (2010). Following your heart or your head: Focusing on
emotions versus information differentially influences the decisions of younger
and older adults. Journal of Experimental Psychology: Applied, 16(1), 87–95.
doi:10.1037/a0018500
Mora, F., Segovia, G., & Del Arco, A. (2007). Glutamate-dopamine-GABA
interactions in the aging basal ganglia. Brain Research Reviews, 52, 340–353.
doi:10.1016/j.brainresrev.2007.10.006
Nielsen, L., Knutson, B., & Carstensen, L. L. (2008). Affect dynamics, affective fore-
casting, and aging. Emotion, 8, 318–330. doi:10.1037/1528-3542.8.3.318
Okun, M. A. (1976). Adult age and cautiousness in decision: A review of the litera-
ture. Human Development, 19, 220–233. doi:10.1159/000271530
Park, D. C., Lautenschlager, G., Hedden, T., Davidson, N. S., Smith, A. D., &
Smith, P. K. (2002). Models of visuospatial and verbal memory across the adult
life span. Psychology and Aging, 17, 299–320. doi:10.1037/0882-7974.17.2.299
Reyna, V. F., & Brainerd, C. J. (2011). Dual processes in decision making and develop-
mental neuroscience: A fuzzy-trace model. Developmental Review, 31, 180–206.
doi:10.1016/j.dr.2011.07.004
Reyna, V. F., & Farley, F. (2006). Risk and rationality in adolescent decision making:
Implications for theory, practice, and public policy. Psychological Science in the
Public Interest, 7(1), 1–44. doi:10.1111/j.1529-1006.2006.00026.x
Reyna, V. F., & Lloyd, F. J. (2006). Physician decision making and cardiac risk:
Effects of knowledge, risk perception, risk tolerance, and fuzzy processing.
Journal of Experimental Psychology: Applied, 12, 179–195. doi:10.1037/1076-
898X.12.3.179
Roalf, D. R., Mitchell, S. H., Harbaugh, W. T., & Janowsky, J. S. (2012). Risk,
reward, and economic decision making in aging. The Journals of Gerontology:
Series B: Psychological Sciences and Social Sciences, 67, 289–298. doi:10.1093/
geronb/gbr099
Roesch, M. R., Bryden, D. W., Cerri, D. H., Haney, Z. R., & Schoenbaum, G.
(2012). Willingness to wait and altered encoding of time-discounted reward
in the orbito­frontal cortex with normal aging. The Journal of Neuroscience, 32,
5525–5533. doi:10.1523/JNEUROSCI.0586-12.2012

140       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 140 11/15/13 1:43 PM

 Salthouse, T. A. (2004). What and when of cognitive aging. Current Directions in
Psychological Science, 13, 140–144. doi:10.1111/j.0963-7214.2004.00293.x
Samanez-Larkin, G. R. (2010, April 22). Incentive processing in the aging brain: Indi-
vidual differences in value-based learning and decision making across the adult life
span (Doctoral dissertation, Stanford University). Retrieved from http://purl.
stanford.edu/vy834vv5149
Samanez-Larkin, G. R. (2011). Decision making over the life span. Hoboken, NJ:
Wiley-Blackwell.
Samanez-Larkin, G. R., & Carstensen, L. L. (2011). Socioemotional functioning and
the aging brain. In J. Decety & J. T. Cacioppo (Eds.), The handbook of social neu-
roscience (pp. 507–521). New York, NY: Oxford University Press. doi:10.1093/
oxfordhb/9780195342161.013.0034
Samanez-Larkin, G. R., Gibbs, S. E. B., Khanna, K., Nielsen, L., Carstensen, L. L., &
Knutson, B. (2007). Anticipation of monetary gain but not loss in healthy older
adults. Nature Neuroscience, 10, 787–791. doi:10.1038/nn1894
Samanez-Larkin, G. R., Kuhnen, C. M., Yoo, D. J., & Knutson, B. (2010). Variability
in nucleus accumbens activity mediates age-related suboptimal financial risk
taking. The Journal of Neuroscience, 30, 1426–1434. doi:10.1523/JNEUROSCI.
4902-09.2010
Samanez-Larkin, G. R., Levens, S. M., Perry, L. M., Dougherty, R. F., & Knutson,
B. (2012). Frontostriatal white matter integrity mediates adult age differences
in probabilistic reward learning. The Journal of Neuroscience, 32, 5333–5337.
doi:10.1523/JNEUROSCI.5756-11.2012
Samanez-Larkin, G. R., Mata, R., Radu, P. T., Ballard, I. C., Carstensen, L. L., &
McClure, S. M. (2011). Age differences in striatal delay sensitivity during inter-
temporal choice in healthy adults. Frontiers in Neuroscience, 5, 126. doi:10.3389/
fnins.2011.00126
Samanez-Larkin, G. R., Wagner, A. D., & Knutson, B. (2011). Expected value infor-
mation improves financial risk taking across the adult life span. Social Cognitive
and Affective Neuroscience, 6, 207–217. doi:10.1093/scan/nsq043
SaveAndInvest.org. (2011). Fighting Fraud 101. Retrieved from http://www.saveand
invest.org/web/groups/sai/@sai/documents/sai_original_content/p036701.pdf
Scheibe, S., Mata, R., & Carstensen, L. L. (2011). Age differences in affective fore-
casting and experienced emotion surrounding the 2008 US presidential election.
Cognition and Emotion, 25, 1029–1044. doi:10.1080/02699931.2010.545543
Schott, B. H., Niehaus, L., Wittmann, B. C., Schütze, H., Seidenbecher, C. I.,
Heinze, H.-J., & Düzel, E. (2007). Aging and early-stage Parkinson’s disease
affect separable neural mechanisms of mesolimbic reward processing. Brain: A
Journal of Neurology, 130, 2412–2424. doi:10.1093/brain/awm147
Segovia, G., Porras, A., Del Arco, A., & Mora, F. (2001). Glutamatergic neuro-
transmission in aging: A critical perspective. Mechanisms of Ageing and Devel-
opment, 122(1), 1–29. doi:10.1016/S0047-6374(00)00225-6

reward, risk, and aging      141

13490-06_Ch05-3rdPgs.indd 141 11/15/13 1:43 PM

 Simon, J. R., Howard, J. H., & Howard, D. V. (2010). Adult age differences in learning
from positive and negative probabilistic feedback. Neuropsychology, 24, 534–541.
doi:10.1037/a0018652
Simon, N. W., Lasarge, C. L., Montgomery, K. S., Williams, M. T., Mendez, I. A.,
Setlow, B., & Bizon, J. L. (2010). Good things come to those who wait: Attenu-
ated discounting of delayed rewards in aged Fischer 344 rats. Neurobiology of
Aging, 31, 853–862. doi:10.1016/j.neurobiolaging.2008.06.004
Spaniol, J., Voss, A., Bowen, H. J., & Grady, C. L. (2011). Motivational incentives
modulate age differences in visual perception. Psychology and Aging, 26, 932–939.
doi:10.1037/a0023297
Spaniol, J., & Wegier, P. (2012). Decisions from experience: Adaptive information
search and choice in younger and older adults. Frontiers in Neuroscience, 6, 36.
doi:10.3389/fnins.2012.00036
Strough, J., Karns, T. E., & Schlosnagle, L. (2011). Decision-making heuristics and
biases across the life span. Annals of the New York Academy of Sciences, 1235(1),
57–74. doi:10.1111/j.1749-6632.2011.06208.x
Van Duijvenvoorde, A. C. K., Jansen, B. R. J., Bredman, J. C., & Huizenga, H. M.
(2012). Age-related changes in decision making: Comparing informed and non-
informed situations. Developmental Psychology, 48(1), 192–203. doi:10.1037/
a0025601
Westbrook, A., Martins, B. S., Yarkoni, T., & Braver, T. S. (2012). Strategic insight
and age-related goal-neglect influence risky decision-making. Frontiers in Neuro­
science, 6, 68. doi:10.3389/fnins.2012.00068
Wood, S., Busemeyer, J., Koling, A., Cox, C. R., & Davis, H. (2005). Older adults
as adaptive decision makers: Evidence from the Iowa gambling task. Psychology
and Aging, 20, 220–225. doi:10.1037/0882-7974.20.2.220
Worthy, D. A., Gorlick, M. A., Pacheco, J. L., Schnyer, D. M., & Maddox, W. T.
(2011). With age comes wisdom: decision making in younger and older adults.
Psychological Science, 22, 1375–1380. doi:10.1177/0956797611420301
Worthy, D. A., & Maddox, W. T. (2012). Age-based differences in strategy use in
choice tasks. Frontiers in Neuroscience, 5, 145. doi:10.3389/fnins.2011.00145
Wu, C. C., Samanez-Larkin, G. R., & Knutson, B. (2012). Neural markers of incentive
anticipation are associated with affective traits. Working Paper.

142       samanez-larkin and knutson

13490-06_Ch05-3rdPgs.indd 142 11/15/13 1:43 PM

 III
Neuropsychology

13490-07_PT3_Ch06-3rdPgs.indd 143 11/15/13 1:43 PM

13490-07_PT3_Ch06-3rdPgs.indd 144 11/15/13 1:43 PM
 6
Mind and Brain in Delay
of Gratification
Vivian Zayas, Walter Mischel, and Gayathri Pandey

A central focus of psychological and behavioral sciences is to identify the fac­

tors that enable and hinder delay of gratification. In this chapter, we review
findings from the original preschool delay of gratification work that identify
key attentional–cognitive control strategies that enable (vs. hinder) delay. We
also describe recent behavioral and neuroscientific findings that investigate the
link between preschool delay of gratification abilities and adult mechanisms of
cognitive control. This work suggests that dispositional abilities to delay grati­
fication are subserved by individual differences in the functioning of prefrontal
cortical and limbic neural systems. We end by discussing the implications of this
work for theory and future research.
Delay of gratification, the ability to forgo an immediate reward for the
sake of obtaining a more desirable but delayed reward, has been shown to
predict successful outcomes in diverse and consequential domains, such
as academics and work, close relationships, and physical and mental well-
being (Mischel, Shoda, & Rodriguez, 1989; Mischel et al., 2011). Moreover,

http://dx.doi.org/10.1037/14322-007
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

145

13490-07_PT3_Ch06-3rdPgs.indd 145 11/15/13 1:43 PM

 the extent to which behaviors are guided by momentary rewards without
consideration of future costs increases the likelihood for drug abuse, risky
sexual behaviors, obesity, criminality, and consumer debt (e.g., Bogg &
Roberts, 2004; Moffitt et al., 2011; Polivy, 1998; Tangney, Baumeister, &
Boone, 2004).
Considerable attention has been devoted to identifying the biologi-
cal, psychological, social, and cultural factors, and the interactions among
them, that enable some individuals to successfully delay gratification and
make doing so difficult for others (Ainslie, 1975; Blair, 2010; Blair & Raver,
2012; Evans & Kim, 2013; Hackman, Farah, & Meaney, 2010; Mischel et al.,
1989). Here, we focus on the psychological and neural mechanisms that
facilitate or hinder the ability to resist temptation. We first review the classic
work finding that individual differences in the ability to delay gratification
observed in early childhood show remarkable stability over the life span
and across situations, predicting a wide array of adaptive life outcomes in
adolescence and adulthood. Next, we review findings from experimental
work on delay of gratification that identified key attentional–cognitive pro-
cesses that enable delay of gratification, specifically distraction (directing
attention away from tempting, appetitive, consummatory, visceral, or “hot”
aspects of the situation) and reconstrual or reappraisal (focusing on qualities
of rewards that are more informational, abstract, and “cooler”). We then
briefly describe recent findings from our research team on the cognitive and
neural circuitry that may subserve dispositional abilities to delay gratifica-
tion throughout the life span.
We conclude by highlighting a key implication of the extant findings
on delay of gratification to date. An important consequence of enacting self-
control strategies is that they dampen the extent to which representations
of temptations, which are inherently appetitive and consummatory (or hot),
are active in working memory. In so doing, they decrease the psychological
pull of the temptation in the situation—even in the face of actual, objective
temptations. Although effortful cognitive control is a prerequisite for the
enactment of self-control strategies, such as distraction and reconstrual, such
an act makes delaying gratification less effortful and indeed much easier. It
enables more “willpower” with less sweat if, as William James (1890/1950)
put it long ago, certain preliminaries are met.

Delay of Gratification

The preschool delay of gratification paradigm (Mischel, Ebbesen, &

Zeiss, 1972; Mischel et al., 1989) has garnered the attention of scholars
from diverse disciplines in social, developmental, cognitive, and clinical

146       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 146 11/15/13 1:43 PM

 psychology, as well as behavioral economics and philosophy, because it cap-
tures a quintessential self-control dilemma: Does one choose a smaller reward
available immediately, such as taking a toke of a cigarette, eating a piece of
the forbidden chocolate cake, or splurging on a trip to Paris, or wait for a
larger reward available at a later time, such as achieving physical health, a
lean physique, or financial security for the golden years? And equally impor-
tant, the delay of gratification paradigm also provides a window into how one
follows through with the decision once made.
In the classic delay of gratification situation, preschoolers are told that
the experimenter needs to leave the room to prepare for an upcoming task
and are presented with the choice of receiving one smaller treat (cookie,
marshmallow, little pretzel stick) immediately or a larger treat (twice the
amount of the treat selected) when the experimenter returns (e.g., Mischel
& Ayduk, 2004; Mischel et al., 1972, 1989). Despite its simplicity, the delay
of gratification situation poses the child with a conflict: Does she choose
what she desires now or does she wait (an unknown amount of time) for what
she desires in the future? Once she chooses, is she able to make good on her
commitment?
Although the majority of preschoolers clearly express to the experi-
menter their intention to wait for the larger, delayed treats, only a subset of
children are actually able to wait the entire 15-minute period. Despite their
best intentions, waiting becomes very difficult in the face of temptation,
and preschoolers experience increasing frustration as they continue to wait.
Thus, many fall short of their stated goal. On average, preschoolers suc-
cumb to the immediate temptation within 1 minute of the start of the delay
period (although 15 minutes were needed to obtain the reward; Mischel &
Ebbesen, 1970).
Remarkably, individual differences in the ability of preschoolers to resist
eating the tempting treats or ringing the bell to end the delay period pre-
dict a multitude of long-term outcomes. Those preschoolers who were able
to delay gratification showed greater cognitive, self-regulatory, and coping
competence later in life (Shoda, Mischel, & Peake, 1990). As adolescents,
these preschoolers were also perceived to be more cognitively and socially
competent, coping better with frustration and stress and achieving higher
scholastic performance (Mischel et al., 1989). As adults, they were less likely
to use cocaine (e.g., Ayduk et al., 2000; Mischel & Ayduk, 2004) and more
likely to have a lower body mass index (Schlam, Wilson, Shoda, Mischel,
& Ayduk, 2013). Such findings are consistent with results obtained in other
cross-sectional and longitudinal studies that have used different samples and
different assessments of delay ability (Bonato & Boland, 1983; Bruce et al.,
2011; Francis & Susman, 2009; Kubzansky, Martin, & Buka, 2009; Moffitt
et al., 2011; Seeyave et al., 2009).

mind and brain in delay of gratification      147

13490-07_PT3_Ch06-3rdPgs.indd 147 11/15/13 1:43 PM

 The fact that preschool delay of gratification ability predicts successful
outcomes in diverse domains (from academics to health) suggests that success
in delaying gratification at age 4 taps into more general cognitive capacities
(e.g., attentional control, inhibitory control, working memory) involved in
the regulation of feelings, thoughts, and behaviors for the sake of achieving
long-term goals. In particular, in order to wait for the larger reward, the pre-
schooler must be able to regulate feelings, thoughts, and behaviors associated
with the frustration elicited by waiting in the presence of the readily avail-
able rewards. Such capacities for regulating psychological and behavioral
responses for the sake of long-term goals are needed to succeed in diverse life
domains, as one must be able to effectively cope with the inevitable setbacks
that arise during goal pursuit.
Given this, Ayduk et al. (2000) reasoned that the ability to regulate
aversive states, as reflected by the delay of gratification ability, may be par-
ticularly beneficial for individuals who possess chronic personal vulner-
abilities that increase their likelihood of experiencing aversive, negative
states. To test the hypothesis that delay ability serves a protective function
among vulnerable groups, in two studies, using two different cohorts, indi-
viduals were assessed in the delay of gratification paradigm in childhood,
and they were assessed again on dispositional rejection sensitivity (RS) as
well as various interpersonally relevant outcome measures (aggression, peer
rejection, drug use) in adulthood approximately 20 years later. Past work
has shown that high RS individuals, who are predisposed to anxiously or
angrily expect, readily perceive, and intensely react to social rejection, are
more likely to report destructive personal and interpersonal outcomes, such
as depression, verbal and nonverbal signs of anger, and physical aggression
(Ayduk, Downey, & Kim, 2001; Ayduk, Downey, Testa, Yen, & Shoda, 1999;
Downey & Feldman, 1996; Downey, Freitas, Michaelis, & Khouri, 1998;
Downey, Lebolt, Rincon, & Freitas, 1998). In Ayduk et al.’s (2000) work,
high RS individuals who were also low in preschool delay ability showed
the predictable negative consequences of RS. They reported lower self-
esteem and self-worth, less effective coping ability, lower education levels,
and greater cocaine and crack use. However, critically, high RS individuals
who were high in preschool delay ability did not show the same negative
consequences. Their outcomes were essentially indistinguishable from low
RS individuals (Ayduk et al., 2000). Similar findings have been obtained in
other samples (Ayduk et al., 2008) and with behavioral measures of reac-
tivity, such as the startle response (Gyurak & Ayduk, 2007). Moreover, as
will be discussed in later sections on the neural correlates subserving delay
ability, the rostral anterior cingulate cortex (rACC), an area of the brain
associated with emotional control (Etkin, Egner, Peraza, Kandel, & Hirsh,

148       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 148 11/15/13 1:43 PM

 2006), is implicated in the buffering effects of attentional control (Gyurak,
Goodkind, Kramer, Miller, & Levenson, 2012).

Cognitive Strategies That Enable

Delay of Gratification

What are the strategies that enable some individuals to resist tempta-
tion while other individuals succumb? Before addressing the psychological
and neural processes that enable or hinder delaying of gratification, it is fruit-
ful to first ask a more basic question: Why is delaying gratification even dif-
ficult? Although this question may appear overly simplistic, it offers insights
into the strategies that may enable delay ability.
Part of the reason that waiting for a larger reward is difficult is because
immediately available rewards are concrete and delayed rewards are abstract.
As such, immediate rewards are highly salient, easily processed, more likely
to automatically capture attention and elicit appetitive behaviors, and more
difficult to ignore and resist (Carretié, Hinojosa, Martín-Loeches, Mercado,
& Tapia, 2004). Immediate rewards are also perceived as subjectively more
valuable (e.g., Ainslie, 1975). In contrast, delayed rewards are hypothetical,
inherently less salient, less motivating, and perceived as less valuable. Thus,
when evaluating between a small reward now versus a larger reward at a later
time, people prefer the former, reflecting the famous proverb that a “bird in
the hand is worth two in the bush.”
Not surprisingly then, by default—in the absence of any attempts at
self-regulation—in quintessential self-control dilemmas, exposures to temp-
tations automatically activate hot, appetitive aspects of the mental represen-
tation of the immediate rewards and spontaneously elicit actions directed
toward them (Metcalfe & Mischel, 1999). These impulsive tendencies must
somehow be regulated in order to obtain the desired but delayed reward.
Starting in the 1970s, a series of experiments focused on identifying the
cognitive strategies that enabled preschoolers to delay immediate gratifica-
tion and wait for the larger, more desired reward (e.g., Mischel et al., 2011).
This work identified various strategies that affected a given individual’s abil-
ity to resist temptation. The simplest strategy was to minimize the extent to
which one attends to the tempting stimulus, either by ignoring it completely
or through distraction in which one directs attention away from aspects of
the situation that elicit appetitive, concrete, or hot representations. A more
complex strategy involved dampening the power of the tempting stimulus via
reframing (reconstruing or reappraising) the meaning of the stimulus to focus

mind and brain in delay of gratification      149

13490-07_PT3_Ch06-3rdPgs.indd 149 11/15/13 1:43 PM

 on the informational, abstract, or “cool” aspects of the situation instead of the
default appetitive, concrete, or hot aspects.

Effect of Distraction Strategies on Delay of Gratification

In several experiments, strategies such as ignoring and distraction that

dampened the salience of the tempting stimulus led to greater delay ability.
For example, if children took part in the delay of gratification procedure when
the rewards were out of sight, they waited longer compared with the standard
condition in which the rewards were plainly in sight (Mischel & Ebbesen,
1970). Thus, an externally enacted strategy of restructuring the physical
situation such that the tempting aspects of the situation were less visible
and unavailable for processing made the task of delaying easier. Likewise, if
children were provided with a distraction (e.g., given a fun toy to play with),
they were able to wait longer than did children who were not provided with
a distraction (Mischel et al., 1972). The effectiveness of this strategy was not
limited to external distractions. Subjective, internal thoughts were equally
effective. When children were simply given the instruction to “think fun,”
the children’s own thoughts served as sufficient distraction, facilitating delay
and successfully thwarting the aversive aspects of the immediate situation
(Mischel et al., 1972).
Most interesting were findings showing that in the absence of any
explicit instructions, preschoolers’ spontaneous enactment of attentional–
control strategies during the standard delay of gratification task in which all
the rewards were fully visible meaningfully predicted delay time (Rodriguez,
Mischel, & Shoda, 1989). Children’s spontaneous behavior during the delay
period was meticulously coded in terms of whether they directed their atten-
tion toward the immediately available tempting reward or toward nontempt-
ing aspects of the situation. The proportion of time spontaneously spent
directing attention away from the rewards and toward other nontempting
aspects of the situation was a strong predictor (r = .49) of delay time.

Effect of Reappraisal on Delay of Gratification

A second strategy enabling delay of gratification is reappraisal. Reconstruing

or reframing the tempting stimuli in terms of its cooler, informational, and
abstract features instead of its naturally salient hot, appetitive, or consumma-
tory features enabled preschoolers to wait longer for the delayed rewards (e.g.,
Mischel et al., 1972, 1989; Mischel & Ayduk, 2004; Mischel & Underwood,
1974). For example, preschoolers who were instructed to imagine the treats
in terms of their cool, abstract, and informational features (e.g., marshmal-
lows look like white puffy clouds) waited longer than those who imagined the

150       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 150 11/15/13 1:43 PM

 treats in terms of their hot, concrete, and appetitive features (e.g., the sweet
taste of marshmallows; Mischel & Baker, 1975). Interestingly, although the
actual presence of the treats increases the salience of the hot aspects of the sit-
uation and thus hinders delay of gratification ability, the symbolic presence of
the treat (a photograph of the treat) increases the salience of the cool aspects
of the rewards and thus promotes delaying gratification (Mischel & Moore,
1973). The specificity of the reward is important. Viewing the photo­graph
of the reward one is waiting for (e.g., two marshmallows) is more motivating
than viewing the photograph of an irrelevant reward (e.g., toy).

Integrating Key Findings From the Experimental

and Longitudinal Work

In short, both the experimental and the longitudinal work speak to the
strategies that allow individuals to delay gratification. On the one hand, the
longitudinal findings suggest that there are important, naturally existing, indi-
vidual differences in delay ability; that this variability predicts consequen-
tial outcomes in adolescence and adulthood in diverse life domains; and that
individual differences in the spontaneous use of attentional–control strategies
(occurring in the absence of any explicit instructions) is a key strategy pre-
dicting delay of gratification (e.g., Eigsti et al., 2006; Rodriguez et al., 1989).
On the other hand, the experimental work suggests that regardless of the par-
ticular strategy (ignore, distraction, reconstrual) and whether it was enacted
externally by the structure of the situation (obscure the treats from sight) or
internally by the preschooler’s own efforts (think about the marshmallows as
clouds), providing preschoolers with self-control strategies served to dampen
the salience of the tempting stimulus, and this in turn enabled delay.
Collectively, the experimental and longitudinal work also move away
from a purely trait, dispositional conceptualization of delay of gratification
ability toward a focus on cognitive strategies and processes. Because such
strategies can be acquired and practiced, they are promising routes for increas-
ing delay of gratification ability. Specifically, the preschooler who might have
succumbed to the tempting aspects of the situation if left to her own devices
might be able to wait for the larger reward if she was provided with an appro-
priate strategy, such as engaging in a distracting activity or focusing on the
cool, cognitive, and abstract qualities of the treats. Similarly, the opposite is
also likely: The preschooler who might have been able to delay gratification
in the absence of any explicit instructions may find it more difficult to wait
for the larger reward if she is provided with an inappropriate strategy, such
as focusing on the immediate reward or its hot, appetitive (consummatory),
and concrete features (Metcalfe & Mischel, 1999; Mischel & Ayduk, 2004).
The critical point is that because the experimental work shows that for the

mind and brain in delay of gratification      151

13490-07_PT3_Ch06-3rdPgs.indd 151 11/15/13 1:43 PM

 majority of children (those with high as well as low delay ability) the appro-
priate strategy can be enacted if made temporarily accessible via external or
internal strategies, it is possible for children who are not spontaneously adept
at delaying gratification to learn to enact more appropriate strategies and
improve their overall ability to delay gratification.

A Hot/Cool Systems Approach to Delay of Gratification

The hot/cool framework (Metcalfe & Mischel, 1999) was developed to

integrate diverse findings from the delay of gratification work and to delin-
eate more precisely the mechanisms by which strategies, such as distraction
and reappraisal, subserve delay of gratification (Mischel & Ayduk, 2002, 2004;
Metcalfe & Mischel, 1999). Similar to other two-system models of related con-
structs, such as risky decision making (Reyna & Farley, 2006; Reyna & Rivers,
2008), temporal discounting (McClure, Laibson, Loewenstein, & Cohen,
2004), impulsivity (Steinberg et al., 2008), and addictions (Bechara, 2005;
Everitt, Dickinson, & Robbins, 2001; Everitt & Robbins, 2005), the hot/
cool framework is based on the idea that there are two modes of informa-
tion processing (e.g., Lieberman, Gaunt, Gilbert, & Trope, 2002; Smith &
DeCoster, 2000): a cool cognitive system involved in more deliberative and
reflective processing and a hot emotional system that is primarily involved in
the spontaneous evaluation of rewards and punishments.
Although the hot/cool framework was first developed as a heuristic,
with no commitment to their anatomical structures, a hope was that these
metaphors might reflect more tangible cognitive and neural processes. As
reflected by the contributions in the present volume, an explosion of research
in cognitive neuroscience over the past two decades has been turning this
idea into a reality (e.g., Heatherton, 2011; Lieberman, 2007; Ochsner &
Gross, 2007). Although an extensive review of the literature is beyond the
scope of this chapter, here we present a brief description of these two systems,
their characteristics and neural substrates, and how variations in the func-
tioning of these neural systems may affect people’s tendency (due to both
dispositional and situational influences) to delay gratification. We then turn
to recent work from our research team identifying the adult cognitive and
neural correlates of preschool delay of gratification.

Hot Limbic System and Cool Cognitive System

The hot, impulsive, or “go” system (also referred to in different frameworks
as “System 1,” automatic, spontaneous, experiential, associative) is seen early in
development (Eisenberger, Smith, Sadovsky, & Spinrad, 2004; Rothbart, Ellis,
& Posner, 2004). It is primarily stimulus-triggered and affective in nature, and it
operates quickly, with little effort, deliberation, or control. As such, it provides

152       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 152 11/15/13 1:43 PM

 a rudimentary analysis of the potential rewards and threats in the environment
(e.g., Cacioppo & Gardner, 1999; Compton, 2003), especially those that are
primary reinforcers and punishers, and enables fast-acting responsive behaviors
(e.g., moving toward rewards and away from threats). Research from cognitive
neuroscience has routinely implicated limbic structures (e.g., Cunningham
et al., 2004), such as the amygdala (e.g., Armony & LeDoux, 2000; Bechara,
2005; Breiter et al., 1996; Metcalfe & Jacobs, 1996, 1998; Morris, Öhman, &
Dolan, 1998; Phelps & LeDoux, 2005), nucleus accumbens (Galvan et al.,
2005), and ventral striatum (Delgado, Locke, Stenger, & Fiez, 2003; Delgado,
Nystrom, Fissell, Noll, & Fiez, 2000; Knutson, Adams, Fong, & Hommer,
2001), with the hot system.
The cool, cognitive, or “know” system (also referred to in different frame-
works as “System 2,” deliberate, rational, rule-based, reflective) is responsible
for a number of abilities that are commonly referred to as cognitive control,
executive functions, central executive, and attentional bias (e.g., Baddeley, 1986;
Cohen & Servan-Schreiber, 1992; Desimone & Duncan, 1995; Shallice,
1988). In contrast to the hot system, the cool system develops later, emerg-
ing by around the age of 4 (Mischel, 1974; see also Eisenberger et al., 2004;
Rothbart et al., 2004). This system is typically viewed as the seat of free will
and self-regulation (Moffitt et al., 2011). It enables an individual to inhibit
automatic, reflexive stimulus-triggered responses; to reflect on and manipulate
(maintain, enhance, and suppress) information in working memory, includ-
ing mental representations of goals and events not present in the immediate
situation; to devise and implement plans for achieving future goals (Baddeley,
1998); and to facilitate effective and appropriate response selection (Rowe,
Toni, Josephs, Frackowiak, & Passingham, 2000). An expansive neuroimag-
ing literature provides evidence that various cognitive control functions are
governed by structures in the prefrontal cortex (PFC; Jackson et al., 2003;
Miller & Cohen, 2001; Ochsner & Gross, 2007; O’Reilly, 2006), such as
the inferior frontal gyrus (IFG), which is involved in maintaining informa-
tion in working memory and in executing planned action, and the lateral
PFC, which is involved in representations of abstract rules (Bunge & Zelazo,
2006), as well as the anterior cingulate cortex, which is hypothesized to serve
conflict-monitoring functions that signal recruitment of PFC control mecha-
nisms (Botvinick, Braver, Barch, Carter, & Cohen, 2001; Etkin et al., 2006;
Heatherton, 2011; Kerns et al., 2004).

Relative Activation of the Hot and Cool System as Mechanisms

Underlying Delay of Gratification
A key assumption of the hot/cool framework is that delay of gratifica-
tion is governed by continuous interactions between the hot and cool systems
(for a discussion, see Metcalfe & Mischel, 1999). From this perspective, a

mind and brain in delay of gratification      153

13490-07_PT3_Ch06-3rdPgs.indd 153 11/15/13 1:43 PM

 given stimulus in the environment can be represented in terms of its con-
crete, appetitive, or hot features or in terms of abstract, informational, or cool
features. Accordingly a given stimulus can activate these features within a
person’s mind, which are represented as a neural network, and are referred to
simply as “hot spots” of the hot system and “cool nodes” of the cool system.
Furthermore, the hot spots of the hot system and the cool nodes of the cool
system are connected with one another, such that activation of cool nodes
(e.g., thinking about a marshmallow as a cloud) can lead to a decrease in the
activation of hot nodes (e.g., focusing on the sensory eliciting qualities of eat-
ing a marshmallow). Thus, the extent to which one engages in greater acti-
vation of the cool (vs. hot) system, for example by means of distraction and
reappraisal strategies, facilitates the ability to wait for the delayed rewards.
In contrast, the extent to which activation of the hot system is dominant,
for example, by increased salience of the temptations, hinders delay. This
idea of two continuously interacting systems whose relative activations have
implications for decision making and behavior resonates with those of other
scholars in the related domains of delay discounting and impulsivity (e.g.,
Blais & Weber, 2001; Figner & Weber, 2011; McClure et al., 2004; Mellers,
Schwartz, & Weber, 1997; Weber, 2001; Zaleskiewicz, 2001).
Not surprisingly, research in cognitive neuroscience is increasingly
focusing on interactions among neural structures subserving the processes
that enable delay of gratification and the relative balance between activation
of limbic structures associated with the processing of affective stimuli and
the PFC structures recruited in regulation of cognition, affect, and behav-
ior (Bechara, 2005; Heatherton, 2011; LeDoux, 1996; McClure et al., 2004,
2007; Ochsner et al., 2004; Sanfey, Hastie, Colvin, & Grafman, 2003; Sanfey,
Loewenstein, McClure, & Cohen, 2006; Somerville & Casey, 2010). Recent
studies show that directing attention away from emotionally salient features
of a situation reduces activation in emotion-generative neural regions, such
as the amygdala and insula (Bantick et al., 2002; McRae et al., 2010). Thus,
similar to the role of distraction in dampening the tempting aspects of the
rewards in the original delaying of gratification experiments, in recent cogni-
tive neuroscience studies, directing attention away from appetitive stimuli
(via shifting gaze or working memory load manipulations) minimizes the
extent to which the stimulus is encoded in working memory. Moreover, a
recent study recording event-related potentials investigated whether atten-
tional deployment is effective even after the affective cue is encoded into
working memory. Thiruchselvam, Hajcak, and Gross (2012) had participants
view an emotion-arousing image (e.g., car crash) and hold the image in work-
ing memory. After the image was in memory, they were instructed to either
focus their attention to a neutral aspect of the representation or an arousing
aspect of the representation. Even though the emotion-arousing image had

154       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 154 11/15/13 1:43 PM

 already been encoded in working memory, attention deployment to neutral
aspects of the representation led to reduced self-reported mood and damp-
ened late positive potential, a neural measure of emotional responding.
Neuroimaging work has also looked at the effect of reconstruing the
meaning of a situation on the activation of neural systems involved in the
processing of affective stimuli and its regulation. For example, Ochsner et al.
(2004) showed that reappraising a negative stimulus (e.g., snake) in more
neutral terms increased activation of the cingulate and PFC, which is impli-
cated in the cool system, and decreased the activation of the amygdala. This
suggests that the PFC and cingulate play a critical role in the reappraisal
strategies attributed to the cool system, and that enactment of such strategies
decreases the sensitivity of the hot system, as reflected by decreased activa-
tion of the amygdala in response to the negative stimulus.

Individual Differences
Contextual factors, both those generated internally by one’s own
thoughts and those generated externally by cues available in the situation,
can influence the relative momentary activation of the hot versus cool sys-
tems and thus a person’s ability to delay. But individuals also differ in the
chronic and stable activations of these two systems. Exposure to environ-
mental factors throughout development can lead to greater dominance of one
system over the other. Chronic stress, for example, through its effects on the
hippocampus (Sapolsky, 1996) or PFC (Brown, Henning, & Wellman, 2005;
Vyas, Bernal, & Chattarji, 2003), could increase the chronic activation of the
hot system (McEwen & Gianaros, 2010). In contrast, routine enactment of
attentional–control and other self-regulation strategies and cognitive capaci-
ties, including higher working memory capacity (Barkley, 1997; Hinson,
Jameson, & Whitney, 2003), could lead to greater chronic activation of the
cool system (Ayduk et al., 2000; Fishbach & Trope, 2005). Biological pre-
dispositions, physiology, drug use, and disease could also affect the extent to
which one system is more chronically active relative to the other. Regardless
of the source, individual differences in the neural structures associated with
either the hot or the cool systems, and their interconnections, are expected
to contribute to chronic, trait-like, dispositional differences in the ability
to delay gratification (Metcalfe & Mischel, 1999; Mischel & Ayduk, 2004).
Indeed, individual differences in the structure and function of these neural
systems may mediate the remarkable predictive ability of preschool delay on
later life outcomes.
Research has routinely found evidence that some individuals are par-
ticularly sensitive to the hot aspects of temptations. For example, through use
of a “cue reactivity” paradigm in which participants are exposed to images

mind and brain in delay of gratification      155

13490-07_PT3_Ch06-3rdPgs.indd 155 11/15/13 1:43 PM

 of primary reinforcers, obese individuals, smokers, and drug addicts show
stronger activation of the mesolimbic reward system in response to tempta-
tions (e.g., image of cigarettes for smokers; Childress et al., 1999; David et al.,
2007; Rothemund et al., 2007). Moreover, increased sensitivity to rewards
has implications for control mechanisms. For example, the typical PFC–
amygdala coupling, which is observed during attempts at emotion regulation
in nonclinical samples, is not observed in clinical populations (e.g., major
depressive disorder, bipolar personality disorder; Johnstone, van Reekum,
Urry, Kalin, & Davidson, 2007). Instead of the inverse coupling suggesting
PFC down-regulation of limbic areas in nonclinical samples, both PFC and
amygdala activation are high in clinical samples. Similarly, in the addiction
literature, addictive behaviors may emerge from a dysregulation in neural
systems in which there is an interplay between a hypersensitive limbic sys-
tem that is highly reactive to rewards along with a hypoactive prefrontal
system (Bechara, 2005) that fails to down-regulate affect (Demos, Kelley, &
Heatherton, 2011). Such dynamics may account for difficulties in delay of
gratification.
Another recent line of inquiry by Gyurak et al. (2012) has been exam-
ining individual differences in the cognitive and neural systems by which
delay ability (and other related capacities such as executive control) might
serve as a protective factor especially for those with chronic vulnerabilities
to adaptively regulate negative emotions and thoughts. Although this work
is not able to investigate the developmental issue, because all measures
were assessed concurrently, it is implicating the role of the rACC, which
is involved in the monitoring of errors and the signaling of the need for
control. Greater activation of the dorsal ACC in a task requiring inhi-
bition of threatening stimuli has been linked to individual differences in
self-reported attentional control (Bishop, Jenkins, & Lawrence, 2007). In
the Gyurak et al. (2007) study, participants viewed images depicting social
rejection (a woman sitting alone in a room) as well as equally negatively
valenced images of a nonsocial nature. Among participants with low self-
esteem (a disposition that is similar to high RS in being highly sensitive
to rejection), those who also had high attentional control reported less
arousal after viewing images of social rejection than did those with low
attentional control. Moreover, results of the functional magnetic resonance
imaging (fMRI) analysis indicated that those with low self-esteem and high
attentional control engaged the rACC, an area of the brain associated with
emotional control, more than their low-self-esteem and low-attentional-
control counter­parts. Activation in the rACC fully mediated the relation-
ship between the interaction of self-esteem and attentional control and
self-reported arousal, suggesting that the rACC activation underlies the
buffering effects of attentional control.

156       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 156 11/15/13 1:43 PM

 Preschool Delay of Gratification and Adult Cognitive and Neural Mechanisms
Despite the growing body of research speaking to the cognitive and
neural mechanisms of the hot and cool systems underlying self-control and
related constructs (Hare, Camerer, & Rangel, 2009; Heatherton, 2011;
McClure et al., 2004), only recently have studies begun to directly inves-
tigate the developmental question of how preschool delay ability relates to
adult cognitive and neural processes. Given that the ability to wait for larger
rewards depends on how one selectively attends to and mentally represents
aspects of the tempting situation, and that enacting these strategies depends
on the ability to recruit cognitive control, a straightforward prediction is that
preschool delay would be a precursor, or marker, of the cognitive and neural
systems underlying cognitive control in adulthood.
In a multisite, multidisciplinary endeavor, our research team has begun
to tackle this developmental question. This work has involved following up
with individuals who took part in the delay of gratification at age 4 and now
in adulthood having the same individuals complete behavioral measures
while fMRI scans are obtained. We briefly review key findings from this new
wave of delay of gratification research here.

Behavioral Findings: Predicting Adult Impulse Control

From Preschool Attentional–Control Strategies
In a follow-up to an existing longitudinal study, we investigated whether
performance in the delay of gratification situation at age 4 predicts perfor-
mance on a standard measure of impulse control (go/no-go task) at age 18
(Eigsti et al., 2006). We reasoned that dispositional abilities in delaying grati-
fication reflect, in part, individual differences in the relative activation of
the cool or “no-go” system over the hot (or go) system and that such stable
differences make it easier for individuals to enact strategies that dampen the
salience of the appetitive, hot, or go-related representations in self-control
situations. Thus, not surprisingly, we chose to use a well-established task of
impulse control, the go/no-go task, which despite striking differences at the
surface level bears some resemblance to the preschool delay of gratification
task (Casey et al., 1997; Fillmore & Rush, 2002; Fillmore, Rush, & Hays,
2002; Kaufman, Ross, Stein, & Garavan, 2003).
In the go/no-go task, participants are instructed to respond to target
(go) stimuli by pressing a response key, but to not press a response key to
other nontarget (no-go) stimuli. The ability to successfully refrain from mak-
ing responses on no-go trials (reflected by false alarm rate on no-go trials)
and the efficiency with which an individual is able to make a response on go
trials (reflected by reaction times on go trials), are behavioral indicators of
the ability to successfully and efficiently inhibit a response. Past work using

mind and brain in delay of gratification      157

13490-07_PT3_Ch06-3rdPgs.indd 157 11/15/13 1:43 PM

 the go/no-go task has found that children ages 7 to 12 show greater difficulty
than adults (twice as many errors overall and slower reaction time; Durston,
Thomas, Yang, et al., 2002), consistent with the idea that impulse control
becomes more efficient with development (Davidson, Amso, Anderson, &
Diamond, 2006; Keating & Bobbitt, 1978; see also Chapter 3, this volume).
Given that the original delay of gratification work showed that pre-
schoolers’ ability to delay gratification was affected by contextual factors
(e.g., when rewards are obscured from sight, delaying is easier than when
rewards are visible), we aimed to mimic this contextual effect in the go/no-go
paradigm. Specifically, we included a parametric manipulation of the number
of preceding consecutive go trials. This parametric manipulation essentially
primes the go response, which, in turn, increases conflict between the two
response options (to go or not to go) and thus increases the need for greater
cognitive control to accurately and efficiently perform the task. For example,
inhibiting a go response following only one go trial is easier than inhibiting a
go response after five consecutive go trials. Indeed, as the number of preced-
ing consecutive go trials increases, the proportion of false alarms on no-go
trials increases (Durston, Thomas, Worden, Yang, & Casey, 2002), as do reac-
tion times on go trials (Liston et al., 2006).
In our follow-up, we therefore asked the following: Would preschoolers’
ability to enact strategies that allow them to delay gratification predict their
ability, as adults, to effectively and efficiently refrain from responding to irrele-
vant stimuli in the go/no-go task? The results from the follow-up study revealed
a clear and remarkable convergence between the spontaneous attentional–
control strategies enacted by preschoolers in the delay of gratification situa-
tion and their adult impulse control abilities as reflected by the go/no-go task
(Eigsti et al., 2006). As shown in Figure 6.1, those who at age 4 directed their
attention away from tempting aspects of the situation (i.e., low temptation
focus), compared with those who directed their attention toward tempting
aspects of the situation (i.e., high temptation focus), were more efficient at sup-
pressing prepotent responses in the go/no-go task in adulthood. Specifically,
consistent with other research showing that longer reaction times on go trials
is a behavioral marker of difficulty with impulse control (Liston et al., 2006),
the low temptation focus group was faster on go responses without making
more errors. Moreover, the predictive ability of individual differences in pre-
school attentional–control strategies was most pronounced in conditions in
which a go trial was preceded by a greater number of consecutive go trials,
which presumably required greater impulse control. The results showed that
the difference between preschoolers with low temptation focus and high
temptation focus increased linearly as the need for impulse control increased
(i.e., when the go stimulus had been preceded by five or six consecutive go
trials). Overall, the results indicate that for individuals who as preschoolers

158       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 158 11/15/13 1:43 PM

 High Temptation Focus
Low Temptation Focus
350

Reaction Time (ms) 340

330

320

310

300

290

280
Go-1,2 Go-3,4 Go-5,6
Context
Figure 6.1. Mean reaction times on go trials as a function of preschool attentional-
control (high vs. low temptation focus) and as a function of the number of consecu-
tive preceding go trials. Bars represent 1 standard error above and below the mean.
Those who at age 4 showed better attentional control (low temptation focus) were
faster at responding to go trials, compared with those with worse attentional control
at age 4 (high temptation focus). This difference between effective versus ineffective
attentional control at age 4 was most pronounced in conditions in which resisting a
prepotent impulse was greatest (i.e., with greater number of consecutive preceding
go trials). ms = milliseconds. From “Attentional Control in Preschool Predicts Cogni-
tive Control at Age Eighteen,” by L.-M. Eigsti et al., 2006. Psychological Science, 17,
p. 482. Copyright 2006 by Wiley-Blackwell. Adapted with permission.

spontaneously enacted ineffective attentional–control strategies (high temp-

tation focus) to perform the go/no-go task accurately as adults, they needed
to compensate by slowing down their speed of responding, particularly in the
most difficult trials (i.e., following a large number of consecutive go trials).

Neural Correlates: Implicating the PFC and the Limbic System

Given that performance on the go/no-go task has been well character-
ized as involving activation in frontostriatal regions of the brain (Booth et al.,
2003; Casey et al., 1997; Davidson et al., 2004; Durston, Thomas, Worden,
et al., 2002; Durston, Thomas, Yang, et al., 2002; Durston et al., 2003; Konishi
et al., 1999; Vaidya et al., 1998), the behavioral findings from the Eigsti et al.
(2006) follow-up provocatively suggest that preschool delay ability may serve
as an early marker of the functional integrity of this circuitry. In recent work
involving a different cohort (Casey et al., 2011), we provide the first empirical
support for this proposition.

mind and brain in delay of gratification      159

13490-07_PT3_Ch06-3rdPgs.indd 159 11/15/13 1:43 PM

 In this study, participants had completed the delay of gratification pro-
cedure at age 4 during the 1960s and 1970s as well as various self-report
measures of self-control in their 20s and 30s. Participants were grouped into
those who exhibited consistently above-average self-control throughout the
various assessments (high delay group) and those who exhibited consistently
below-average self-control (low delay group). A subset of this sample agreed
to take part in the follow-up study.
In the first part of the study, participants completed two versions of the
go/no-go task: a cool and a hot version. The cool version was designed to
assess the ability to inhibit responses to cool stimuli, that is, stimuli that do
not inherently possess appetitive properties. Accordingly, stimuli were photo­
graphs of individual men and women with neutral emotional expressions.
One gender served as the go stimuli, and the opposite gender served as the
no-go stimuli (counterbalanced). The hot version of the go/no-go task was
designed to assess the ability to inhibit responses to stimuli that inherently
possess appetitive properties. Thus, the hot go/no-go task was identical to the
cool go/no-go task except that the go stimuli were photographs of faces with
emotional expressions (e.g., happy) and the no-go stimuli were photographs
of faces with a different emotional expression (e.g., fearful). Happy faces
are appetitive and automatically trigger approach-related responses (Hare,
Tottenham, Davidson, Glover, & Casey, 2005), whereas fearful faces trigger
attentional vigilance (Davis & Whalen, 2001) and defensive responses when
encountered in situations of anxiety (Grillon & Charney, 2011).
Given a growing body of research distinguishing between hot and
cool executive functions (Figner, Mackinlay, Wilkening, & Weber, 2009;
Somerville & Casey, 2010; see also Chapter 3, this volume), we predicted
that preschool delay ability would affect performance most strongly in the
hot (vs. cool) go/no-go task. Moreover, we reasoned that trials in which par-
ticipants were instructed to go on fearful faces and to not go on happy faces
would be particularly difficult, because this condition required inhibiting the
appetitive appeal of happy faces. Thus, we predicted that preschool delay
ability would be particularly diagnostic in these more difficult trials.
As shown in Figure 6.2, left panel, at the behavioral level, the high delay
group and the low delay group did not differ appreciably in their ability to
withhold a response in the cool go/no-go task, as expected. However, the high
delay group, compared with the low delay group, was much more effective at
withholding a response in the hot version of the task. Specifically, the primary
difference between the high and low delayers emerged in situations in which
they were required to inhibit a behavioral response to happy faces serving as
no-go stimuli, with high delayers exhibiting fewer false alarms than low delayers.
(The difference between the high and low delay groups was also observed when
the task was performed inside the scanner; Figure 6.2 right panel.)

160       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 160 11/15/13 1:43 PM

 Figure 6.2. Mean false alarm rate in the behavioral study (experiment 1: left panel)
and in the imaging study (experiment 2: right panel). In the go/no-go task, when the
cues are “cool” (neutral facial expressions), there was no significant difference in
performance between high versus low delayers. However, when the cues are “hot”
(emotional faces) low delayers made more errors than did high delayers. Error bars
denote 1 standard error above and below the mean. From “Behavioral and Neural
Correlates of Delay of Gratification 40 Years Later,” by B. J. Casey et al., 2011. Pro-
ceedings of the National Academy of Sciences of the United States of America, 108,
p. 15000. Copyright 2011 by the National Academy of Sciences. Adapted with
permission.

An investigation of the neural substrates of these differences focused

particularly on the role of the IFG, a region of the PFC involved in the reso-
lution of conflict between competing motor responses (Aron, Robbins, &
Poldrack, 2004; Casey, Tottenham, & Fossella, 2002) and among competing
representations in memory (Jonides & Nee, 2006). We hypothesized that in
the basic task facing participants—that is, resolving the decision of whether
to go or not to go—the IFG would be recruited more heavily, and more impor-
tant, that high delayers, compared with low delayers, would recruit the IFG
more heavily. Consistent with these hypotheses, as shown in Color Plate 9
(Panel A), the IFG was recruited more heavily on no-go trials compared with
go trials (and so was the left motor cortex, Panel B, and cerebellum, Panel C).
Moreover, as shown in Figure 6.3, compared with their low delaying counter­
parts, high delayers showed greater IFG activation in trials requiring the inhi-
bition of a response (nogo trials) than trials that did not (go trials).
In addition, given that the behavioral findings showed that the two
groups were statistically indistinguishable when the cues to be inhibited were
neutral but differed primarily in situations in which the cues to be inhibited

mind and brain in delay of gratification      161

13490-07_PT3_Ch06-3rdPgs.indd 161 11/15/13 1:43 PM

 Figure 6.3. Percent change in magnetic resonance (MR) signal to go and no-go
trials in the inferior frontal gyrus (IFG) in high and low delaying groups. High delay
ability predicts greater recruitment of the IFG when correctly inhibiting responses to
no-go trials (positive social cues) relative to go trials 40 years later.

were appetitive and prepotent, we investigated the extent to which limbic

structures were recruited in the go/no-go task. We hypothesized that acti-
vation of neural regions involved in the processing of rewards, such as the
ventral striatum, which is routinely involved in the processing of appetitive
stimuli (Delgado, 2007), would also distinguish the two groups. As shown in
Color Plate 10, the neural data indicated that the low delay group, compared
with the high delay group, showed greater activation of the ventral striatum
when happy faces served as no-go stimuli.

Implications for Theory and Research

on Delay of Gratification

There has been an explosion of research in cognitive neuroscience,

social neuroscience, and neuroeconomics focusing on the neural structures
involved in decision making, temporal discounting, cognitive control, and
reward sensitivity. The emerging picture is that delay of gratification ability is
affected by the interplay of limbic structures associated with the processing of
tempting cues, on the one hand, and the PFC and cingulate cortex associated
with the regulation of cognition, affect, and behavior, on the other (Bechara,
2005; LeDoux, 1996; McClure et al., 2004; McClure, Erikson, Laibson,
Loewenstein, & Cohen, 2007; Ochsner et al., 2004; Sanfey et al., 2003,
2006; Somerville & Casey, 2010). The recent findings from our research team
(Casey et al., 2011; Eigsti et al., 2006) further clarify the cognitive and neural

162       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 162 11/15/13 1:43 PM

 mechanisms enabling effective delay of gratification and how these mecha-
nisms relate to preschool delay ability assessed over two decades earlier. We
discuss below how this recent work contributes to current theorizing of the
neural mechanisms underlying delay of gratification.

Implications for the Cool System: The Role of Cognitive Control

The classic findings from the original preschooler delay of gratification

studies and recent work utilizing behavioral and neural measures to probe
cognitive control in adulthood converge to support the idea that delay of
gratification requires cognitive control. To effectively wait the entire 15 min-
utes in the classic delay of gratification situation as well as to efficiently and
correctly perform the computer-based go/no-go task, one must be able to
inhibit behaviors that promote impulsive action: In the delay task, a child
must inhibit the urge to eat the treat or ring the bell, and in the go/no-go
task, the adult must inhibit the impulse to press a button in response to a
prepotent, but task-irrelevant, cue.
Moreover, these findings indicate that individuals differ in their ability
to spontaneously recruit cognitive control, and there is evidence for some
remarkable stability in this ability over the life span. In the absence of any
explicit instructions, in two studies with two different cohorts (Casey et al.,
2011; Eigsti et al., 2006), the ability to spontaneously recruit strategies (e.g.,
distraction) in the delay of gratification situation in preschool predicted the
ability to efficiently inhibit responses to distracting, potent cues in the go/no-go
task in adulthood. Furthermore, providing more direct empirical evidence that
delay of gratification requires cognitive control, in the recent follow-up (Casey
et al., 2011), high delayers recruited areas of the IFG, which is involved in the
resolution of conflict, to a greater extent than low delayers. This stability is
impressive given the 20-year time span, the drastically different methods of
assessment (behavioral vs. reaction-based computer task), and the distinct
contexts and cues (food vs. happy faces).

Implications for the Hot System

In the recent imaging work (Casey et al., 2011), the low delay group
(compared with the high delay group) showed greater activation of limbic-
related neural structures, such as ventral striatum, involved in the processing
of rewards when happy faces served as no-go stimuli. One straightforward
explanation for this finding is that the representation of happy faces (a social
reward) was more salient and affective for low delayers than for high delayers.
This would be in line with the idea that low delayers may be more sensitive to
appetitive cues (happy faces) than are high delayers, and that this sensitivity

mind and brain in delay of gratification      163

13490-07_PT3_Ch06-3rdPgs.indd 163 11/15/13 1:43 PM

 promotes greater impulsive responding and makes the task of inhibiting
responses to happy faces more difficult (Herzberger & Dweck, 1978). Thus,
those with poorer delay ability may have chronically overactive limbic sys-
tems that tip the relative balance of the hot/cool system.
It is worth noting that preschool delay of gratification predicted perfor-
mance in the go/no-go task primarily in situations in which the cues to be
inhibited were appetitive and prepotent. In the Casey et al. (2011) follow-up,
the two groups were statistically indistinguishable when the cues to be inhib-
ited were neutral. That delay ability is most clearly observed in situations that
involve inhibition of affective, hot, prepotent stimuli, not just irrelevant (but
cool) stimuli, supports a growing body of research distinguishing between hot
and cool executive functions (Figner et al., 2009; Somerville & Casey, 2010;
see also Chapter 3, this volume).
Moreover, these findings are particularly interesting in light of develop-
mental models of cognitive control (Casey, Getz, & Galvan, 2008; Somerville
& Casey, 2010; see also Chapter 4, this volume) that aim to understand the
neural mechanisms that may heighten risky decision making during adoles-
cence. From this developmental perspective, although the PFC cognitive con-
trol system matures in a relatively linear fashion, the limbic affective system is
highly reactive during adolescence. Thus, the greater development of limbic
structures relative to PFC-related cognitive control structures biases sensitivity
to and the processing of immediate rewards (relative to delayed rewards) and
leads to greater risky decision making and behaviors. Such developmental mod-
els could be fruitfully applied to understanding individual differences (Carlson,
Zayas, & Guthormsen, 2009). In this view, adults with poor self-control might
be viewed as developmentally less mature and similar to adolescents in pos-
sessing less mature frontostriatal circuitry that predisposes them to greater risk.

Interactions Between the Cool and Hot Systems

Increasingly, cognitive neuroscience is focusing on interactions among

systems. So too did the initial conceptualization of the hot/cool framework.
An underlying theme of these perspectives is the focus on the balance between
the affective systems on the one hand and control systems on the other, and
how the two systems continuously interact and influence one another. Thus,
these approaches raise another, although not mutually exclusive, account
of the recent follow-up (Casey et al., 2011). They raise the possibility that
perhaps the differential activation of the ventral striatum is not simply a
cause but also a consequence of differences in the ability to engage in cognitive
control processes regulated by the PFC cognitive system. For example, high
delayers’ ability to effectively recruit cognitive control processes, and thus
resolve conflict between competing representations, may have dampened the

164       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 164 11/15/13 1:43 PM

 sensitivity to and processing of appetitive stimuli. Applied to Casey et al.’s
(2011) findings, this reasoning suggests that the ability to efficiently resolve
conflict between go responses to fearful faces and no-go responses to happy
faces may have weakened the appetitive aspects of the no-go stimuli (happy
faces). That is, if the conflict between the behavioral options to the stimuli is
resolved quickly, then the representation of the distracting, appetitive no-go
stimuli (happy faces) will be quickly expunged from working memory and
be less likely to activate limbic structures involved in the processing of pre-
potent rewards. In contrast, if there is greater difficulty in resolving the con-
flict between the two behavioral options, the representation of the no-go
stimuli (happy faces), which should be inhibited, will remain in working
memory for a longer period of time and thus more strongly activate limbic
structures involved in the processing of rewards. Thus, it is possible that high
delayers’ ability to engage in cognitive control decreased the salience of the
appetitive stimuli and made the task of inhibiting a response less difficult.
Although further research is needed, such a possibility is evident in past
work showing that an important benefit of effortful acts of control is that they
make the very task of resisting temptation easier. They do so, for example, by
preventing in-depth processing of the appetitive stimuli (via ignore/distraction
strategies) and by dampening the appetitive and hot representations of tempta-
tions in working memory (via reconstrual). These strategies, in effect, decrease
the psychological temptation, or “pull,” of the situation and thus make the
goal of delaying less effortful. Recent work provides additional evidence that
difficulties in enacting self-control may enhance sensitivity to affective cues.
Experimentally manipulating the ability to engage in self-regulation (e.g., via
cognitive load manipulations) leads to more impulsive choices (Friese, Hoffman,
& Wanke, 2008; Hofmann, Friese, & Strack, 2009; Shiv & Feodorikhin, 2002;
Ward & Mann, 2000). Vohs, Baumeister, and Schmeichel (2012) showed that
participants induced to engage in less self-regulation reported stronger emo-
tional reactions to valenced images and film clips, stronger feelings of pain
in a cold pressor task, and a greater desire to eat cookies during a tasting test.
Although effortful cognitive control is a prerequisite for the enactment of self-
control strategies, such acts make delaying gratification much easier. Indeed,
it is ironic that to make delaying of gratification easier, and thereby lessen
the need for cognitive control, one needs cognitive control.

Conclusion

The ability to delay gratification is necessary for achieving success in a

number of life domains. A growing body of research, starting with the origi-
nal delay of gratification experiments, has been identifying the psychological

mind and brain in delay of gratification      165

13490-07_PT3_Ch06-3rdPgs.indd 165 11/15/13 1:43 PM

 and neural mechanisms that enable individuals to engage in these more
challenging goal pursuits. Speaking to the psychological level, this work
shows that strategies such as distraction and reappraisal enable delay of
gratification. Moreover, the ability to implement these strategies depends
critically on cognitive control—the ability to regulate (suppress, replace,
enhance, maintain) actions, thoughts, and feelings for the sake of long-term
goals. Recent behavioral and neuroscientific findings are consistent with
the hypothesis that those with difficulties in delaying gratification have less
effective cognitive control abilities. During a task that requires inhibition
of affective cues, those with difficulties delaying gratification are also more
strongly affected by the appetitive aspects of immediate rewards. The cumu-
lative effect is that those with poor delay ability are likely to place greater
relative weight on the present, immediately available rewards than on the
imagined, delayed rewards and decrease their ability to delay gratification
(McClure et al., 2004). These recent neuroscientific findings are particu-
larly interesting in light of models of cognitive control (Casey et al., 2008;
Somerville & Casey, 2010; see also Chapter 4, this volume) and suggest that
adults with poor self-control may possess less mature frontostriatal circuitry
that predisposes them to greater risk.

References

Ainslie, G. (1975). Specious reward: A behavioral theory of impulsiveness and

impulse control. Psychological Bulletin, 82, 463–496. doi:10.1037/h0076860
Armony, J. L. and LeDoux, J. E. (1997) How the brain processes emotional informa-
tion. Annals of the New York Academy of Sciences 821, 259–270.
Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition and the right
inferior frontal cortex. Trends in Cognitive Sciences, 8, 170–177. doi:10.1016/
j.tics.2004.02.010
Ayduk, O., Downey, G., & Kim, M. (2001). Rejection sensitivity and depressive
symptoms in women. Personality and Social Psychology Bulletin, 27, 868–877.
doi:10.1177/0146167201277009
Ayduk, O., Downey, G., Testa, A., Yen, Y., & Shoda, Y. (1999). Does rejection
elicit hostility in rejection sensitive women? Social Cognition, 17, 245–271.
doi:10.1521/soco.1999.17.2.245
Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey, G., Peake, P., & Rodriguez,
M. (2000). Regulating the interpersonal self: Strategic self-regulation for cop-
ing with rejection sensitivity. Journal of Personality and Social Psychology, 79,
776–792. doi:10.1037/0022-3514.79.5.776
Ayduk, O., Zayas, V., Downey, G., Cole, A. B., Shoda, Y., & Mischel, W. (2008).
Rejection sensitivity and executive control: Joint predictions of borderline per-

166       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 166 11/15/13 1:43 PM

 sonality features. Journal of Research in Personality, 42, 151–168. doi:10.1016/
j.jrp.2007.04.002
Baddeley, A. (1998). Recent developments in working memory. Current Opinion in
Neurobiology, 8, 234–238. doi:10.1016/S0959-4388(98)80145-1
Baddeley, A. D. (1986). Working memory. Oxford, England: Clarendon Press.
Bantick, S. J., Wise, R. G., Ploghaus, A., Clare, S., Smith, S. M., & Tracey, I. (2002).
Imaging how attention modulates pain in humans using functional MRI. Brain:
A Journal of Neurology, 125, 310–319. doi:10.1093/brain/awf022
Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive
functions: Constructing a unifying theory of ADHD. Psychological Bulletin, 121,
65–94. doi:10.1037/0033-2909.121.1.65
Bechara, A. (2005). Decision making, impulse control and loss of willpower to
resist drugs: A neurocognitive perspective. Nature Neuroscience, 8, 1458–1463.
doi:10.1038/nn1584
Bishop, S. J., Jenkins, R., & Lawrence, A. D. (2007). Neural processing of fearful
faces: Effects of anxiety are gated by perceptual capacity limitations. Cerebral
Cortex, 17, 1595–1603. doi:10.1093/cercor/bhl070
Blair, C. (2010). Stress and the development of self-regulation in context. Child
Development Perspectives, 4, 181–188. doi:10.1111/j.1750-8606.2010.00145.x
Blair, C., & Raver, C. C. (2012). Child development in the context of adversity.
American Psychologist, 67, 309–318. doi:10.1037/a0027493
Blais, A. R., & Weber, E. U. (2001). Domain-specificity and gender differences in deci-
sion making. Risk Decision & Policy, 6, 47–69. doi:10.1017/S1357530901000254
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and health-related behaviors:
A meta-analysis of the leading behavioral contributors to mortality. Psycho­
logical Bulletin, 130, 887–919. doi:10.1037/0033-2909.130.6.887
Bonato, D. P., & Boland, F. J. (1983). Delay of gratification in obese children. Addic­
tive Behaviors, 8, 71–74. doi:10.1016/0306-4603(83)90059-X
Booth, J. R., Burman, D. D., Meyer, J. R., Lei, Z., Trommer, B. L., Davenport,
N. D., . . . Mesulam, M. M. (2003). Neural development of selective atten-
tion and response inhibition. NeuroImage, 20, 737–751. doi:10.1016/S1053-
8119(03)00404-X
Botvinick, M. M., Braver, T., Barch, D., Carter, C., & Cohen, J. (2001). Con-
flict monitoring and cognitive control. Psychological Review, 108, 624–652.
doi:10.1037/0033-295X.108.3.624
Breiter, H. C., Etcoff, N. L., Whalen, P. J., Kennedy, W. A., Rauch, S. L., Buckner,
R. L., . . . Rosen, B. R. (1996). Response and habituation of the human amygdala
during visual processing of facial expression. Neuron, 17, 875–887. doi:10.1016/
S0896-6273(00)80219-6
Brown, S. M., Henning, S., & Wellman, C. L. (2005). Short-term, mild stress alters
dendritic morphology in rat medial prefrontal cortex. Cerebral Cortex, 15,
1714–1722. doi:10.1093/cercor/bhi048

mind and brain in delay of gratification      167

13490-07_PT3_Ch06-3rdPgs.indd 167 11/15/13 1:43 PM

 Bruce, A. S., Black, W. R., Bruce, J. M., Daldalian, M., Martin, L. E., & Davis, A. M.
(2011). Ability to delay gratification and BMI in preadolescence. Obesity, 19,
1101–1102. doi:10.1038/oby.2010.297
Bunge, S. A., & Zelazo, P. D. (2006). A brain-based account of the development of
rule use in childhood. Current Directions in Psychological Science, 15, 118–121.
doi:10.1111/j.0963-7214.2006.00419.x
Cacioppo, J. T., & Gardner, W. L. (1999). Emotion. Annual Review of Psychology, 50,
191–214. doi:10.1146/annurev.psych.50.1.191
Carlson, S. M., Zayas, V., & Guthormsen, A. (2009). Neural correlates of decision
making on a gambling task. Child Development, 80, 1076–1096. doi:10.1111/
j.1467-8624.2009.01318.x
Carretié, L., Hinojosa, J. A., Martín-Loeches, M., Mercado, F., & Tapia, M. (2004).
Automatic attention to emotional stimuli: Neural correlates. Human Brain
Mapping, 22, 290–299. doi:10.1002/hbm.20037
Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent brain. Developmental
Review, 28, 62–77. doi:10.1016/j.dr.2007.08.003
Casey, B. J., Somerville, L. H., Gotlib, I., Ayduk, O., Franklin, N., Askren,
M. K., . . . Shoda, Y. (2011). Behavioral and neural correlates of delay of gratifi-
cation 40 years later. Proceedings of the National Academy of Sciences of the United
States of America, 108, 14998–15003. doi:10.1073/pnas.1108561108
Casey, B. J., Tottenham, N., & Fossella, J. (2002). Clinical, imaging, lesion, and
genetic approaches toward a model of cognitive control. Developmental Psycho­
biology, 40, 237–254. doi:10.1002/dev.10030
Casey, B. J., Trainor, R. J., Orendi, J. L., Schubert, A. B., Nystrom, L. N., Giedd,
J. N., . . . Rapoport, J. L. (1997). A developmental functional MRI study of pre-
frontal activation during performance of a go-no-go task. Journal of Cognitive
Neuroscience, 9, 835–847. doi:10.1162/jocn.1997.9.6.835
Childress, A. R., Mozley, P. D., McElgin, W., Fitgerald, J., Reivich, M., & O’Brien,
C. P. (1999). Limbic activation during cue-induced cocaine craving. The Ameri­
can Journal of Psychiatry, 156, 11–18.
Cohen, J. D., & Servan-Schreiber, D. (1992). Context, cortex and dopamine: A
connectionist approach to behavior and biology in schizophrenia. Psychological
Review, 99, 45–77. doi:10.1037/0033-295X.99.1.45
Compton, R. J. (2003). The interface between emotion and attention: A review of
evidence from psychology and neuroscience. Behavioral and Cognitive Neuro­
science Reviews, 2, 115–129. doi:10.1177/1534582303002002003
Cunningham, W. A., Johnson, M. K., Raye, C. L., Gatenby, J. C., Gore, J. C., &
Banaji, M. R. (2004). Separable neural components in the processing of Black
and White faces. Psychological Science, 15, 806–813.
David, S. P., Munafò, M. R., Johansen-Berg, H., MacKillop, J., Sweet, L. H., Cohen,
R. A., . . . Walton, R. T. (2007). Effects of acute nicotine abstinence on cue-
elicited ventral striatum/nucleus accumbens activation in female cigarette

168       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 168 12/9/13 11:56 AM

 smokers: A functional magnetic resonance imaging study. Brain Imaging and
Behavior, 1, 43–57. doi:10.1007/s11682-007-9004-1
Davidson, M. C., Amso, D., Anderson, L. A., & Diamond, A. (2006). Development
of cognitive control and executive functions from 4 to 13 years: Evidence from
manipulations of memory, inhibition, and task switching. Neuropsychologia, 44,
2037–2078. doi:10.1016/j.neuropsychologia.2006.02.006
Davidson, M. C., Horvitz, J. C., Tottenham, N., Fossella, J. A., Watts, R., Ulug,
A. M., & Casey, B. J. (2004). Differential cingulate and caudate activation
following unexpected nonrewarding stimuli. NeuroImage, 23, 1039–1045.
doi:10.1016/j.neuroimage.2004.07.049
Davis, M., & Whalen, P. J. (2001). The amygdala: Vigilance and emotion. Molecular
Psychiatry, 6, 13–34. doi:10.1038/sj.mp.4000812
Delgado, M. R. (2007). Reward-related responses in the human striatum. Annals of
the New York Academy of Sciences, 1104, 70–88. doi:10.1196/annals.1390.002
Delgado, M. R., Locke, H. M., Stenger, V. A., & Fiez, J. A. (2003). Dorsal striatum
responses to reward and punishment: Effects of valence and magnitude manipu-
lations. Cognitive, Affective & Behavioral Neuroscience, 3, 27–38. doi:10.3758/
CABN.3.1.27
Delgado, M. R., Nystrom, L. E., Fissell, C., Noll, D. C., & Fiez, J. A. (2000). Tracking
the hemodynamic responses to reward and punishment in the striatum. Journal
of Neurophysiology, 84, 3072–3077.
Demos, K. E., Kelley, W. M., & Heatherton, T. F. (2011). Dietary restraint violations
influence reward responses in nucleus accumbens and amygdala. Journal of Cog­
nitive Neuroscience, 23, 1952–1963. doi:10.1162/jocn.2010.21568
Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual atten-
tion. Annual Review of Neuroscience, 18, 193–222. doi:10.1146/annurev.
ne.18.030195.001205
Downey, G., & Feldman, S. (1996). Implications of rejection sensitivity for inti-
mate relationships. Journal of Personality and Social Psychology, 70, 1327–1343.
doi:10.1037/0022-3514.70.6.1327
Downey, G., Freitas, A., Michaelis, B., & Khouri, H. (1998). The self-fulfilling
prophecy in close relationships: Rejection sensitivity and rejection by romantic
partners. Journal of Personality and Social Psychology, 75, 545–560. doi:10.1037/
0022-3514.75.2.545
Downey, G., Lebolt, A., Rincon, C., & Freitas, A. L. (1998). Rejection sensitivity
and children’s interpersonal difficulties. Child Development, 69, 1074–1091.
Durston, S., Thomas, K. M., Worden, M. S., Yang, Y., & Casey, B. J. (2002). The
effect of preceding context on inhibition: An event-related fMRI study. Neuro­
Image, 16, 449–453. doi:10.1006/nimg.2002.1074
Durston, S., Thomas, K. M., Yang, Y., Ulug, A. M., Zimmerman, R. D., & Casey, B. J.
(2002). A neural basis for the development of inhibitory control. Developmental
Science, 5, F9–F16. doi:10.1111/1467-7687.00235

mind and brain in delay of gratification      169

13490-07_PT3_Ch06-3rdPgs.indd 169 11/15/13 1:43 PM

 Durston, S., Tottenham, N. T., Thomas, K. M., Davidson, M. C., Eigsti, I. M.,
Yang, Y., . . . Casey, B. J. (2003). Differential patterns of striatal activation in
young children with or without ADHD. Biological Psychiatry, 53, 871–878.
doi:10.1016/S0006-3223(02)01904-2
Eigsti, I.-M., Zayas, V., Mischel, W., Shoda, Y., Ayduk, O., Dadlan, M. B., . . . Casey,
B. J. (2006). Attentional control in preschool predicts cognitive control
at age eighteen. Psychological Science, 17, 478–484. doi:10.1111/j.1467-
9280.2006.01732.x
Eisenberger, N., Smith, C. L., Sadovsky, A., & Spinrad, T. L. (2004). Effortful
control: Reactions with emotion regulation, adjustment, and socialization in
childhood. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation:
Research, theory, and applications (pp. 259–282). New York, NY: Guilford Press.
Etkin, A., Egner, T., Peraza, D. M., Kandel, E. R., & Hirsch, J. (2006). Resolving emo-
tional conflict: A role for the rostral anterior cingulate cortex in modulating activ-
ity in the amygdala. Neuron, 51, 871–882. doi:10.1016/j.neuron.2006.07.029
Evans, G. W., & Kim, P. (2013). Childhood poverty, chronic stress, self-regulation,
and coping. Child Development Perspectives, 7, 43–48.
Everitt, B. J., Dickinson, A., & Robbins, T. W. (2001). The neuropsychological
basis of addictive behaviour. Brain Research Reviews, 36, 129–138. doi:10.1016/
S0165-0173(01)00088-1
Everitt, B. J., & Robbins, T. W. (2005). Neural systems of reinforcement for drug addic-
tion: From actions to habits to compulsion. Nature Neuroscience, 8, 1481–1489.
doi:10.1038/nn1579
Figner, B., Mackinlay, R. J., Wilkening, F., & Weber, E. U. (2009). Affective and
deliberative processes in risky choice: Age differences in risk taking in the
Columbia Card Task. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 35, 709–730. doi:10.1037/a0014983
Figner, B., & Weber, E. U. (2011). Who takes risks, when, and why? Determi-
nants of risk taking. Current Directions in Psychological Science, 20, 211–216.
doi:10.1177/0963721411415790
Fillmore, M. T., & Rush, C. R. (2002). Impaired inhibitory control of behavior in
chronic cocaine users. Drug and Alcohol Dependence, 66, 265–273. doi:10.1016/
S0376-8716(01)00206-X
Fillmore, M. T., Rush, C. R., & Hays, L. (2002). Acute effects of oral cocaine on
inhibitory control of behavior in humans. Drug and Alcohol Dependence, 67,
157–167. doi:10.1016/S0376-8716(02)00062-5
Fishbach, A., & Trope, Y. (2005). The substitutability of external control and self-
control in overcoming temptation. Journal of Experimental Social Psychology, 41,
256–270. doi:10.1016/j.jesp.2004.07.002
Francis, L. A., & Susman, E. J. (2009). Self-regulation and rapid weight gain in chil-
dren from age 3 to 12 years. Archives of Pediatrics & Adolescent Medicine, 163,
297–302. doi:10.1001/archpediatrics.2008.579

170       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 170 11/15/13 1:43 PM

 Friese, M., Hofmann, W., & Wänke, M. (2008). When impulses take over: Moder-
ated predictive validity of implicit and explicit attitude measures in predicting
food choice and consumption behavior. British Journal of Social Psychology, 47,
397–419
Galvan, A., Hare, T. A., Davidson, M., Spicer, J., Glover, G., & Casey, B. J. (2005).
The role of ventral frontostriatal circuitry in reward-based learning in humans. The
Journal of Neuroscience, 25, 8650–8656. doi:10.1523/JNEUROSCI.2431-05.2005
Grillon, C., & Charney, D. R. (2011). In the face of fear: Anxiety sensitizes defensive
responses to fearful faces. Psychophysiology, 48, 1745–1752. doi:10.1111/j.1469-
8986.2011.01268.x
Gyurak, A., & Ayduk, O. (2007). Defensive physiological reactions to rejection: The
effect of self-esteem and attentional control on startle responses. Psychological
Science, 18, 886–892. doi:10.1111/j.1467-9280.2007.01996.x
Gyurak, A., Goodkind, M. S., Kramer, J. H., Miller, B. L., & Levenson, R. W. (2012).
Executive functions and the down-regulation and up-regulation of emotion.
Cognition & Emotion, 26, 103–118.
Hackman, D. A., Farah, M. J., & Meaney, M. J. (2010). Socioeconomic status
and the brain: Mechanistic insights from human and animal research. Nature
Reviews Neuroscience, 11, 651–659. doi:10.1038/nrn2897
Hare, T. A., Camerer, C. F., & Rangel, A. (2009). Self-control in decision-making
involves modulation of the vmPFC valuation system. Science, 324, 646–648.
doi:10.1126/science.1168450
Hare, T. A., Tottenham, N., Davidson, M. C., Glover, G. H., & Casey, B. J. (2005).
Contributions of amygdala and striatal activity in emotion regulation. Biological
Psychiatry, 57, 624–632. doi:10.1016/j.biopsych.2004.12.038
Heatherton, T. F. (2011). Neuroscience of self and self-regulation. Annual Review of
Psychology, 62, 363–390. doi:10.1146/annurev.psych.121208.131616
Herzberger, S. D., & Dweck, C. S. (1978). Attraction and delay of gratification. Jour­
nal of Personality, 46, 215–227. doi:10.1111/j.1467-6494.1978.tb00176.x
Hinson, J. M., Jameson, T. L., & Whitney, P. (2003). Impulsive decision making
and working memory. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 29, 298–306. doi:10.1037/0278-7393.29.2.298
Hofmann, W., Friese, M., & Strack, F. (2009). Impulse and self-control from a
dual-systems perspective. Perspectives on Psychological Science, 4, 162–176.
doi:10.1111/j.1745-6924.2009.01116.x
Jackson, D. C., Muller, C. J., Dolski, I., Dalton, K. M., Nitschke, J. B., Urry,
H. L., . . . Davidson, R. J. (2003). Now you feel it, now you don’t: Frontal brain
electrical asymmetry and individual differences in emotion regulation. Psycho­
logical Science, 14, 612–617. doi:10.1046/j.0956-7976.2003.psci_1473.x
James, W. (1950). The principles of psychology. New York, NY: Dover. (Original work
published 1890).

mind and brain in delay of gratification      171

13490-07_PT3_Ch06-3rdPgs.indd 171 11/15/13 1:43 PM

 Johnstone, T., van Reekum, C. M., Urry, H. L., Kalin, N. H., & Davidson, R. J.
(2007). Failure to regulate: Counterproductive recruitment of top-down
prefrontal-subcortical circuitry in major depression. Journal of Neuroscience,
27, 8877–8884. doi:10.1523/JNEUROSCI.2063-07.2007
Jonides, J., & Nee, D. E. (2006). Brain mechanisms of proactive interference in
working memory. Neuroscience, 139, 181–193. doi:10.1016/j.neuroscience.
2005.06.042
Kaufman, J. N., Ross, T. J., Stein, E. A., & Garavan, H. (2003). Cingulate hypo­activity
in cocaine users during a GO–NOGO task as revealed by event-related functional
magnetic resonance imaging. The Journal of Neuroscience, 23, 7839–7843.
Keating, D. P., & Bobbitt, B. L. (1978). Individual and developmental differences
in cognitive-processing components of mental ability. Child Development, 49,
155–167. doi:10.2307/1128604
Kerns, J. G., Cohen, J. D., MacDonald, A. W., III, Cho, R. Y., Stenger, V. A., &
Carter, C. S. (2004). Anterior cingulate conflict monitoring and adjustments in
control. Science, 303, 1023–1026. doi:10.1126/science.1089910
Knutson, B., Adams, C. A., Fong, C. W., & Hommer, D. (2001). Anticipation of
increasing monetary reward selectively recruits nucleus accumbens. The Journal
of Neuroscience, 21, RC159.
Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M., & Miyashita, Y.
(1999). Common inhibitory mechanism in human inferior prefrontal cortex
revealed by event-related functional MRI. Brain: A Journal of Neurology, 122,
981–991. doi:10.1093/brain/122.5.981
Kubzansky, L. D., Martin, L. T., & Buka, S. L. (2009). Early manifestations of person-
ality and adult health: A life course perspective. Health Psychology, 28, 364–372.
doi:10.1037/a0014428
LeDoux, J. (1996). Emotional networks and motor control: A fearful view. Progress in
Brain Research, 107, 437–446. doi:10.1016/S0079-6123(08)61880-4
Lieberman, M. D. (2007). The X- and C-systems: The neural basis of automatic and
controlled social cognition. In E. Harmon-Jones & P. Winkelman (Eds.), Funda­
mentals of social neuroscience (pp. 290–315). New York, NY: Guilford Press.
Lieberman, M. D., Gaunt, R., Gilbert, D. T., & Trope, Y. (2002). Reflexion and reflec­
tion: A social cognitive neuroscience approach to attributional inference. In M. P.
Zanna (Ed.), Advances in experimental social psychology (Vol. 34, pp. 199–249).
New York, NY: Academic Press. doi:10.1016/S0065-2601(02)80006-5
Liston, C., Miller, M. M., Goldwater, D. S., Radley, J. J., Rocher, A. B., Hof,
P. R., . . . McEwen, B. S. (2006). Stress-induced alterations in prefrontal corti-
cal dendritic morphology predict selective impairments in perceptual atten-
tional set-shifting. The Journal of Neuroscience, 26, 7870–7874. doi:10.1523/
JNEUROSCI.1184-06.2006
McClure, S. M., Erikson, K. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D.
(2007). Time discounting for primary rewards. The Journal of Neuroscience, 27,
5796–5804. doi:10.1523/JNEUROSCI.4246-06.2007

172       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 172 11/15/13 1:43 PM

 McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D. (2004). Separate
neural systems value immediate and delayed monetary rewards. Science, 306,
503–507. doi:10.1126/science.1100907
McEwen, B. S., & Gianaros, P. J. (2010). Central role of the brain in stress and adap-
tation: Links to socioeconomic status, health, and disease. Annals of the New York
Academy of Sciences, 1186, 190–222. doi:10.1111/j.1749-6632.2009.05331.x
McRae, K., Hughes, B., Chopra, S., Gabrieli, J. D. E., Gross, J. J., & Ochsner, K. N.
(2010). The neural bases of distraction and reappraisal. Journal of Cognitive
Neuro­science, 22, 248–262. doi:10.1162/jocn.2009.21243
Mellers, B., Schwartz, A., & Weber, E. (1997). Do risk attitudes reflect in the eye of
the beholder? In A. A. J. Marley (Ed.), Choice, decision, and measurement: Essays
in honor of R. Duncan Luce (pp. 57–71). Mahway, NJ: Erlbaum.
Metcalfe, J., & Jacobs, W. J. (1996). A “hot-system/cool-system” view of memory
under stress. PTSD Research Quarterly, 7(2), 1–6.
Metcalfe, J., & Jacobs, W. J. (1998). Emotional memory: The effects of stress on
“cool” and “hot” memory systems. In D. L. Medin (ed.), The psychology of learn-
ing and motivation: Advances in research and theory (Vol. 38, pp. 187–222). San
Diego, CA: Academic Press.
Metcalfe, J., & Mischel, W. (1999). A hot/cool system analysis of delay of gratifica-
tion: Dynamics of willpower. Psychological Review, 106, 3–19. doi:10.1037/0033-
295X.106.1.3
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex
function. Annual Review of Neuroscience, 24, 167–202. doi:10.1146/annurev.
neuro.24.1.167
Mischel, W. (1974). Process in delay of gratification. In L. Berkowitz (Ed.), Processes
in delay of gratification (Vol. 7, pp. 249–292). Advances in experimental social psy-
chology. San Diego, CA: Academic Press.
Mischel, W., & Ayduk, O. (2002). Self-regulation in a cognitive-affective personal-
ity system: Attentional control in the service of the self. Self and Identity, 1,
113–120.
Mischel, W., & Ayduk, O. (2004). Willpower in a cognitive-affective processing sys-
tem: The dynamics of delay of gratification. In R. F. Baumeister & K. D. Vohs
(Eds.), Handbook of self-regulation: Research, theory, and applications (pp. 99–129).
New York, NY: Guilford Press.
Mischel, W., Ayduk, O. N., Berman, M., Casey, B. J., Jonides, J., Kross, E., . . . Shoda, Y.
(2011). “Willpower” over the life span: Decomposing impulse control. Social
Cognitive and Affective Neuroscience, 6, 252–256. doi:10.1093/scan/nsq081
Mischel, W., & Baker, N. (1975). Cognitive appraisals and transformations in delay
behavior. Journal of Personality and Social Psychology, 31, 254–261. doi:10.1037/
h0076272
Mischel, W., & Ebbesen, E. (1970). Attention in delay of gratification. Journal of
Personality and Social Psychology, 16, 329–337. doi:10.1037/h0029815

mind and brain in delay of gratification      173

13490-07_PT3_Ch06-3rdPgs.indd 173 12/9/13 11:57 AM

 Mischel, W., Ebbesen, E. B., & Zeiss, A. R. (1972). Cognitive and attentional mech-
anisms in delay of gratification. Journal of Personality and Social Psychology, 21,
204–218. doi:10.1037/h0032198
Mischel, W., & Moore, B. (1973). Effects of attention to symbolically presented
rewards on self. Journal of Personality and Social Psychology, 28, 172–179.
doi:10.1037/h0035716
Mischel, W., Shoda, Y., & Rodriguez, M. L. (1989). Delay of gratification in children.
Science, 244, 933–938. doi:10.1126/science.2658056
Mischel, W., & Underwood, B. (1974). Instrumental ideation in delay of gratifica-
tion. Child Development, 45, 1083–1088.
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington,
H., . . . Caspi, A. (2011). A gradient of childhood self-control predicts health,
wealth, and public safety. Proceedings of the National Academy of Sciences of the
United States of America, 108, 2693–2698. doi:10.1073/pnas.1010076108
Morris, J. S., Öhman, A., & Dolan, R. J. (1998). Conscious and unconscious emotional
learning in the human amygdala. Nature, 393, 467–470. doi:10.1038/30976
Ochsner, K. N., & Gross, J. J. (2007). The neural architecture of emotion regulation.
In J. J. Gross & R. Buck (Eds.), The handbook of emotion regulation (pp. 87–109).
New York, NY: Guilford Press.
Ochsner, K. N., Ray, R. D., Cooper, J. C., Robertson, E. R., Chopra, S., Gabrieli,
J. D. E., & Gross, J. J. (2004). For better or for worse: Neural systems supporting
the cognitive down- and up-regulation of negative emotion. NeuroImage, 23,
483–499. doi:10.1016/j.neuroimage.2004.06.030
O’Reilly, R. C. (2006). Biologically based computational models of high level cogni-
tion. Science, 314, 91–94. doi:10.1126/science.1127242
Phelps, E. A., & LeDoux, J. E. (2005). Contributions of the amygdala to emotion
processing: From animal models to human behavior. Neuron, 48, 175–187.
doi:10.1016/j.neuron.2005.09.025
Polivy, J. (1998). The effects of behavioral inhibition: Integrating internal cues,
cognitive behavior, and affect. Psychological Inquiry, 9, 181–204. doi:10.1207/
s15327965pli0903_1
Reyna, V. F., & Farley, F. (2006). Risk and rationality in adolescent decision making:
Implications for theory, practice, and public policy. Psychological Science in the
Public Interest, 7, 1–44.
Reyna, V. F., & Rivers, S. E. (2008). Current theories of risk and rational decision
making. Developmental Review, 28, 1–11. doi:10.1016/j.dr.2008.01.002
Rodriguez, M. L., Mischel, W., & Shoda, Y. (1989). Cognitive person variables in
the delay of gratification of older children at risk. Journal of Personality and Social
Psychology, 57, 358–367. doi:10.1037/0022-3514.57.2.358
Rothbart, M. K., Ellis, L. K., & Posner, M. I. (2004). Temperament and self-regulation.
In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research,
theory, and applications (pp. 357–370). New York, NY: Guilford Press.

174       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 174 11/15/13 1:43 PM

 Rothemund, Y., Preuschhof, C., Bohner, G., Bauknecht, H. C., Klingebiel, R.,
Flor, H., & Klapp, B. F. (2007). Differential activation of the dorsal striatum by
high-calorie visual food stimuli in obese individuals. NeuroImage, 37, 410–421.
doi:10.1016/j.neuroimage.2007.05.008
Rowe, J. B., Toni, I., Josephs, O., Frackowiak, R. S. J., & Passingham, R. E. (2000).
The prefrontal cortex: Response selection or maintenance within working
memory? Science, 288, 1656–1660. doi:10.1126/science.288.5471.1656
Sanfey, A. G., Hastie, R., Colvin, M. K., & Grafman, J. (2003). Phineas gauged:
Decision-making and the frontal lobes. Neuropsychologia, 41, 1218–1229.
doi:10.1016/S0028-3932(03)00039-3
Sanfey, A. G., Loewenstein, G., McClure, S. M., & Cohen, J. D. (2006). Neuro-
economics: Cross-currents in research on decision-making. Trends in Cognitive
Sciences, 10, 108–116. doi:10.1016/j.tics.2006.01.009
Sapolsky, R. M. (1996). Why stress is bad for your brain. Science, 273, 749–750.
doi:10.1126/science.273.5276.749
Schlam, T. R., Wilson, N. L., Shoda, Y., Mischel, W., & Ayduk, O. (2013). Pre-
schoolers’ delay of gratification predicts their body mass 30 years later. The Jour-
nal of Pediatrics, 162, 90–93. doi:10.1016/j.jpeds.2012.06.049
Seeyave, D. M., Coleman, S., Appugliese, D., Corwyn, R. F., Bradley, R. H., Davidson,
N. S., . . . Lumeng, J. C. (2009). Ability to delay gratification at age 4 years and
risk of overweight at age 11 years. Archives of Pediatrics & Adolescent Medicine,
163, 303–308. doi:10.1001/archpediatrics.2009.12
Shallice, T. (1988). From neuropsychology to mental structure. Cambridge, England:
Cambridge University Press. doi:10.1017/CBO9780511526817
Shiv, B., & Fedorikhin, A. (1999), Heart and mind in conflict: The interplay of affect
and cognition in consumer decision making. Journal of Consumer Research, 26,
278–292.
Shoda, Y., Mischel, W., & Peake, P. K. (1990). Predicting adolescent cognitive
and social competence from preschool delay of gratification: Identifying diag-
nostic conditions. Developmental Psychology, 26, 978–986. doi:10.1037/0012-
1649.26.6.978
Smith, E. R., & DeCoster, J. (2000). Dual-process models in social and cognitive
psychology: conceptual integration and links to underlying memory systems.
Personality and Social Psychology Review, 4, 108–131. doi:10.1207/S153279
57PSPR0402_01
Somerville, L. H., & Casey, B. J. (2010). Developmental neurobiology of cognitive
control and motivational systems. Current Opinion in Neurobiology, 20, 236–241.
doi:10.1016/j.conb.2010.01.006
Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., & Woolard, J. (2008).
Age differences in sensation seeking and impulsivity as indexed by behavior and
self-report: Evidence for a dual systems model. Developmental Psychology, 44,
1764–1778. doi:10.1037/a0012955

mind and brain in delay of gratification      175

13490-07_PT3_Ch06-3rdPgs.indd 175 12/9/13 11:57 AM

 Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts
good adjustment, less pathology, better grades, and interpersonal success. Journal
of Personality, 72, 271–322. doi:10.1111/j.0022-3506.2004.00263.x
Thiruchselvam, R., Hajcak, G., & Gross, J. J. (2012). Looking inwards: Shifting
attention within working memory representations alters emotional responses.
Psychological Science, 23, 1461–1466. doi:10.1177/0956797612449838
Vaidya, C. J., Austin, G., Kirkorian, G., Ridlehuber, H. W., Desmond, J. E., Glover,
G. H., & Gabrieli, J. D. E. (1998). Selective effects of methylphenidate in
attention deficit hyperactivity disorder: A functional magnetic resonance
study. PNAS Proceedings of the National Academy of Sciences of the United States
of America, 95, 14494–14499. doi:10.1073/pnas.95.24.14494
Vohs, K. D., Baumeister, R. F., & Schmeichel, B. J. (2012). Motivation, personal
beliefs, and limited resources all contribute to self-control. Journal of Experimen-
tal Social Psychology, 48, 943–947.
Vyas, A., Bernal, S., & Chattarji, S. (2003). Effects of chronic stress on dendritic arbo-
rization in the central and extended amygdala. Brain Research, 965, 290–294.
doi:10.1016/S0006-8993(02)04162-8
Ward, A., & T. Mann. (2000). Don’t mind if I do: Disinhibited eating under cogni-
tive load. Journal of Personality and Social Psychology, 78, 753–763.
Weber, E. U. (2001). Personality and risk taking. In N. J. Smelser & P. B. Baltes (Eds.),
International encyclopedia of the social and behavioral sciences (pp. 11274–11276).
Oxford, England: Elsevier Science. doi:10.1016/B0-08-043076-7/01782-4
Zaleskiewicz, T. (2001). Beyond risk seeking and risk aversion: Personality and the dual
nature of economic risk taking. European Journal of Personality, 15, S105–S122.
doi:10.1002/per.426

176       zayas, mischel, and pandey

13490-07_PT3_Ch06-3rdPgs.indd 176 12/9/13 11:57 AM

 7
The Neuroscience of Dual
(and Triple) Systems
in Decision Making
Samantha M. W. Wood and Antoine Bechara

With the emergence of functional neuroimaging and the field of neuro­

economics, the debate about “single system” or “dual system” models for valua­
tion is beginning to intensify. Proponents of a single system for valuation maintain
that one neural region is devoted to evaluating costs and rewards, while dual
system models argue decision making is a balance between fast, impulsive pro­
cessing and slow, reflective processes. Using evidence from the clinical side of
the argument, we will argue in this chapter that not only are the traditional two
processes supported by the clinical facts, but in addition the evidence points to a
third process, thus calling for the notion of “triple process” models. We also will
highlight the importance of a third neural system that translates homeostatic,
bodily signals into feelings of craving, which in turn modifies the dynamics of
the traditional dual systems.
Dual process models have had considerable influence in psychologi-
cal research. In the decision-making and self-control literature, dual pro-
cess models have been invoked to explain why individuals make rational,

http://dx.doi.org/10.1037/14322-008
The Neuroscience of Risky Decision Making, by V. F. Reyna and V. Zayas (Editors)
Copyright © 2014 by the American Psychological Association. All rights reserved.

177

13490-08_Ch07-3rdPgs.indd 177 11/15/13 1:43 PM

 analytical decisions in some circumstances but fail to use logical reasoning
in other situations (e.g., Baumeister, Masicampo, & Vohs, 2011; Chaiken &
Trope, 1999; Evans, 2008; Morewedge & Kahneman, 2010; Reyna, 2004).
Often referred to as System 1 and System 2, the current popular conception
of the dual processes posits an intuitive system that is quick, automatic, and
emotional and a reasoning system that is effortful, controlled, and nonemo-
tional. Although research on decision making has traditionally posited two
psychological systems used in evaluations (System 1 vs. System 2, which we
refer to elsewhere as impulsive versus reflective systems; Bechara, 2005), more
recent evidence suggests that a largely overlooked structure, the insula, plays
a key role in modulating the dynamics of the systems. More specifically, the
insula translates bottom-up, interoceptive signals into what subjectively may
be experienced as an urge or craving.1 This craving system potentiates the
activity of the impulsive system and weakens cognitive resources that are
needed for the normal operation of the reflective system. Thus, when consid-
ering physiological states that involve deprivation, withdrawal of an appeti-
tive stimulus, stress, anxiety, or any condition associated with homeostatic
perturbation, a third process (governed by the insula) emerges that works to
correct the homeostatic imbalance. This third process directly affects the
functionality of the traditional dual systems.
In this chapter, we detail how dual process models have evolved and
discuss the neural underpinnings of these systems. Finally, we outline the role
of the insula in decision making and describe how the insula interacts with
the other two systems to modify decisions.

History of Dual Process Models

The present conception of the two systems is not a perfect reflection of

how dual processes were originally conceived. Initially, dual process models
were used to explain findings about how individuals process information.
Psychologists in the field of vision noticed a dissociation between automatic
detection and controlled-search mechanisms (Shiffrin & Schneider, 1977).
Their dual process theories claimed that automatic detection requires a great
deal of training and would be difficult to suppress once learned. The term
dual process has become less popular in vision science, but the theoretical
contributions from Shiffrin and Schneider still resonate in models of atten-
tion. Contemporary theories of attention still include a bottom-up automatic

1As we describe later in the chapter, the insula is generally responsible for internal, subjective feelings.
These feelings are not limited to cravings and other homeostatic signals but can include more abstract
feelings, for example, disgust, admiration, love, and indignation.

178       wood and bechara

13490-08_Ch07-3rdPgs.indd 178 11/15/13 1:43 PM

 mechanism for attending to salient stimuli and a top-down voluntary control
of attention (Knudsen, 2007). Notably, this initial formulation of dual pro-
cess models was completely agnostic to the role of emotion. Psychologists
studying vision were interested in perceptual attention cues rather than the
role of emotion in attention.
Social psychologists brought dual process theories to more affective
questions. They used dual process theories to address questions related to per-
suasion and evaluations. For example, Chaiken’s (1980) heuristic-systematic
model of information processing and Petty and Cacioppo’s (1984) elabora-
tion likelihood model both argued that persuasive messages could be pro-
cessed heuristically by simple cues or processed systematically using effortful
cognitive activity. In addition to explaining attitudes about persuasive mes-
sages, two systems were used to explain evaluations of people. Inspired by
Shiffrin and Schneider’s model of two systems in information processing,
Devine (1989) used a similar theory to explore stereotypes. Stereotypes are
well-learned and automatically activated, like automatic detection of visual
stimuli; however, unlike automatic visual detection, they can be overridden
by controlled processes with cognitive effort. In these earlier social psychol-
ogy examples of dual process models, theories did not explicitly identify a
single system as being emotional. Theories about how people and messages
are evaluated assumed that both systems must handle affective content. The
major distinction between the two systems was that one was automatic while
the other was controlled.
Although the dichotomy between emotional and cognitive systems (or
“hot” and “cool” systems, per Metcalfe & Mischel, 1999) is now common-
place in the decision making literature, this portrayal of dual processes is
relatively new. The early reviews that attempted to integrate various dual
process theories of reasoning into a single System 1 and System 2 frame-
work did not include emotional components (Stanovich & West, 2000, see
Figure 7.1).2 This framework explained why judgments sometimes maximize
standard expected utility, but at other times individuals rely on heuristics
and biases. While intuitive judgments might violate principles of classical
economics, scholars have emphasized the computational efficiency of using
heuristic-based reasoning (Gigerenzer & Goldstein, 1996). By using fuzzy and
intuitive reasoning rather than strict computational logic, individuals can
reduce information to smaller, more manageable dimensions (Reyna, 2004).
Indeed, this intuitive, gist-based reasoning is used more often in individuals

2Many dual process theories of decision making had been proposed prior to Stanovich and West (2000),
but this is the earliest major review paper typically cited in regard to dual process theories, likely because
of Stanovich and West’s thorough review as well as the responses from many influential proponents
of dual process theories. However, the hot/cool dichotomy does predate this work (e.g., Metcalfe &
Mischel, 1999; Nisbett & Ross, 1980).

the neuroscience of dual systems      179

13490-08_Ch07-3rdPgs.indd 179 11/15/13 1:43 PM

 System 1 ("Impulsive") System 2 ("Reflective")

Associative Rule-based

Holistic Analytic

Automatic Controlled

Relatively undemanding of cognitive

Demanding of cognitive capacity
capacity

Relatively fast Relatively slow

Acquisition by biology, exposure, and Acquisition by culture and formal

personal experience tuition

Highly contextualized task construal Decontextualized

Figure 7.1. Summary of System 1 and System 2 attributes (for a review of System 1
and System 2 properties, see Stanovich & West, 2000).

as cognitive development (Reyna & Ellis, 1994) or the development of

expertise (Reyna, Lloyd, & Brainerd, 2003) advances.
Scholars quickly began including emotions under System 1. Emotions
may also be evoked rapidly and automatically and may conflict with our ana-
lytic reasoning (Epstein, 1994). Metcalfe and Mischel (1999) hypothesized
that an internal tug-of-war between emotional and cognitive processes could
explain the struggle between temptation and willpower:
We propose that there are two types of processing—hot and cool—
involving distinct interacting systems. . . . The cool cognitive system is
specialized for complex spatiotemporal and episodic representation
and thought. We call it the “know” system. The hot emotional system is
specialized for quick emotional processing and responding on the basis
of unconditional or conditional trigger features. We call it the “go”
system. (p. 4)
Mischel used this framework to generate strategies for improving self-control.
In a series of experiments, preschool children were given the choice to eat
one marshmallow now or wait for a delay period to get two marshmallows.
When both the delayed and immediate rewards were placed in front of the

180       wood and bechara

13490-08_Ch07-3rdPgs.indd 180 11/15/13 1:43 PM

 children during the delay period, the children could only delay themselves for
an average of 1 minute (Kross & Mischel, 2010). Yet, children were able to
wait longer, approximately 11 minutes, if all reward stimuli were obstructed
from view or the child did not attend to them (allowing the children to rely
on their less impulsive system).
However, not all research on System 1 and System 2 in delayed dis-
counting suggests that emotional cues will induce temporal myopia. Although
presenting children with an actual reward reduces how long they can delay
reward, presenting children with pictures of the food rewards actually increased
their ability to wait (Mischel & Moore, 1973). These findings indicate that
emotional cues can facilitate either System 1 (impulsive) or System 2 (reflec-
tive and controlled) decisions.
The intuitive separation of emotion and cognition has become typical
of System 1 and System 2 descriptions. However, as we will see when exam-
ining the neurological evidence, this dissociation of emotion and cognition
does not exist in the brain. Instead, the neural systems underlying System 1
and System 2 decision making are both affective.

Dual Process Models in Decision Making

Individuals can often maintain incongruous goals simultaneously. We can

desire a piece of chocolate cake while sincerely wanting to remain successful
dieters. We may crave the thrill of skydiving but still value our physical safety.
Dividing evaluative processes into two systems helps researchers account for
this subjective feeling that we consist of multiple selves with diverging moti-
vations. Because of its theoretical usefulness, the concept of System 1 and
System 2 has contributed to a range of findings in the psychology of decision
making.
In studies of temporal discounting, participants are required to choose
between a smaller but immediate reward or a larger but delayed reward.
Mischel’s marshmallow study (Mischel, Shoda, & Rodriguez, 1989) is an
example of a temporal discounting task for children. Children have difficulty
executing the self-regulatory strategies required to wait for a delayed reward.
Yet, adults also choose impatiently. Paradoxically, individuals are more likely
to choose $10 today over $11 tomorrow than to choose $10 in a year over
$11 in a year and day (Frederick, Loewenstein, & O’Donoghue, 2002). In
both choice scenarios the time delay (1 day) and the monetary increase ($1)
are identical, but the delay is preferred when both payoff options are in the
future. Similar time discounting patterns hold true even when the future
payoffs are ensured; for example, when subjects are given post-dated checks.
Dual processing may be the underlying cause of this preference reversal.

the neuroscience of dual systems      181

13490-08_Ch07-3rdPgs.indd 181 11/15/13 1:43 PM

 McClure, Laibson, Loewenstein, and Cohen (2004) argued that the lim-
bic system responds to immediate rewards while another system, governed
by the prefrontal cortex, responds to any abstract rewards. Individuals with
more active limbic systems would have more difficulty resisting an immediate
reward. Accordingly, in Frederick et al.’s (2002) temporal discounting para-
digm, only the $10 today would activate the limbic system, while the other
rewards ($11 tomorrow, $10 in a year, and $11 in a year and a day) would rely
on the prefrontal cortex.
Reliance on the automatic and intuitive System 1 may also account for
preference reversal in risk taking. In the classic Asian disease problem, par-
ticipants show greater preference for a policy that will save 200 people (out
of 600) with 1/3 probability than the identical policy in which 400 people
(out of 600) will die with 1/3 probability (Tversky & Kahneman, 1981). Even
outside of risk assessment, the way in which choices are framed can have a
significant effect on our decisions. These framing effects result from using
System 1 instead of the more analytical (and less context-biased) System 2.
Dual process models have even influenced theories of moral reason-
ing. Moral philosophers use logical, syllogistic reasoning to arrive at moral
principles. However, philosophical laymen show reliance on emotional intu-
itions to resolve moral dilemmas, especially under conditions of cognitive
load (Greene, Morelli, Lowenberg, Nystrom, & Cohen, 2008; Haidt, 2001).
Greene et al. and Haidt have argued for a dual process account of moral rea-
soning with one system using intuitive moral reasoning fueled by emotional
reactions to moral dilemmas and a second system using cognitive resources
to calculate utilitarian outcomes of moral dilemmas.

Neuroscience and Dual Processes

Clinical neuroscience provides evidence for at least two separate but

interconnected brain systems involved in the evaluation of choices. The first
brain region is the amygdala–striatal system. The amygdala has strong pro-
jections to the striatum (a structure that consists of the nucleus accumbens,
caudate, putamen, fundus, and olfactory tubule) (Everitt et al., 1999). While
the amygdala and striatum are separate structures with some dissociable
functions, some scholars refer to a neural macrostructure called the extended
amygdala, containing the nucleus accumbens shell and central nucleus of
the amygdala due to similarities in connectivity and morphology (Alheid &
Heimer, 1988; Koob & Le Moal, 2001). Both the striatum and the amygdala
are critical pieces of the brain’s reward circuitry, responding to changes in the
neurotransmitter dopamine (Kelley, Schiltz, & Landry, 2005). The amygdala–
striatum corresponds to the traditional System 1 because it responds quickly

182       wood and bechara

13490-08_Ch07-3rdPgs.indd 182 11/15/13 1:43 PM

 and automatically on the basis of innate and learned associations. These
associations are relatively inflexible in that, once a stimulus value is learned,
the motivational value is difficult to change (Everitt & Robbins, 2005).
The amygdala responds to both conditioned (learned) and unconditioned
(innately motivational) stimuli (LeDoux, 2000). Thus, the amygdala–striatal
system is responsible for the expression of motivated responses (e.g., fear,
Corcoran & Quirk, 2007; and reward, Everitt & Robbins, 2005), as well as
transferring controlled behaviors and preferences into habits.
Clinical studies of patients with lesions of the amygdala–striatal sys-
tem provide additional insights about the System 1 of dual process models.
Notably, the amygdala governs the bottom-up generation of emotions in
response to “primary inducers” (Bechara & Damasio, 2005; McRae, Misra,
Prasad, Pereira, & Gross, 2012; Ochsner et al., 2009). Primary inducers are
stimuli that are physically present in the current environment. For example,
if you were out for a walk and jumped in fear upon encountering a snake in
the path, then the snake would be a primary inducer. Conditioned stimuli
that have been associated with a provoking unconditioned stimulus are also
primary inducers. So, if the snake in our example were always under the same
bench, that bench could also become a primary inducer eliciting fear.
Individuals with damage to the amygdala have a striking disruption
in their primary induction of emotion. Consider one of the best-researched
examples of amygdala damage, patient SM. SM has normal IQ, memory, lan-
guage, and perception. Yet, SM exhibited no fear during exposure to live
snakes and spiders or during a tour of a haunted house (Feinstein, Adolphs,
Damasio, & Tranel, 2011). In the haunted house,
[t]he hidden monsters attempted to scare SM numerous times, but to no
avail. She reacted to the monsters by smiling, laughing, or trying to talk
to them. In contrast, their scare tactics typically elicited loud screams of
fright from the other members of the group. (Feinstein et al., p. 35)

These effects are not specific to SM. War veterans with focal amygdala
lesions generally show fewer signs of posttraumatic stress disorder (Koenigs
et al., 2008). Conversely, amygdala stimulation increases corticosterone levels
and symptoms of fear, anxiety, and attention (Applegate, Kapp, Underwood,
& McNall, 1983; Kapp, Supple, & Whalen, 1994; see Phillips, Drevets,
Rauch, & Lane, 2003, for an excellent review).
Amygdala lesions are typically associated with impairment in evaluat-
ing negative stimuli (e.g., Berntson, Bechara, Damasio, Tranel, & Cacioppo,
2007). However, functional neuroimaging studies and single-cell recordings
indicate that the amygdala is also involved in evaluating rewards (for a review,
see Murray, 2007). The amygdala activates more in response to positive, neg-
ative, and unusual or interesting stimuli than in response to neutral stimuli

the neuroscience of dual systems      183

13490-08_Ch07-3rdPgs.indd 183 11/15/13 1:43 PM

 (Cunningham & Brosch, 2012). Its affective tuning is flexible, and participants
show greater amygdala activation in response to positive stimuli when they
are asked to evaluate only the positivity of the stimuli and greater amygdala
activation in response to negative stimuli when they are asked to evaluate only
the negativity of the stimuli (Cunningham, Van Bavel, & Johnsen, 2008).
Thus, Cunningham and colleagues argued that the amygdala responds to goal
salience rather than a stimulus’s valence. Yet, single-cell recordings in rhesus
macaques indicate that positive and negative value are encoded in distinct but
spatially integrated neurons (Paton, Belova, Morrison, & Salzman, 2006). This
evidence points to the amygdala’s role in forming affective associations across
both positive and negative valences. Finally, this system has also been shown
to be critical for evaluating certain types of risky decisions using monetary gains
and losses (e.g., Bechara, Damasio, Damasio, & Lee, 1999; Weller, Levin, Shiv,
& Bechara, 2007).3
Similarly, the striatum has been implicated in reward prediction (Schultz,
2000) in tracking how actual reward receipt compares with expected reward.
Thus, the striatum recognizes patterns of rewards and calculates their probabil-
ity to bias behavior (Bechara & Damasio, 2005). For example, a neuro­logical
patient who had damage to the entire medial temporal lobe4 but retained an
intact striatum was able to learn the nonconscious affective value of faces
despite being unable to learn overt, factual knowledge about the faces (Tranel
& Damasio, 1993).
Unlike System 1, the second dual process system has been associated
with prioritizing rewards that are less concrete: outcomes in the future, out-
comes that affect other individuals, and outcomes that are hypothetical.
Here, we refer to these outcomes as secondary inducers. Secondary inducers
are hypothetical or remembered emotional triggers (e.g., thoughts, memories,
reflections). In the example of the snake, the next time you go for a walk,
you might remember the snake in your path and decide to choose a different
route. In this case, the memory of the snake is a secondary inducer.
The brain region that triggers an emotional state in response to second-
ary inducers is the ventromedial prefrontal cortex (vmPFC).5 The vmPFC
provides a link between the amygdala and the area of the brain involved in
working memory (the dorsolateral prefrontal cortex). Since secondary inducers
are not present in the current environment, they must be held online in work-
ing memory. To create an emotional state, the information about the memory
3Weller et al. (2007) found that patients with amygdala lesions showed a deficit in risky decision making
for choices involving gains but not losses. This is particularly interesting given that the majority of studies
implicate the amygdala in evaluating negative rather than positive stimuli.
4The amygdala and hippocampus are both housed in the medial temporal lobe. The patient’s lesion also

encompassed a portion of the orbital prefrontal cortex and the anterior cingulate.
5Here, we use vmPFC to refer to the ventral medial sector of the brain as well as the medial sector of the

orbitofrontal cortex. Thus, this area encompasses Brodman’s areas 25; lower 24; 32; and medial aspects
of 11, 12, and 10.

184       wood and bechara

13490-08_Ch07-3rdPgs.indd 184 11/15/13 1:43 PM

 must be relayed to the amygdala and striatum, structures that induce somatic
responses.
Accordingly, individuals with damage to the vmPFC have selective
deficits for avoiding negative future consequences in personal and social
matters. The classic neurological case study of vmPFC damage comes from
a 19th century railroad worker named Phineas Gage (Damasio, Grabowski,
Frank, Galaburda, & Damasio, 1994). While he was working, a meter-long
metal tamping rod shot through Gage’s skull. Miraculously, Gage survived the
accident and was still able to hear, see, speak, and use all of his motor skills.
Despite this apparent full recovery, his friends felt that Gage was no longer the
same person. After his brain damage, Gage had trouble holding a regular job,
anticipating the future, acting in socially appropriate ways, and having a sense
of responsibility. Modern patients with vmPFC damage show similar impair-
ments. They repeatedly ignore negative future consequences when making
decisions and fail to learn from these mistakes. Yet, these patients have IQ
scores in the normal range and show normal performance on the Wisconsin
Card Sorting Test, tasks that require self-ordering, cognitive estimations,
and judgments of recency and frequency (Bechara, Damasio, Damasio, &
Anderson, 1994).
One task that has proven particularly effective in disentangling primary
and secondary inducers is the Iowa gambling task (IGT). In this task, partici-
pants are asked to make choices from among four decks of cards. Each card
choice is associated with winning some amount of money (a certain reward);
however, some cards also take money away (a probabilistic loss). Unknown to
the participant, two of the decks have cards that bring high rewards, but the
losses are also high. Because the losses are so large, these decks actually have
a negative net value. If a participant chooses only from those “disadvanta-
geous” decks, she will lose more money than she wins. On the other hand, in
the other two “advantageous” decks, each card brings only a small reward, but
when there is a loss, it is also very small. Because the probabilistic losses are
much smaller, the net value of these advantageous decks is positive.
When playing the IGT, both patients with amygdala damage and
patients with vmPFC damage tend to perform poorly, but for different rea-
sons. Behavioral performance for the two patient groups is nearly identical,
but psychophysiological methods tell a more complicated story. Skin con-
ductance responses (SCRs) give a measure of emotional arousal by recording
the sweat secretion by eccrine sweat glands, controlled by the autonomic
nervous system. A large SCR suggests that the associated stimulus has strong
motivational value (Hugdahl, 1995). During the IGT, normal, neurologically
healthy participants show large SCRs before choosing from a disadvanta-
geous deck (an anticipatory response to a secondary inducer) as well as after
actually receiving a reward or punishment (a primary inducer). However,

the neuroscience of dual systems      185

13490-08_Ch07-3rdPgs.indd 185 11/15/13 1:43 PM

 patients with vmPFC lesions did not show large SCRs before choosing from a
disadvantageous deck (Bechara & Damasio, 2005; Bechara et al., 1999). The
secondary inducer did not produce emotional arousal when the vmPFC was
damaged. These patients still responded to primary inducers by showing large
SCRs in response to winning or losing money in the game. These patients
had a System 1 that was intact and able to assess the motivational value of
concrete rewards and losses in the here and now.
The SCRs of the patients with amygdala damage are even more reveal-
ing. One might expect a double dissociation: that amygdala patients would
have high SCRs in anticipation of choosing from a disadvantageous deck
(a secondary inducer), but not in response to winning or losing (a primary
inducer). Instead, patients with amygdala damage did not show large SCRs in
either situation (Bechara & Damasio, 2005; Bechara et al., 1999). This find-
ing suggests that emotional responses to a secondary inducer are generated
by using information from primary responses prospectively. A large monetary
loss triggers an affective reaction associated with a neural pattern of activa-
tion during the game. When patients with amygdala damage imagine the
outcome of choosing from that deck again, the vmPFC triggers a reactiva-
tion of that original neural pattern of activation. Thus, the vmPFC does not
calculate the emotional saliency of a stimulus by itself. Instead, the vmPFC
relies on the amygdala–striatal system to enact somatic changes.
In this sense, neuroscientific evidence contradicts our intuitions about
self-control. Colloquialisms on self-control tend to imply that we use two
completely separable systems to weigh short-term and long-term goals. We
feel torn between two options; we wrestle with ourselves and listen to either
the devil or the angel on our shoulder. In actuality, though, self-control is not
a shouting match between two isolated brain regions. The amygdala–striatal
region and the prefrontal cortex are interconnected systems. Patients with
amygdala lesions are unable to produce emotional responses to primary and
secondary stimuli, implying that the amygdala is necessary to trigger response
in the vmPFC (Weller et al., 2007).

Recent Opposition

Despite the popularity of using two systems in models of decision mak-

ing, dual process theories are not without opposition. Recently, the inter­
disciplinary field of neuroeconomics has applied the methods of neuroscience
to the study of economics. With the advent of neuroeconomics, researchers
began to search for a single source of valuation in the brain. In economic
marketplaces, currency is useful for placing comparable values on disparate
goods and services. Single process theories argue for a single neural currency

186       wood and bechara

13490-08_Ch07-3rdPgs.indd 186 11/15/13 1:43 PM

 for evaluating options. These attempts to uncover a single system for deci-
sion making mostly pinpointed the vmPFC as the neural correlate of value.
By examining changes in vmPFC BOLD signal, researchers could predict
individuals’ preferences for a wide variety of stimuli including soft drinks
(McClure, Li, et al., 2004) snacks (Plassmann, O’Doherty, & Rangel, 2007),
car brands (Schaefer, Berens, Heinze, & Rotte, 2006), and wine (Plassmann,
O’Doherty, Shiv, & Rangel, 2008).
Although these functional neuroimaging studies provide a strong argu-
ment for the vmPFC and adjacent regions as being a center of valuation,
neurological evidence demonstrates that a large bilateral lesion that wipes
out this area of the brain does not entirely abolish valuation. According to
clinical observations, these patients do not generally make decisions that lead
to concrete and immediate harm to themselves and others. Their impair-
ments are specific to decisions about consequences that are delayed, probabi-
listic, and abstract (Bechara & Damasio, 2005). To illustrate, although these
patients do not properly value future or probabilistic outcomes in the IGT,
they still prefer high immediate gain or avoid high immediate loss, which
suggests that their valuations of these immediate contingencies are not dis-
rupted. Indeed, by the end of the IGT, many patients with vmPFC damage
are actually able to report which decks were advantageous or disadvantageous
when asked directly, but they persisted in choosing disadvantageous decks
(Bechara et al., 1997).

Addiction: A Clinical Example

We propose that addiction is the result of an imbalance of the two sys-

tems described above: the impulsive amygdala–striatal system and the reflec-
tive prefrontal cortex system (Bechara, 2005). Due to dysregulation of one
or both of these systems, addicts behave similarly to individuals with vmPFC
damage. They tend to ignore future long-term consequences and fail to learn
from their past mistakes. This description of addiction is consistent with stud-
ies of structural brain abnormalities in addicts. Cocaine addicts have signifi-
cantly reduced amygdala (Makris et al., 2004), striatal (Barrós-Loscertales
et al., 2011), and vmPFC (Franklin et al., 2002) volume compared with
healthy controls.
Furthermore, a wide range of addicts show impaired decision making
as measured by the IGT. The IGT mimics real-life decisions by creating a
conflict between an immediate, luring reward and a probabilistic punish-
ment. The IGT paradigm has pointed to decision-making deficits in alcohol
(Fein, Klein, & Finn, 2004), ecstasy (Hanson, Luciana, & Sullwold, 2008),
cocaine (Verdejo-Garcia et al., 2007), and cannabis (Whitlow et al., 2004)

the neuroscience of dual systems      187

13490-08_Ch07-3rdPgs.indd 187 11/15/13 1:43 PM

 abusers. This decision-making impairment is not selective to drug abuse and
can also be found in behavioral addictions including pathological gambling
(Cavedini, Riboldi, Keller, D’Annucci, & Bellodi, 2002) and overeating/
obesity (Davis, Levitan, Muglia, Bewell, & Kennedy, 2004).
Addicts’ performance on the IGT and measures of their skin conduc-
tance while playing can give us more information about the neural disruptions
underlying their addictions. When Bechara, Dolan, and Hindes (2002) asked
substance abusers to play the IGT, the participants fell into three categories.
The first group was a minority of participants who performed similarly to nor-
mal, healthy individuals. Based on their lack of decision making impairment
as measured by the IGT, this group may consist of mainly “functional” addicts.
Such addicts have generally experienced minimal consequences of their drug
use in their everyday life and may not be poor decision makers per se.
A second portion of the addicts played similarly to individuals with
vmPFC lesions. These players showed impaired behavioral performance as
well as reduced SCRs in anticipation of choosing from the disadvantageous
decks. Although they still showed large SCRs in response to losing large
sums of money, they did not turn these emotional responses into prospective
signals of the riskiness of each deck of cards.
The final group of addicts showed a novel signature of SCRs while play-
ing the IGT. Like the second group, these individuals showed impaired behav-
ioral performance. However, these individuals had SCRs similar to those of
normal, healthy participants, with the exception that they had heightened
emotional responses (SCRs) in response to winning large sums of money.
Unlike individuals with lesions to the vmPFC who display myopia for the
future, these individuals seem to fail at the IGT due to a hypersensitivity to
reward (for a discussion on hypersensitivity to reward in adolescence, see
Chapter 3, this volume).
An open question in studying the neurobiology of addiction is the
extent to which these brain abnormalities are due to genetic predisposition
and neurotoxicity. Some studies suggest that individuals who have not used
drugs but who are at high risk for drug abuse (e.g., a family history of alco-
hol abuse) are more likely than other healthy individuals to have decision-
making impairments (Fein & Chang, 2008; Lovallo et al., 2006). Moreover,
both drug-dependent individuals and their nonaddicted siblings have front­
ostriatal brain abnormalities, suggesting that these abnormalities are not caused
entirely by drug use (Ersche et al., 2012). Nonetheless, because drug use cannot
be experimentally manipulated in humans, animal models are typically used to
separate the influences of genetics and environment. Animal studies indi-
cate that brain abnormalities in addicts are part of a vicious circle. Rats that
are bred without the dopamine D1 receptor show increased vulnerability
to develop drug self-administration (Caine et al., 2007). Studies of human

188       wood and bechara

13490-08_Ch07-3rdPgs.indd 188 11/15/13 1:43 PM

 addicts indicate that some brain volume losses are correlated with years of
use, implying that drug use also causes neurological changes (Goldstein &
Volkow, 2002). Dysfunction in reward processing makes individuals more
likely to use and become addicted to drugs; drug use then further disrupts
reward processing.
Although we have used addiction as a case study, disruption in the
prefrontal cortex and amygdala–striatal system is not unique to addiction.
Dysfunction in these brain regions has been linked to impulsivity more gen-
erally. In addition to addiction, other clinical disorders have been linked
to impulsive behavior, including attention-deficit/hyperactivity disorder
(ADHD), mania, and antisocial behavior (Dalley, Everitt, & Robbins, 2011).
For instance, ADHD is associated with hypoactivity in the frontal cortex
and basal ganglia, among other regions (Dickstein, Bannon, Castellanos, &
Milham, 2006). Patients with vmPFC damage show impairments in moral
judgment (Young et al., 2010), pointing to a role for the vmPFC in antisocial
behavior.

The Insular Cortex—A Missing Piece?

Dual process theories have remained popular models due to their explan-
atory power. However, we argue that the neuroscientific evidence points to
an additional system involved in decision making. The neural basis for this
third system is the insular cortex, or insula. The cognitive function of the
insula is translating bodily states into conscious feelings. Although the amyg-
dala is responsible for associative learning and emotions, the insula governs
the conscious feeling of those emotions. As explained by Damasio (1994),
emotions can be disentangled from the feeling of emotions. Emotions are
the physical reactions to stimuli, but we may have that emotional reaction
without translating the somatic reactions into a feeling. This sentiment is not
unique to the modern somatic marker hypothesis; it was echoed by James and
Lange decades earlier:
Common sense says, we lose our fortune, are sorry and weep; we meet
a bear, are frightened and run; we are insulted by a rival, are angry
and strike. The hypothesis here to be defended says that this order of
sequence is incorrect . . . and that the more rational statement is that
we feel sorry because we cry, angry because we strike, afraid because we
tremble . . . Without the bodily states following on the perception, the
latter would be purely cognitive in form, pale, colorless, destitute of emo-
tional warmth. We might then see the bear, and judge it best to run,
receive the insult and deem it right to strike, but we should not actually
feel afraid or angry. (James, 1890, p. 449–450)

the neuroscience of dual systems      189

13490-08_Ch07-3rdPgs.indd 189 11/15/13 1:43 PM

 The insula system adds a crucial missing element to the dual pro-
cess model: a subjective feeling of craving. As we explain below, craving is
the result of the brain attempting to correct an imbalance in homeostasis.
Recognizing the role of the insula in decision making highlights the impor-
tance of the situational context, the way we feel in the moment of the deci-
sion. Consider the adage that one should never go grocery shopping while
hungry. Hunger is associated with a dearth of insulin, which signals a homeo-
static imbalance to the brain. As a result, the brain treats food as a dispropor-
tionately more valuable reward, prompting the hungry shopper to buy foods
that will seem far less appetizing later.

The Insula: Anatomy and Function

We recognize five senses that are responsible for perceiving the exter-
nal world. Sight, hearing, touch, smell, and taste allow us to engage in
exteroception—awareness of stimuli outside the body. The insula is responsible
for a “sixth sense,” that of interoception—perception of what is happening
inside the body.
Typically, interoception is defined as awareness of the state of one’s
internal organs. For example, the insula responds to changes in cardiovascular
function (Critchley, Wiens, Rotshtein, Öhman, & Dolan, 2004), stimula-
tion of the gastrointestinal tract (Aziz et al., 2000), and distention (Wang
et al., 2008). Yet, the insula also translates internal bodily signals that are not
specific to organs. The Valsalva maneuver (Henderson et al., 2003), sensual
touch (Arnow et al., 2002), and exercise (Williamson, McColl, & Mathews,
2003) are all associated with insula activation. Clinically, there is ongoing
activity in the insula in chronic pain patients (Kupers, Gybels, & Gjedde,
2000). Also, dysfunction in the insula can lead patients to feel bodily signals
when none exist; patients with neuropathic pain show evoked activity in the
insula (Peyron, Laurent, & Garcia-Larrea, 2000).
The insula is part of an evolutionarily old interoceptive system in the
brain responsible for maintaining the body’s homeostasis. Sympathetic and
parasympathetic afferents are relayed from the brainstem through the thal-
amus to the posterior insula (Craig, 2002). These inputs are necessary for
somatosensory, vestibular, and motor functions. However, these physiological
states also have motivational value. Homeostatic signals are only valuable
insofar as they prompt regulatory behavior to correct imbalances. According
to Craig’s (2002) research on insula function, the physiological states acti-
vated in the posterior insula are re-represented in the anterior insula to pro-
duce subjective feelings. To provide an example of this re-representation,
Craig, Chen, Bandy, and Reiman (2000) examined activity in the insula
when applying innocuous cool stimuli to subjects’ right hands. Activity in

190       wood and bechara

13490-08_Ch07-3rdPgs.indd 190 11/15/13 1:43 PM

 the posterior insula was linearly related to the actual temperature of the stim-
uli. Intriguingly, activity in the anterior insula was correlated with subjects’
subjective ratings of the intensity of the stimulus.
The anterior insula is a phylogenetically newer brain region that acts as
a neural highway connecting interoceptive information (from the posterior
insula) with information about existing motivations, reward cues, environ-
mental stimuli, and social conditions (for a review on the evolution of insular
pathways, see Craig, 2002). The anterior insula has reciprocal connections
to regions including the amygdala, ventral striatum, vmPFC, and anterior
cingulate cortex. Rather than merely representing the state of viscera, the
translation of signals in the anterior insula represents all subjective feelings.
Past studies have elicited insula activity using a variety of emotional (rather
than purely somatic) feelings including “the feeling of knowing” (a subjective
sense of retrievability), love, anger, fear, disgust, happiness, beauty, unfairness
and indignation, social exclusion, empathy, trust, and “a state of union with
God” (for an excellent review, see Craig, 2009). Patients with right insular
lesions report altered subjective feelings: anergia, underactivity, and tired-
ness, perhaps due to the insula’s connections to brain regions that are neces-
sary for willed actions (Manes, Paradiso, & Robinson, 1999).

Craving and the Insula

Subjective feelings are a critical component of normal decision making.

Cases of self-control provide a useful illustration of the role of the insula and
subjective feelings in decision making. In these instances, the insula mediates
a feeling of craving that can affect the reward value of stimuli.
The insula has been implicated in processing the value of a variety of
rewards. Injecting a rat insula with low doses of SB-334867 (a receptor agonist
that blocks activity in the insula) leads to a reduction in intake of nicotine
(Hollander, Lu, Cameron, Kamenecka, & Kenny, 2008) and higher doses reduce
intake of palatable, high-fat food as well as total baseline food (Nair, Golden,
& Shaham, 2008; Rodgers et al., 2001). Similarly, temporarily inactivating
the rat posterior insula disrupts drug seeking in amphetamine-experienced rats
(Contreras, Ceric, & Torrealba, 2007). Rats lack the phylogenetically newer
insular structures underlying the evaluation of interoceptive signals through
re-representation in the anterior insula (Craig, 2002). Therefore, these find-
ings suggest that reward-seeking behavior requires physical sensations of crav-
ing, but not necessarily the subjective feeling of craving. In humans, insula
activity during a decision-making game can predict relapse in recovering
methamphetamine addicts (Paulus, 2007). In addition, chronic cocaine users
have a smaller ratio of gray to white matter in the insula (Franklin et al.,
2002). The density decrease was not correlated with years of cocaine use,

the neuroscience of dual systems      191

13490-08_Ch07-3rdPgs.indd 191 11/15/13 1:43 PM

 so the brain abnormality may have been preexisting rather than gradually
induced with increased drug exposure.
Patients with insula lesions provide striking evidence that the changes
in reward seeking are due to a disruption in craving. Naqvi, Rudrauf, Damasio,
and Bechara (2007) examined nicotine addiction in patients who had acquired
brain lesions from strokes or surgical resectioning. All of the patients had been
cigarette smokers prior to their brain damage. However, patients with insula
damage (compared with all other lesions) were more likely to undergo a “dis-
ruption of smoking addiction. A patient was considered to have a “disrupted”
addiction if she quit smoking after the lesion and did not relapse, rated the
difficulty of quitting smoking as relatively easy (less than 3 on a scale of 1 to
7), and reported no urges to smoke since quitting. Smokers with damage to
the insula were nearly 8 times more likely to undergo a disruption of addiction
than were smokers with damage to other brain regions. One insula patient
reported that he was smoking more than 40 cigarettes per day prior to his
stroke. After his stroke, he abruptly stopped smoking, explaining that, “I for-
got that I was a smoker,” elaborating that “my body forgot the urge to smoke.”
Cravings and urges are represented in the insula, so when insula function was
interrupted, the nicotine addiction was as well.

Role of the Insula: A Balance Of Power

The standard account offered by dual process models suggests that there
are two separate systems that compete to determine what decision will be
made—favoring a concrete reward in the present or an abstract reward that
is in the future, probabilistic, or benefiting someone else. However, the insula
affects decision making by representing the subjective feeling of cravings and
maintaining homeostasis—a role that is not accounted for in standard dual
process theories. We propose that dual process models should be modified to
include a third system: cravings. These cravings are elicited by the brain and
body’s joint efforts to correct a homeostatic imbalance.
Moreover, the neuroscientific evidence does not provide support for
separate competing systems. Instead, the involved brain regions are highly
interconnected and affect decision making by modifying each other’s activ-
ity. These forces are sometimes depicted as conflicting parties; however a
better analogy may be that of a “neural balance of power.” Balance of power
has been a popular model to regulate governmental decision making. The
government is divided into three branches (typically legislature, execu-
tive, and judiciary); each of the branches can check the power of the other
branches. For instance, the legislature can pass laws, but if those laws are

192       wood and bechara

13490-08_Ch07-3rdPgs.indd 192 11/15/13 1:43 PM

 vmPFC

Execu on of
(a)
behavior
(c)

Amygdala

(b)

Insula

Figure 7.2. A diagram of the triple process model. The amygdala–striatal system,
which we have referred to as the impulsive system or System 1, triggers emotional
responses to the present environment. The vmPFC triggers responses to secondary
inducers (hypothetical or remembered states). The relationship between the amyg-
dala and vmPFC is reciprocal (Amaral, Price, Pitkänen, & Carmichael, 1992; Milad &
Quirk, 2002). (a) Amygdala activation is necessary to trigger vmPFC response, but
the reflective system can affect the impulsive system through several mechanisms
of impulse control. During homeostatic imbalance, the insula can alter the dynamics
between the amygdala and vmPFC by (b) sensitizing the impulsive system and
(c) disabling the reflective system.

unjust they may be vetoed by the executive or overturned by the judiciary.

The brain regions involved in decision making can also check each other
(see Figure 7.2). The amygdala rapidly and automatically evaluates stimuli
based on associative learning, but those evaluations can be modified by the
vmPFC (based on the evaluation of the stimuli’s abstract properties) and the
insula (based on how the stimuli could fulfill homeostatic needs). Certainly
not every decision needs to invoke the vmPFC and insula, just as not every
law needs to be evaluated by the executive and judiciary. Everyday life would
become too burdensome if every habit from how we brush our teeth or how
we tie our shoes needed deliberation.
The prefrontal cortex modifies decision making through mechanisms
of impulse control that modulate amygdala activity. For example, the ventro­
lateral prefrontal cortex modulates activity in the amygdala and nucleus
accumbens during emotional reappraisal tasks (Wager, Davidson, Hughes,
Lindquist, & Ochsner, 2008). The insula operates in a similar way. The

the neuroscience of dual systems      193

13490-08_Ch07-3rdPgs.indd 193 11/15/13 1:43 PM

 insula modulates activity of the vmPFC and amygdala to maintain internal
biological (and perhaps emotional) homeostasis. A need to restore homeo-
stasis (e.g., due to hunger, drug withdrawal) activates the insula and is expe-
rienced as a feeling of craving. The current needs of the body outweigh
long-term or abstract goals; correspondingly, the insula may inhibit activity
in the reflective system (vmPFC) and increase activity in the impulsive
system (amygdala–striatal). This reflects the adaptive notion that future
rewards are inconsequential if the body cannot survive current situations.

Conclusion

The standard dual process accounts of decision making depict two

competing systems: a cold and calculating system that is slow and effort-
ful against a hot and emotional system that is quick and automatic. Based
on this account, our decisions are a result of whether rational or emotional
cognition prevails. Past dual process models did not always presuppose that
automatic processes are emotional while controlled processes are devoid
of emotions. In fact, neurological evidence indicates that both decision-
making systems depend on brain regions related to emotional processing: the
amygdala–striatal system and the vmPFC. Both brain regions are critical for
forming somatic markers. The amygdala responds to primary inducers of emo-
tion while the vmPFC responds to secondary inducers of emotion. However,
the vmPFC evokes emotional reactions to secondary inducers by modulating
activity in the amygdala. Rather than being competitive, these brain regions
are highly interconnected and modulate each other’s activity. One or both
of these systems may be affected in cases of addiction (a clinical example of
dysfunctional reward processing).
In addition to the two systems that are typically discussed, recent
evidence suggests that a third system needs to be considered. This third
decision-making system reflects important contextual information about
someone’s current state. It incorporates information about physical and
emotional homeostasis, from feelings of hunger to feelings of compassion.
These feelings may alter the motivational value of rewards, for example,
ascribing higher reward value to food or drugs when we are hungry or going
through withdrawal. The insula has been implicated in interoception, or
the representation of internal, bodily signals. As a third neural system in
decision making, the insula modulates activity in the vmPFC and amygdala,
affecting the perceived value of rewards and costs. To give a full account of
decision making, we need to go beyond dual process models to include this
third system.

194       wood and bechara

13490-08_Ch07-3rdPgs.indd 194 11/15/13 1:43 PM

 References

Alheid, G. F., & Heimer, L. (1988). New perspectives in basal forebrain organization
of special relevance for neuropsychiatric disorders: The striatopallidal, amygda-
loid, and corticopetal components of substantia innominata. Neuroscience, 27,
1–39. doi:10.1016/0306-4522(88)90217-5
Amaral, D. G., Price, J. L. Pitkänen A., & Carmichael S. T. (1992). Anatomical
organization of the primate amygdaloid complex. In J. P. Aggleton (Ed.), The
amygdala: Neuropsychological aspects of emotion, memory, and mental dysfunction
(pp. 1–66). New York, NY: Wiley-Liss.
Applegate, C. D., Kapp, B. S., Underwood, M. D., & McNall, C. L. (1983).
Autonomic and somatomotor effects of amygdala central nucleus stimula-
tion in awake rabbits. Physiology & Behavior, 31, 353–360. doi:10.1016/0031-
9384(83)90201-9
Arnow, B. A., Desmond, J. E., Banner, L. L., Glover, G. H., Solomon, A., Polan,
M. L., . . . Atlas, S. W. (2002). Brain activation and sexual arousal in healthy,
heterosexual males. Brain: A Journal of Neurology, 125, 1014–1023. doi:10.1093/
brain/awf108
Aziz, Q., Thompson, D. G., Ng, V. W. K., Hamdy, S., Sarkar, S., Brammer, M. J., . . .
Williams, S. C. R. (2000). Cortical processing of human somatic and visceral
sensation. The Journal of Neuroscience, 20, 2657–2663.
Barrós-Loscertales, A., Garavan, H., Bustamante, J. C., Ventura-Campos, N., Llopis,
J. J., Belloch, V., & Ávila, C. (2011). Reduced striatal volume in cocaine-dependent
patients. NeuroImage, 56, 1021–1026. doi:10.1016/j.neuroimage.2011.02.035
Baumeister, R. F., Masicampo, E. J., & Vohs, K. D. (2011). Do conscious thoughts
cause behavior? Annual Review of Psychology, 62, 331–361. doi:10.1146/annurev.
psych.093008.131126
Bechara, A. (2005). Decision making, impulse control and loss of willpower to
resist drugs: A neurocognitive perspective. Nature Neuroscience, 8, 1458–1463.
doi:10.1038/nn1584
Bechara, A., & Damasio, A. R. (2005). The somatic marker hypothesis: A neural theory
of economic decision. Games and Economic Behavior, 52, 336–372. doi:10.1016/
j.geb.2004.06.010
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994). Insensitivity
to future consequences following damage to human prefrontal cortex. Cognition,
50(1–3), 7–15. doi:10.1016/0010-0277(94)90018-3
Bechara, A., Damasio, H., Damasio, A. R., & Lee, G. P. (1999). Different contribu-
tions of the human amygdala and ventromedial prefrontal cortex to decision-
making. The Journal of Neuroscience, 19, 5473–5481.
Bechara, A., Dolan, S., & Hindes, A. (2002). Decision-making and addiction (part II):
Myopia for the future or hypersensitivity to reward? Neuropsychologia, 40,
1690–1705. doi:10.1016/S0028-3932(02)00016-7

the neuroscience of dual systems      195

13490-08_Ch07-3rdPgs.indd 195 11/15/13 1:43 PM

 Berntson, G. G., Bechara, A., Damasio, H., Tranel, D., & Cacioppo, J. T. (2007).
Amygdala contribution to selective dimensions of emotion. Social Cognitive and
Affective Neuroscience, 2, 123–129. doi:10.1093/scan/nsm008
Caine, S. B., Thomsen, M., Gabriel, K. I., Berkowitz, J. S., Gold, L. H., Koob,
G. F., . . . Xu, M. (2007). Lack of self-administration of cocaine in dopamine
D1 receptor knock-out mice. The Journal of Neuroscience, 27, 13140–13150.
doi:10.1523/JNEUROSCI.2284-07.2007
Cavedini, P., Riboldi, G., Keller, R., D’Annucci, A., & Bellodi, L. (2002). Frontal
lobe dysfunction in pathological gambling patients. Biological Psychiatry, 51,
334–341. doi:10.1016/S0006-3223(01)01227-6
Chaiken, S. (1980). Heuristic versus systematic information-processing and the use
of source versus message cues in persuasion. Journal of Personality and Social Psy­
chology, 39, 752–766. doi:10.1037/0022-3514.39.5.752
Chaiken, S., & Trope, Y. (Eds.). (1999). Dual process theories in social psychology. New
York, NY: Guilford Press.
Contreras, M., Ceric, F., & Torrealba, F. (2007). Inactivation of the interoceptive
insula disrupts drug craving and malaise induced by lithium. Science, 318,
655–658. doi:10.1126/science.1145590
Corcoran, K. A., & Quirk, G. J. (2007). Activity in prelimbic cortex is necessary for
the expression of learned, but not innate fears. The Journal of Neuroscience, 27,
840–844. doi:10.1523/JNEUROSCI.5327-06.2007
Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological
condition of the body. Nature Reviews Neuroscience, 3, 655–666.
Craig, A. D. (2009). How do you feel—now? The anterior insula and human aware-
ness. Nature Reviews Neuroscience, 10(1), 59–70. doi:10.1038/nrn2555
Craig, A. D., Chen, K., Bandy, D., & Reiman, E. M. (2000). Thermosensory activa-
tion of insular cortex. Nature Neuroscience, 3, 184–190. doi:10.1038/72131
Critchley, H. D., Wiens, S., Rotshtein, P., Öhman, A., & Dolan, R. J. (2004). Neural
systems supporting interoceptive awareness. Nature Neuroscience, 7, 189–195.
doi:10.1038/nn1176
Cunningham, W. A., & Brosch, T. (2012). Motivational salience: Amygdala tuning
from traits, needs, values, and goals. Current Directions in Psychological Science,
21, 54–59. doi:10.1177/0963721411430832
Cunningham, W. A., Van Bavel, J. J., & Johnsen, I. R. (2008). Affective flexibility:
Evaluative processing goals shape amygdala activity. Psychological Science, 19,
152–160. doi:10.1111/j.1467-9280.2008.02061.x
Dalley, J. W., Everitt, B. J., & Robbins, T. W. (2011). Impulsivity, compulsivity,
and top-down cognitive control. Neuron, 69, 680–694. doi:10.1016/j.neuron.
2011.01.020
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the human brain. New York,
NY: Putnam.

196       wood and bechara

13490-08_Ch07-3rdPgs.indd 196 11/15/13 1:43 PM

 Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., & Damasio, A. R. (1994).
The return of Gage, Phineas—Clues about the brain from the skull of a famous
patient. Science, 264, 1102–1105. doi:10.1126/science.8178168
Davis, C., Levitan, R. D., Muglia, P., Bewell, C., & Kennedy, J. L. (2004). Decision-
making deficits and overeating: A risk model for obesity. Obesity Research, 12,
929–935. doi:10.1038/oby.2004.113
Devine, P. G. (1989). Stereotypes and prejudice—Their automatic and con-
trolled components. Journal of Personality and Social Psychology, 56(1), 5–18.
doi:10.1037/0022-3514.56.1.5
Dickstein, S. G., Bannon, K., Castellanos, F. X., & Milham, M. P. (2006). The neural
correlates of attention deficit hyperactivity disorder: An ALE meta-analysis.
Journal of Child Psychology and Psychiatry, 47, 1051–1062. doi:10.1111/j.1469-
7610.2006.01671.x
Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious.
American Psychologist, 49, 709–724. doi:10.1037/0003-066X.49.8.709
Ersche, K. D., Jones, P. S., Williams, G. B., Turton, A. J., Robbins, T. W., & Bullmore,
E. T. (2012). Abnormal brain structure implicated in stimulant drug addiction.
Science, 335, 601–604. doi:10.1126/science.1214463
Evans, J. S. B. T. (2008). Dual-processing accounts of reasoning, judgment, and
social cognition. Annual Review of Psychology, 59, 255–278. doi:10.1146/
annurev.psych.59.103006.093629
Everitt, B. J., Parkinson, J. A., Olmstead, M. C., Arroyo, M., Robledo, P., & Robbins,
T. W. (1999). Associative processes in addiction and reward—The role of
amygdala–ventral striatal subsystems. Annals of the New York Academy of Sci­
ences, 877, 412–438. doi:10.1111/j.1749-6632.1999.tb09280.x
Everitt, B. J., & Robbins, T. W. (2005). Neural systems of reinforcement for drug addic-
tion: From actions to habits to compulsion. Nature Neuroscience, 8, 1481–1489.
doi:10.1038/nn1579
Fein, G., & Chang, M. (2008). Smaller feedback ERN amplitudes during the BART
are associated with a greater family history density of alcohol problems in
treatment-naive alcoholics. Drug and Alcohol Dependence, 92(1–3), 141–148.
doi:10.1016/j.drugalcdep.2007.07.017
Fein, G., Klein, L., & Finn, P. (2004). Impairment on a simulated gambling task in
long-term abstinent alcoholics. Alcoholism: Clinical and Experimental Research,
28, 1487–1491. doi:10.1097/01.ALC.0000141642.39065.9B
Feinstein, J. S., Adolphs, R., Damasio, A., & Tranel, D. (2011). The human amyg-
dala and the induction and experience of fear. Current Biology, 21(1), 34–38.
doi:10.1016/j.cub.2010.11.042
Franklin, T. R., Acton, P. D., Maldjian, J. A., Gray, J. D., Croft, J. R., Dackis,
C. A., . . . Childress, A. R. (2002). Decreased gray matter concentration in the
insular, orbitofrontal, cingulate, and temporal cortices of cocaine patients. Bio­
logical Psychiatry, 51, 134–142. doi:10.1016/S0006-3223(01)01269-0

the neuroscience of dual systems      197

13490-08_Ch07-3rdPgs.indd 197 11/15/13 1:43 PM

 Frederick, S., Loewenstein, G., & O’Donoghue, T. (2002). Time discounting and
time preference: A critical review. Journal of Economic Literature, 40, 351–401.
doi:10.1257/002205102320161311
Gigerenzer, G., & Goldstein, D. G. (1996). Reasoning the fast and frugal way: Models
of bounded rationality. Psychological Review, 103, 650–669. doi:10.1037/0033-
295X.103.4.650
Goldstein, R. Z., & Volkow, N. D. (2002). Drug addiction and its underlying neuro­
biological basis: Neuroimaging evidence for the involvement of the frontal
cortex. The American Journal of Psychiatry, 159, 1642–1652. doi:10.1176/appi.
ajp.159.10.1642
Greene, J. D., Morelli, S. A., Lowenberg, K., Nystrom, L. E., & Cohen, J. D. (2008).
Cognitive load selectively interferes with utilitarian moral judgment. Cognition,
107, 1144–1154. doi:10.1016/j.cognition.2007.11.004
Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach
to moral judgment. Psychological Review, 108, 814–834. doi:10.1037/0033-
295X.108.4.814
Hanson, K. L., Luciana, M., & Sullwold, K. (2008). Reward-related decision-making
deficits and elevated impulsivity among MDMA and other drug users. Drug and
Alcohol Dependence, 96(1–2), 99–110. doi:10.1016/j.drugalcdep.2008.02.003
Henderson, L. A., Woo, M. A., Macey, P. M., Macey, K. E., Frysinger, R. C., Alger,
J. R., . . . Harper, R. M. (2003). Neural responses during Valsalva maneuvers in
obstructive sleep apnea syndrome. Journal of Applied Physiology, 94, 1063–1074.
Hollander, J. A., Lu, Q., Cameron, M. D., Kamenecka, T. M., & Kenny, P. J. (2008).
Insular hypocretin transmission regulates nicotine reward. Proceedings of the
National Academy of Sciences of the United States of America, 105, 19480–19485.
doi:10.1073/pnas.0808023105
Hugdahl, K. (1995). Psychophysiology: The mind–body perspective. Cambridge, MA:
Harvard University Press.
James, W. (1890). The Principles of Psychology. (Vol. 2). New York, NY: Holt.
Kapp, B. S., Supple, W. F., & Whalen, P. J. (1994). Effects of electrical stimulation of
the amygdalaloid central nucleus on neocortical arousal in the rabbit. Behavioral
Neuroscience, 108, 81–93. doi:10.1037/0735-7044.108.1.81
Kelley, A. E., Schiltz, C. A., & Landry, C. F. (2005). Neural systems recruited by
drug- and food-related cues: Studies of gene activation in corticolimbic regions.
Physiology & Behavior, 86(1–2), 11–14. doi:10.1016/j.physbeh.2005.06.018
Knudsen, E. I. (2007). Fundamental components of attention. Annual Review of
Neuro­science, 30, 57–78. doi:10.1146/annurev.neuro.30.051606.094256
Koenigs, M., Huey, E. D., Raymont, V., Cheon, B., Solomon, J., Wassermann, E. M.,
& Grafman, J. (2008). Focal brain damage protects against post-traumatic stress
disorder in combat veterans. Nature Neuroscience, 11, 232–237. doi:10.1038/
nn2032

198       wood and bechara

13490-08_Ch07-3rdPgs.indd 198 11/15/13 1:43 PM

 Koob, G. F., & Le Moal, M. (2001). Drug addiction, dysregulation of reward,
and allostasis. Neuropsychopharmacology, 24, 97–129. doi:10.1016/S0893-
133X(00)00195-0
Kross, E., & Mischel, W. (2010). From stimulus control to self-control: Toward
an integrative understanding of the processes underlying willpower. In
R. Hassin, K. Oschner, & Y. Trope (Eds.), Self-control in society, mind, and brain
(pp. 329–337). New York, NY: Oxford University Press. doi:10.1093/acprof:
oso/9780195391381.003.0023
Kupers, R. C., Gybels, J. M., & Gjedde, A. (2000). Positron emission tomography
study of a chronic pain patient successfully treated with somatosensory thalamic
stimulation. Pain, 87, 295–302. doi:10.1016/S0304-3959(00)00295-5
LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience,
23, 155–184. doi:10.1146/annurev.neuro.23.1.155
Lovallo, W. R., Yechiam, E., Sorocco, K. H., Vincent, A. S., & Collins, F. L. (2006).
Working memory and decision-making biases in young adults with a family his-
tory of alcoholism: Studies from the Oklahoma Family Health Patterns Project.
Alcoholism: Clinical and Experimental Research, 30, 763–773. doi:10.1111/j.1530-
0277.2006.00089.x
Makris, N., Gasic, G. P., Seidman, L. J., Goldstein, J. M., Gastfriend, D. R., Elman,
I., . . . Breiter, H. C. (2004). Decreased absolute amygdala volume in cocaine
addicts. Neuron, 44, 729–740. doi:10.1016/j.neuron.2004.10.027
Manes, F., Paradiso, S., & Robinson, R. G. (1999). Neuropsychiatric effects of insu-
lar stroke. Journal of Nervous and Mental Disease, 187, 707–712. doi:10.1097/
00005053-199912000-00001
McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D. (2004). Separate
neural systems value immediate and delayed monetary rewards. Science, 306,
503–507. doi:10.1126/science.1100907
McClure, S. M., Li, J., Tomlin, D., Cypert, K. S., Montague, L. M., & Montague,
P. R. (2004). Neural correlates of behavioral preference for culturally familiar
drinks. Neuron, 44, 379–387. doi:10.1016/j.neuron.2004.09.019
McRae, K., Misra, S., Prasad, A. K., Pereira, S. C., & Gross, J. J. (2012). Bottom-up
and top-down emotion generation: Implications for emotion regulation. Social
Cognitive and Affective Neuroscience, 7, 253–262. doi:10.1093/scan/nsq103
Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of delay of gratifica-
tion: Dynamics of willpower. Psychological Review, 106, 3–19. doi:10.1037/0033-
295X.106.1.3
Milad, M. R., & Quirk, G. J. (2002). Neurons in medial prefrontal cortex signal
memory for fear extinction. Nature, 420, 70–74. doi:10.1038/nature01138
Mischel, W., & Moore, B. (1973). Effects of attention to symbolically presented
rewards of self-control. Journal of Personality and Social Psychology, 28, 172–179.
doi:10.1037/h0035716

the neuroscience of dual systems      199

13490-08_Ch07-3rdPgs.indd 199 11/15/13 1:43 PM

 Mischel, W., Shoda, Y., & Rodriguez, M. L. (1989). Delay of gratification in children.
Science, 244, 933–938. doi:10.1126/science.2658056
Morewedge, C. K., & Kahneman, D. (2010). Associative processes in intuitive judg-
ment. Trends in Cognitive Sciences, 14, 435–440. doi:10.1016/j.tics.2010.07.004
Murray, E. A. (2007). The amygdala, reward and emotion. Trends in Cognitive Sci­
ences, 11, 489–497. doi:10.1016/j.tics.2007.08.013
Nair, S. G., Golden, S. A., & Shaham, Y. (2008). Differential effects of the hypo-
cretin 1 receptor antagonist SB 334867 on high-fat food self-administration
and reinstatement of food seeking in rats. British Journal of Pharmacology, 154,
406–416. doi:10.1038/bjp.2008.3
Naqvi, N. H., Rudrauf, D., Damasio, H., & Bechara, A. (2007). Damage to the insula
disrupts addiction to cigarette smoking. Science, 315, 531–534. doi:10.1126/
science.1135926
Nisbett, R. E., & Ross, L. D. (1980). Human inference: Strategies and shortcomings of
social judgment. Englewood Cliffs, NJ: Prentice-Hall.
Ochsner, K. N., Ray, R. R., Hughes, B., McRae, K., Cooper, J. C., Weber, J., . . . Gross,
J. J. (2009). Bottom-up and top-down processes in emotion generation: Com-
mon and distinct neural mechanisms. Psychological Science, 20, 1322–1331.
doi:10.1111/j.1467-9280.2009.02459.x
Paton, J. J., Belova, M. A., Morrison, S. E., & Salzman, C. D. (2006). The primate
amygdala represents the positive and negative value of visual stimuli during
learning. Nature, 439, 865–870. doi:10.1038/nature04490
Paulus, M. P. (2007). Decision-making dysfunctions in psychiatry—Altered homeo-
static processing? Science, 318, 602–606. doi:10.1126/science.1142997
Petty, R. E., & Cacioppo, J. T. (1984). Source factors and the elaboration likelihood
model of persuasion. Advances in Consumer Research, 11, 668–672.
Peyron, R., Laurent, B., & Garcia-Larrea, L. (2000). Functional imaging of brain
responses to pain. A review and meta-analysis (2000). Neurophysiologie Clinique-
Clinical Neurophysiology, 30, 263–288. doi:10.1016/S0987-7053(00)00227-6
Phillips, M. L., Drevets, W. C., Rauch, S. L., & Lane, R. (2003). Neurobiology of
emotion perception I: The neural basis of normal emotion perception. Biological
Psychiatry, 54, 504–514. doi:10.1016/S0006-3223(03)00168-9
Plassmann, H., O’Doherty, J., & Rangel, A. (2007). Orbitofrontal cortex encodes
willingness to pay in everyday economic transactions. The Journal of Neuro­
science, 27, 9984–9988. doi:10.1523/JNEUROSCI.2131-07.2007
Plassmann, H., O’Doherty, J., Shiv, B., & Rangel, A. (2008). Marketing actions can
modulate neural representations of experienced pleasantness. Proceedings of the
National Academy of Sciences of the United States of America, 105, 1050–1054.
doi:10.1073/pnas.0706929105

200       wood and bechara

13490-08_Ch07-3rdPgs.indd 200 11/15/13 1:43 PM

 Reyna, V. F. (2004). How people make decisions that involve risk–A dual-processes
approach. Current Directions in Psychological Science, 13, 60–66. doi:10.1111/
j.0963-7214.2004.00275.x
Reyna, V. F., & Ellis, S. C. (1994). Fuzzy-trace theory and framing effects in chil-
dren’s risky decision-making. Psychological Science, 5, 275–279. doi:10.1111/
j.1467-9280.1994.tb00625.x
Reyna, V. F., Lloyd, F. J., & Brainerd, C. J. (2003). Memory, development, and ratio-
nality: An integrative theory of judgment and decision-making. In S. Schneider
& J. Shanteau (Eds.), Emerging perspectives on judgment and decision research
(pp. 201–245). New York, NY: Cambridge University Press. doi:10.1017/
CBO9780511609978.009
Rodgers, R. J., Halford, J. C. G., de Souza, R. L. N., de Souza, A. L. C., Piper, D. C.,
Arch, J. R. S., . . . . Blundel, J. E. (2001). SB-334867, a selective orexin-1
receptor antagonist, enhances behavioural satiety and blocks the hyperphagic
effect of orexin-A in rats. European Journal of Neuroscience, 13, 1444–1452.
doi:10.1046/j.0953-816x.2001.01518.x
Schaefer, M., Berens, H., Heinze, H. J., & Rotte, M. (2006). Neural correlates of cultur-
ally familiar brands of car manufacturers. NeuroImage, 31, 861–865. doi:10.1016/
j.neuroimage.2005.12.047
Schultz, W. (2000). Multiple reward signals in the brain. Nature Reviews Neuro­
science, 1, 199–207. doi:10.1038/35044563
Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human infor-
mation processing. II. Perceptual learning, automatic attending, and a general
theory. Psychological Review, 84, 127–190. doi:10.1037/0033-295X.84.2.127
Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Impli-
cations for the rationality debate? Behavioral and Brain Sciences, 23, 645–665.
doi:10.1017/S0140525X00003435
Tranel, D., & Damasio, A. R. (1993). The covert learning of affective valence does
not require structures in hippocampal system or amygdala. Journal of Cognitive
Neuroscience, 5(1), 79–88. doi:10.1162/jocn.1993.5.1.79
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology
of choice. Science, 211, 453–458. doi:10.1126/science.7455683
Verdejo-Garcia, A., Benbrook, A., Funderburk, F., David, P., Cadet, J. L., & Bolla,
K. I. (2007). The differential relationship between cocaine use and marijuana
use on decision-making performance over repeat testing with the Iowa gam-
bling task. Drug and Alcohol Dependence, 90(1), 2–11. doi:10.1016/j.drugalcdep.
2007.02.004
Wager, T. D., Davidson, M. L., Hughes, B. L., Lindquist, M. A., & Ochsner, K. N.
(2008). Prefrontal–subcortical pathways mediating successful emotion regula-
tion. Neuron, 59, 1037–1050. doi:10.1016/j.neuron.2008.09.006

the neuroscience of dual systems      201

13490-08_Ch07-3rdPgs.indd 201 11/15/13 1:43 PM

 Wang, G. J., Tomasi, D., Backus, W., Wang, R., Telang, F., Geliebter, A., . . . Volkow,
N. E. (2008). Gastric distention activates satiety circuitry in the human brain.
NeuroImage, 39, 1824–1831. doi:10.1016/j.neuroimage.2007.11.008
Weller, J. A., Levin, I. P., Shiv, B., & Bechara, A. (2007). Neural correlates of adap-
tive decision making for risky gains and losses. Psychological Science, 18, 958–964.
doi:10.1111/j.1467-9280.2007.02009.x
Whitlow, C. T., Liguori, A., Livengood, L. B., Hart, S. L., Mussat-Whitlow, B. J.,
Lamborn, C. M., . . . Porrino, L. J. (2004). Long-term heavy marijuana users
make costly decisions on a gambling task. Drug and Alcohol Dependence, 76(1),
107–111. doi:10.1016/j.drugalcdep.2004.04.009
Williamson, J. W., McColl, R., & Mathews, D. (2003). Evidence for central com-
mand activation of the human insular cortex during exercise. Journal of Applied
Physiology, 94, 1726–1734.
Young, L., Bechara, A., Tranel, D., Damasio, H., Hauser, M., & Damasio, A. (2010).
Damage to ventromedial prefrontal cortex impairs judgment of harmful intent.
Neuron, 65, 845–851. doi:10.1016/j.neuron.2010.03.003

202       wood and bechara

13490-08_Ch07-3rdPgs.indd 202 11/15/13 1:43 PM

 Index

Note: “C.P.” in the index refers to Color Aging brain. See Older adults
Plates. Albert, D., 83, 107
Alcohol, 12, 86, 187
Abstract rewards, concrete vs., 149 American Psychological Association
ACC. See Anterior cingulate cortex (APA), xii
ACC/dmPFC BOLD signal, 22, 23, 28 Amygdala, 21–22
Accumulator, 23–24 in adolescents, 78–79
Adaptive adolescent period, 109 and delay of gratification, 153–156
Adaptive decision making, 31 and dual/triple processes, 182–187,
Addiction, 98, 152, 156, 187–189, 194 189, 191
ADHD (attention-deficit/hyperactivity fear response of, 45
disorder), 189 in fMRI studies, 80
Adolescence, 73–87 Anderson, S. W., 60
adaptive adolescent period in, 109 Animal studies, 17, 53, 101, 103, 104,
brain imaging and risk-taking 184, 188, 191
research in, 80–84 Anterior cingulate cortex (ACC), 20,
brain maturation in, 96–98 22, 26, 60
cognitive development in, 99–100, in adolescence, 99
C.P. 4 and delay gratification, 156
defined, 73 dorsal, 20
evidence for increases in risk taking in fMRI studies, 80
in, 75–78 rostral, 148–149, 156
hormonal factors during, 83–84 Anterior insula, 20, 28
juvenile justice system and develop- Anterior parahippocampal gyrus, 26
mental factors in, 107–109 Anterior temporal cortex, 79
neurobiological models of risk taking Antisaccade (AS) tasks, 81, 103
during, 78–80 Anxiety, 20, 22, 79
neuroimaging evidence of unique APA (American Psychological
reward processing in, 101–107 Association), xii
peer influences in, 83–84 Asian disease problem, 44, 47, 182
phases of, 73–74 Association for Psychological Science,
sensation-seeking period in, 93–111 xiii
social reorientation during, 79 Attention, in adolescents, 100
Adolescents Attentional bias, 153
directions for future research involv- Attentional control, 149, 157–159
ing, 85–87 Attention-deficit/hyperactivity disorder
impulsivity in, 29–30 (ADHD), 189
mortality among, 74 Attribute framing, 48–51
risk taking by, 74–75, 94–96 Autism, 32
ventral striatum responses to reward Ayduk, O., 148
in, 80–83, C.P. 2
Adrenergic receptors, 97 Bandy, D., 21, 190
Affective node, 79 Basal ganglia, 16, 189
Affective processing, and age, 124 development of, in adolescence, 96
Affective systems, framing effects and orbitofrontal cortex and, 101
activation of, 44–45 Basten, U., 21, 23

203

13490-09_Index-2ndPgs.indd 203 11/15/13 1:43 PM

 Bauer, A. S., 60 Childhood, 73. See also Adolescence
Bault, N., 77 Chronic pain patients, 190
Baumeister, R. F., 53 Cigarette smoking, 95, 191, 192
Bechara, A., 21, 60, 188 Cingulate cortex, 162
Beck, H., 57 Cingulum, 96
Becker–DeGroot–Marschak method, 19 Circadian rhythms, 52–54
Behaviorism, 31 Cocaine addiction, 187, 191–192
Betz, N., 18–19 Cognition, age-related changes in,
BIAS Task, 129 124
Biele, G., 20–21 Cognitive control, 153–156
Bilateral caudate, 20 Cognitive development, in adolescence,
Bilateral hippocampus, 26 99–100, C.P. 4
Bilateral parietal cortex, 26 Cognitive-regulatory node, 79
Bipolar personality disorder, 156 Cognitive strategies, for delay of
Bird, G., 86 gratification, 149–162
Bjork, J. M., 82 Cognitive theories, of decision making,
Blais, A.-R., 18–19 75–76
Blakemore, S. J., 77, 86 Cohen, J. D., 22–23, 182
Blood oxygen dependent (BOLD) Cohen, J. R., 105
signal, 19, 22–23, 187 Common-currency hypothesis, 17–18
Body mass index, 147 Comparison of options, 22–23
Bowman, C. R., 25–26 Concrete rewards, abstract vs., 149
Brain, 13. See also specific regions Conditioned stimuli, 183
framing effects and activation of Conflict between options, 23
affective systems in, 44–45 Control
glucose use by, 53 attentional, 149
and handedness, 54–55 cognitive, 153–156
maturation of, in adolescence, 96–98 self-, 28–30, 191
and reward/valuation, 16–19 Cool cognitive system, 153, 163–165,
and uncertainty, 15 180
white matter integrity, C.P. 8 Corbin, J., 57
Brain imaging, and risk-taking research Coricelli, G., 77
in adolescence, 80–84 Cornell University, xiii
Brammer, M., 81–82 Cortico-subcortical connectivity, 97
Braun, K. A., 48 Coupon redemption, 48–49
Bronfenbrenner, Uri, xii–xiii Craig, A. D., 21, 190
Bruine de Bruin, W., 58 Craving, 191–192
Burnett, S., 77, 86 Craving system, 178
Criminal behavior, 12
Cacioppo, J. T., 179 Crone, E. A., 76
Cake gambling task, 81 Cue processing, 103
Cannabis, 187–188 Cue reactivity paradigm, 155–156
Card gambling, 77 Cunningham, W. A., 184
Casey, B. J., 31, 78, 164–165
Ceci, Stephen, xii D1 receptors, 98
Central executive, 153 D2 receptors, 98
Chaiken, S., 179 D’Acremont, M., 20
Chein, J., 83, 107 Damasio, A. R., 189
Chen, G., 82 Damasio, H., 60
Chen, K., 21, 190 Davis, F. C., 53

204       index

13490-09_Index-2ndPgs.indd 204 11/15/13 1:43 PM

 Decision aids, for enhancement of Dorsal striatum, 16
choice, in older adults, 133–134 Dorsolateral prefrontal cortex (dlPFC),
Decision making. See also specific 27–30, 105–106
headings DOSPERT (Domain-Specific Risk-
absence of global decline in, in older Taking) scale, 18
adults, 132–133 Down-regulation, 29
adaptive, 31 Drug addiction, 187–189
cognitive theories of, 75–76 DSM–IV–REV (Diagnostic and Statistical
dual process models of, 177–194 Manual of Mental Disorders), 15
and handedness, 54–55 Dual process (term), 178
Decisions, functionally significant Dual process models, 177–194
features of, 30–31 addiction clinical example, 187–189
Decision strategies, effect of, on delay of with adolescents, 78–79, 82
gratification, 150 in decision making, 181–182
Decision value, 105 history of, 178–181
Delay of gratification, 145–166 and insular cortex, 189–194
cognitive strategies enabling, neurological evidence for, 182–186
149–162 opposition to, 186–187
definition of, 145 Dvorak, R. D., 54
effect of decision strategies on, 150
effect of reappraisal on, 150–151 Economic approach (to defining risky
experimental findings on, 151–152 decision making), 14–16
future research, directions for, Eigsti, I.-M., 159
162–165 Einarson, A., 48
hot/cool systems approach to, Elaboration likelihood model, 179
152–165 Emotion regulation, 22
longitudinal findings on, 151 Emotion suppression, 51–52, 62
preschool, 146–149 Endocannabinoids, 97
De Martino, B., 22, 23 Environmental factors, in adolescent
Dennis, N. A., 25–26 risk taking, 95–96
Detection node, 79 Ernst, M., 83
Devine, P. G., 179 Esctasy (narcotic), 187
Diagnostic and Statistical Manual of Estrada, S. M., 27
Mental Disorders (DSM–IV– EV. See Expected value
REV), 15 Executive control, in adolescents, 99,
Diamond, W. D., 48 100
Dickinson, D. L., 53 Executive functions, 153
Diffusion tensor imaging studies, 96–97, Expected utility, 14, 24–25
131 Expected value (EV), 14–15, 20, 24–25,
Distractions, 150 27, 75–78, 133
Distractors, 99 Experience, learning from, 23–24
dlPFC (Dorsolateral prefrontal cortex), Explicit memory, 131
27–30, 105–106 Extended amygdala, 182
Dolan, S., 188 Exteroception, 190
Domain-Specific Risk-Taking Eye-tracking technology, 57
(DOSPERT) scale, 18
Dopamine, 97–98, 108, 131 Faces, 184
Dopaminergic activity, 16 False memory, 25
Dorsal anterior cingulate cortex, 20 Fear, 22, 45, 58, 79
Dorsal hypothalamic region, 17 Fecteau, S., 29

index      205

13490-09_Index-2ndPgs.indd 205 11/15/13 1:43 PM

 Feedback, 77, 82, 127, 129 Fuzzy trace theory, 12–16, 20, 21, 24–28,
FEF (frontal eye field), 103 30–33, 35, 38–40, 45, 56–57, 62,
Fetal risk, 48 68, 76, 85, 90, 100, 118, 132, 134,
Fiebach, C. J., 21 140, 152, 174, 175, 178–180,
Figner, B., 77 200, 201
Firearms, 12
Fischhoff, B., 58 GABA (gamma-aminobutyric acid), 97
Fixed economy, 107 Gaeth, G. J., 48
fMRI. See Functional magnetic Gage, Phineas, 185
resonance imaging Gain and loss learning, in older adults,
fMRI studies. See Functional MRI 127–128
studies Gains, neural substrates differentiating
Food rewards, money rewards vs., 18 losses, probabilities, and, 19–22
Forbes, E. E., 84 Gain trials, 58–59
Foster, R. G., 53 Gallo, M., 48
Framing effects, 26–27, 43–63 Galvan, A., 81, 86
and activation of affective systems in Gambling, 18, 46, 76, 81, 188. See also
brain, 44–45 Iowa Gambling Task (IGT)
attribute, 48–51 Gamma-aminobutyric acid (GABA), 97
and dual process models, 45, 182 Geier, C. F., 107
and emotion suppression, 51–52 Giampetro, V., 81–82
future research, directions for, 62–63 Gianotti, L. R., 28–29
and handedness, 54–55 Gist-based reasoning, 56–57, 76, 134,
with impaired vs. unimparied older 179–180
decision makers, 59–61 Gist memory representations, 23–28,
with older adults, 57–61 31, 133
and physiological vs. behavioral Glimcher, P. W., 17–18
measures, 55–57 Glucose, 52–54
reflection effects vs., 46 Glutamate, 97
risky-choice, 47, 50, 52–54 Gmax, 27, 28
and System 1 vs. System 2 decision Goals, incongruous, 181
making, 45–46 Goel, R., 48
ubiquity of, 44 Go/no-go tasks, 157–158, 160, C.P.s
Frederick, S., 45, 182 9–10
Frontal eye field (FEF), 103 Graham v. Florida, 108
Frontal lobe hypothesis of aging, 58 Gratification, delay of. See Delay of
Functionally significant features, of gratification
decisions, 30–31 Gross, J. J., 29, 51, 154
Functional magnetic resonance imaging Gyurak, A., 156
(fMRI), 19, 20, 60, 62, 125
Functional MRI (fMRI) studies, 80 Hajcak, G., 154
of reward learning, in older adults, Halari, R., 81–82
127 Handedness, 54–55
of reward processing, 103–108 Head Start, xiii
and self-esteem/attentional control, Heerkeren, H. R., 20–21
156 Heuristic-systematic model of
of working memory, 99 information processing, 179
Fusiform face area, 79 High temptation focus, 158
Fuzzy memory representations, 24 Hindes, A., 188
Fuzzy reasoning, 179 Hippocampus, 24, 26, 155

206       index

13490-09_Index-2ndPgs.indd 206 11/15/13 1:43 PM

 Homicide, 12 International Statistical Classification of
Hommer, D. W., 82 Diseases and Related Health Prob-
Hormonal factors, in adolescence, 80, lems (ICD–10), 15
83–84, C.P. 3 Interoception, 190
Hot/cool framework (delay of gratifica- Intuition, 14, 24
tion), 152–165 Intuitive system, 178
cool cognitive system in, 153, 163 Iowa Gambling Task (IGT), 24–26,
and development, 157–159 59–60, 77, 132, 185–188
hot limbic system in, 152–153,
163–164 Jackson v. Hobbs, 108
individual differences in, 155–156 James, William, 45, 189
interaction between cool and hot James McKeen Catell Fellow Award,
systems in, 164–165 xiii
neural correlates in, 159–162 Jansen, B. R. J., 77
with preschoolers, 157 Jasper, J. D., 48
relative activation of hot and cool Jones, R. M., 78
systems in, 153–155 Juvenile justice system, 107–109
Hot emotional system, 180
Hot limbic system, 152–153, 163–165 Kahneman, D., 44, 45, 47, 55
Huettel, S. A., 28, 60 Koren, G., 48
Huizenga, H. M., 77 Kühberger, A., 27, 57
Hunt, L. T., 19
Hutcherson, C. A., 29 Laboratory experiments, 12–13, 31
Laibson, D. I., 182
IFG. See Inferior frontal gyrus Lateral parietal cortex, 25
IGT. See Iowa Gambling Task Lateral prefrontal cortices, 58
Illicit substances, use of, 86 Learning from experience, 23–24
Immaturity, in adolescents, 108–109 Levenson, R. J., 51
Impaired decision making, framing Levin, I. P., 46, 48, 49
effects with, 59–61 Levy, D. J., 17–18
Impulse control, 157–159 Life expectancies, 49
Impulsive system, 178 Lmin, 27, 28
Impulsivity, 13, 28–31, 152, 154, 189 Loewenstein, G., 182
Incarcerated youths, 29–30 Loss anticipation, 126
Incentives, 78–79 Loss aversion, 15–16, 21–22, 28
Incongruous goals, 181 Losses
Individual differences and memory representations of risk
in hot/cool framework, 155–156 preference, 26–27
in older adults, 132 neural substrates differentiating
Inferior frontal gyrus (IFG), 153, 161, gains, probabilities, and,
163 19–22
Information processing Low temptation focus, 158
heuristic-systematic model of, 179 Luna, B., 107
social, 79–80, 86
Inhibition, 28–30 Mackinlay, R. J., 77
Inhibitory control, in adolescents, Magnetic resonance imaging (MRI),
99–100, 104 96. See also Functional magnetic
Insula (insular cortex), 20–22, 28, 86, 154 resonance imaging (fMRI)
and dual process models, 189–194 Magnetoencephalography, 19
and fear response, 45 Major depressive disorder, 156

index      207

13490-09_Index-2ndPgs.indd 207 11/15/13 1:43 PM

 Making Human Beings Human Nucleus accumbens, 16, 80, 86, 101,
(Bronfenbrenner), xiii 153
Marijuana, 86 Nystrom, L. E., 22–23
Masicampo, E. J., 53
McClelland, Peter, xii O’Brien, L., 83, 107
McClure, E. B., 79 Occipital cortex, 26
McClure, S. M., 182 Ochsner, K. N., 155
McElroy, T., 53, 55, 57, 62 Oculomotor control regions, 103–104
McMahon, A. J., 54 O’Doherty, J., 19
Medial orbitofrontal cortex (mOFC), OFC. See Orbitofrontal cortex
19, 21, 23 Older adults, 123–135
Medial prefrontal cortex (MPFC), 20, absence of global decline in decision
25, 60, 125, C.P. 5 making in, 132–133
Medial temporal lobe (MTL), 24, 25 and changes in neural network,
Memory representations, 24–28 130–131
encoding of, 24 decision aids for enhancement of
of risk preference, in decision choice in, 133–134
making, 26–28 framing effects in, 57–61
Menaker, M., 53 future research, directions for, 135
Metcalfe, J., 180 gain and loss learning in, 127–128
Midbrain dopamine areas, 16 individual differences in, 132
MID (monetary incentive delay) processing of monetary gains/losses
tasks, 82 in, 125–127, C.P.s 5–6
Miller v. Alabama, 108 and research applicability to real-
Mills, B. A., 27 world tasks, 134
Mischel, W., 180, 181 risky decision making in, 128–130,
Moen, Phyllis, xii C.P. 7
Mohr, P. N., 20–21 Op de Macks, Z. A., 84
Monetary gains/losses, processing of, in Orbitofrontal cortex (OFC), 26
older adults, 125–127, C.P.s 5–6 in fMRI studies, 80
Monetary incentive delay (MID) tasks, in older adults, 58
82, 84 and reward processing in adoles-
Monetary incentives, 106–107 cence, 101, 105–106
Money rewards, food rewards vs., 18
Motivation, in adolescents, 99, 105–106 Parahippocampal regions, 24
Motor response stage, 103 Pardo, J. V., 60
MPFC. See Medial prefrontal cortex Parietal cortex, 20, 23, 24
MRI (magnetic resonance imaging), bilateral, 26
96. See also Functional magnetic lateral, 25
resonance imaging (fMRI) posterior, 27–28
MTL (medial temporal lobe), 24, 25 Parietal eye field, 103
Myelin sheath, 96 Parker, A. M., 58
Payne, J. W., 49
Naqvi, N. H., 192 Peer influences, 83–84
Nelson, E. E., 79, 80, 86 Petty, R. E., 179
Neural network, changes in, in older PFC. See Prefrontal cortex
adults, 130–131 Phantom recollection, 25
Neurochemical maturation, during ado- Phenomenology, 25
lescence, 97–98 Pine, D. S., 79
Nicotine addiction, 191, 192 Plassman, H., 19, 29

208       index

13490-09_Index-2ndPgs.indd 208 11/15/13 1:43 PM

 Pmax, 27 Reasoning system, 178
Pochon, J. B., 22–23 Reconstruing, 149–150
Positive feedback, 127 Reflection effects, 46
Positivity effects, 125, 126 Reflective system, 178
Posterior cingulate cortex, 17–18, 25 Reiman, E., 21, 190
Posterior parietal cortex, 27–28 Rejection sensitivity (RS), 148
Posttraumatic stress disorder (PTSD), Response preference, 23–30
183 and impulsivity/inhibition of
Precentral cortex, 26 responses, 28–30
Prediction error, 105 and memory representations, 24–28
Preference reversal, 181–182 Reward(s)
Prefrontal cortex (PFC), 16, 189. See concrete vs. abstract, 149, 151
also Ventromedial prefrontal and expected value, 14
cortex (vmPFC) neural substrates of, 16–19
in adolescents, 79 representation/remembrance of, 13
and cognitive control, 153–156, 161 Reward accrual, 103
development of, in adolescents, Reward anticipation, 82
96–98 Reward circuit (of brain), 16, 22
functional maturation of, 58 Reward prediction, 184
medial, 20, 25, 60, 125, C.P. 5 Reward processing
superior, 26 in aging brain. See Older adults
Prefrontal systems, in adolescence, 99, neuroimaging evidence of, in adoles-
100 cence, 101–107, C.P. 1
Pregnancy, 48 Reward receipt, ventral striatum and,
Preschoolers, delay of gratification in, 81–83
146–152, 157–159 Reward sensitivity, 13, 31, 76, 103, 107,
Primate studies, 103, 104 162
Probabilities, neural substrates differen- Reward types, risk preferences across, 17
tiating gains, losses, and, 19–22 Reyna, V. F., 27, 76
Prospect theory, 56 Rhesus macaques, 184
Psychological approach (to defining Right anterior insula, 20
risky decision making), 14–16 Riis, J., 22–23
Psychopathy, 12 Risk attitude, 14, 16, 18
PTSD (posttraumatic stress disorder), Risk aversion, 12–17
183 Risk perception, 13
Puberty, 74, 80, 83–84, 93, 109 Risk preference, 13, 14
Public health, risky decision making as across reward types, 17
threat to, 12 memory representations of,
in decision making, 26–28
rACC. See Rostral anterior cingulate in older adults, 129
cortex Risks, defined, 74
Ralph, M. R., 53 Risk taking
Rangel, A., 19, 29 by adolescents, 74–87, 94–96
Reaction times (in choice selection), 23 healthy vs. unhealthy, 15
Real-world risky behaviors, 13 Risky-choice framing, 47, 50, 52–54
Real-world tasks, research applicability Risky decision making, 152
to, in older adults, 134 economic vs. psychological
Reappraisal, effect of, on delay of approaches to defining, 14–16
gratification, 150–151 in laboratory vs. real world, 12–13
Reappraising, 149–150 in older adults, 58–59, 128–130

index      209

13490-09_Index-2ndPgs.indd 209 11/15/13 1:43 PM

 Roper v. Simmons, 108 Socioemotional selectivity theory,
Rostral anterior cingulate cortex 125–126
(rACC), 148–149, 156 Somerville, L. H., 78, 79, 83
RS (rejection sensitivity), 148 Stanovich, K. E., 45–46, 49, 179, 180
Rubia, K., 81–82 The State of Americans (Bronfenbrenner
et al.), xii–xiii
Samanez-Larkin, G. R., 126 Steinberg, L., 83, 107
Sanfey, A. G., 22–23 Stereotypes, 53, 179
Sanyal, A., 48 Striatum, 22, 182–186. See also Ventral
Satisficing, 133 striatum (VS)
SB-34867 receptor agonist, 191 in cocaine addicts, 187
Scheel, M. H., 54 in common-currency hypothesis, 17
Schneider, W., 178 dorsal, 16
Schonberg, T., 29 and reward processing in adolescence,
Schott, B. H., 126–127 105, 106
SCN (suprachiasmatic nucleus), underrecruitment of, in middle
52–53 adolescence, 82
SCRs (skin conductance responses), Stroke, 192
185–186, 188 “Strong-handedness,” 55
Secondary inducers, 184 Subcortical brain, 58, 78
Self-control, 28–30, 191 Superior longitudinal fasciculus, 96
Self-esteem, 156 Superior parietal cortex, 26
Sensation seeking, 13, 16, 18, 76, 84 Superior prefrontal cortex, 26
Sensation-seeking period, adolescent, Superior temporal sulcus, 79
93–111 Supplementary eye field, 103
and adaptive adolescent period, 109 Suprachiasmatic nucleus (SCN), 52–53
and brain maturation, 96–98 Sustained attention tasks, 81–82
and cognitive development, 99–100 Synaptic pruning, 96
and juvenile justice system, 107–109 System 1 decision making, 45–46, 62,
and neuroimaging evidence of 178–183
unique reward processing, System 2 decision making, 45–46, 52,
101–107 62, 178–181
and risk taking, 94–96
Serotonin, 97 Tanner, C., 27, 57
Seta, J., 55 Teens, homicide among, 12
Sexual risk taking, 74, 76, 86, 146 Temporal discounting, 81, 152, 181
Shannon, B. J., 29 Temporal lobe, medial, 24, 25
Shiffrin, R. M., 178 Temporal sulcus, superior, 79
Single process theories, 186–187 Temptation, 146–150
Skin conductance responses (SCRs), Temptation focus, 158
185–186, 188 Testosterone, 84, C.P. 3
Smith, A. B., 81–82 Thalamus, C.P. 8
Smith, A. R., 82 Thiruchselvam, R., 154
Social information processing model, Thompson, S., 86
79–80, 86 Transmagnetic cranial stimulation
Social reorientation, during (TMS), 29
adolescence, 79 Triadic model, 79
Society for Neuroeconomics, xii Triple process model, 79, 192–194
Society for Research in Child True memory, 25
Development, xii Tversky, A., 44, 47, 55

210       index

13490-09_Index-2ndPgs.indd 210 11/15/13 1:43 PM

 Uckert, K., 83, 107 in cocaine addicts, 187
Uncertainty, 15–16 in common-currency hypothesis,
Uncinate fasciculus, 96 17, 18
Unique reward processing, neuro­ and down-/up-regulation, 29
imaging evidence of, and dual processing, 184–188
in adolescence, 101–107 and fear response, 45
Up-regulation, 29 fMRI studies, 19
Urie Bronfenbrenner Award for Life- and IGT performance, 60
time Contribution to Develop- and reward processing in adoles-
mental Psychology in the Service cence, 105–106
of Science and Society, xiii in triple process model, 193–194
U.S. Supreme Court, 108 Verbatim memory representations,
24–28, 56, 134
Valuation, neural substrates of, 16–19 Vigilance, 22, 160
Vandekar, S. N., 25–26 Visser, I., 77
van Duijvenvoorde, A. C. K., 77 Visual cortex, 26, 96
van Leijenhorst, L., 76, 81 vmPFC. See Ventromedial prefrontal
Variance, 14 cortex
Venkatraman, V., 27, 28 vmPFC/mOFC signal, 19, 21, 23
Ventral striatum (VS), 16 Vohs, K. D., 165
in adolescents, 78, 80–83, C.P. 2 VS. See Ventral striatum
and anticipation of monetary losses,
in older adults, 125, C.P. 5 Wang, X. T., 54
and available reward processing Weber, E. U., 18–19, 77
studies in adolescence, C.P. 1 Weller, J. A., 58, 184n3
in fMRI studies, 80 West, R. F., 45–46, 49, 179, 180
and “hot system,” 153 Westenberg, P. M., 76
responses of, to reward, in Wethington, Elaine, xii
adolescents, 80–83 Wheel of fortune task, 81
and reward processing in White matter integrity, C.P. 8
adolescence, 101–105, 108 Wilkening, F., 77
and reward processing in older Wisconsin Card Sorting Test, 185
adults, 127 Wood, S. M. W., 21
and risky decision making, in older Working memory
adults, 130 fMRI studies of, 99
Ventromedial prefrontal cortex inferior frontal gyrus and, 153
(vmPFC), 16, 28
and amygdala, 22, 30 Zanolie, K., 81

index      211

13490-09_Index-2ndPgs.indd 211 11/15/13 1:43 PM

13490-09_Index-2ndPgs.indd 212 11/15/13 1:43 PM
 About the Editors

Valerie F. Reyna, PhD, is director of the Human Neuroscience Institute

at Cornell University and former president of the Society for Judgment and
Decision Making, professor and codirector of the Center for Behavioral
Economics and Decision Research at Cornell University, and codirector of
the Cornell University Magnetic Resonance Imaging Facility. She is a devel-
oper of fuzzy trace theory, a model of memory, decision making, and develop-
ment that is widely applied in law, medicine, and public health. Her recent
work has focused on numeracy, medical decision making, risk perception
and risk taking, neurobiological models of development, and neurocognitive
impairment and genetics. Dr. Reyna has been a leader in using memory prin-
ciples such as accessibility and mathematical models of memory to explain
judgment and decision making. Among her theoretical proposals, she is par-
ticularly well known for a model of intuition that places it at the apex of
judgment and decision making, rather than treating it as a developmentally
primitive process. She also helped to initiate what is now a burgeoning area
of research on developmental differences in judgment and decision making.
Her research supports an evidence-based explanation of neural and psycho-
logical processes of risk taking in adolescence and adulthood, which predicts

213

13490-10_AboutED-3rdPgs.indd 213 11/15/13 1:44 PM

 real-world behaviors. The author of more than 175 publications that have
been cited more than 8,000 times, Dr. Reyna is a fellow of numerous scien-
tific societies and has served on scientific panels of the National Research
Council, the National Science Foundation, the National Institutes of Health,
the MacArthur Foundation, and the National Academy of Sciences.

Vivian Zayas, PhD, is an associate professor of psychology at Cornell

University. Her research examines the cognitive and affective processes
involved in delay of gratification and the interplay between attachment
and affiliative processes, on the one hand, and self-control processes, on the
other, using theoretical frameworks and methods that cross traditionally
defined boundaries between social and personality psychology and cognitive
psychology and cognitive neuroscience and developmental psychology. Her
research has appeared in journals such as Psychological Science, the Journal
of Personality and Social Psychology, Proceedings of the National Academy
of Sciences, Personality and Social Psychology Bulletin, Child Development,
Nature Neuroscience, and the Journal of Personality. She has received fund-
ing from the National Science Foundation and the National Institutes of
Health.

214       about the editors

13490-10_AboutED-3rdPgs.indd 214 11/15/13 1:44 PM

 60 40 20 0 -20 -40 -60 -80 -100

1 May (2004)

MNI coordinates (y axis)

2 Bjork (2004) 60 60

3 Ernst (2005)
40 40
4 Galvan (2006)
6
5 Van Leijenhorst-(2010a) 10
8
20 2 12 20
3 4 7
6 Van Leijenhorst-(2010b) 6 5 9 1
11 7
7 Geier (2010)
0 0
8 Bjork (2010)
9 Chein (2011) -20 -20
10 Smith (2011)
11 Christakou (2011) -40 -40
12 Padmanabhan (2011)

60 40 20 0 -20 -40 -60 -80 -100

MNI coordinates (z axis)

Age-related decrease Adolescent transition

Age-related increase No age difference

Color Plate 1. Meta-analysis of available reward processing studies in adolescence,

with a specific focus on the ventral striatum.

Alone
Peer present
Estimated % Signal Change

Color Plate 2. Left panel: Age (adolescents, young adults, adults) × Social Condition
(peer present vs. alone) interaction in the right ventral striatum (vs. MNI peak coordi­
nates, x = 9, y = 12, z = -8). Right panel: Mean estimated BOLD signal change (beta
coefficients) in adolescents, young adults, and adults under ALONE (blue bars) and
PEER PRESENT (red bars) conditions. Error bars indicate standard error of the
mean. Adapted from “Peers Increase Adolescent Risk Taking by Enhancing Activity
in the Brain’s Reward Circuitry,” by J. Chein, D. Albert, L. O’Brien, K. Uckert, and
L. Steinberg, 2011, Developmental Science, 14, p. F6. Copyright 2011 by Wiley-
Blackwell. Adapted with permission.

13490-11_Color Plates-3rdPgs.indd 1 11/15/13 3:43 PM

 8
4
6
3

4
2

2 1

0 0
p<.001 p<.005
Boys (unc.) Girls (unc.)

Color Plate 3. Regions of activation for reward > loss with testosterone as predictor
included the bilateral ventral striatum in boys (left) and left ventral striatum in girls
(right), at a threshold of p < .005, uncorrected. Colored bars represent t values.
Adapted from “Testosterone Levels Correspond With Increased Ventral Striatum
Activation in Response to Monetary Rewards in Adolescents,” by Z. A. Op de Macks,
B. Gunther Moor, S. Overgaauw, B. Guroglu, R. E. Dahl, and E. A. Crone, 2011,
Developmental Cognitive Neuroscience, 1, p. 512. Copyright 2011 by Elsevier.
Adapted with permission.

13490-11_Color Plates-3rdPgs.indd 2 11/15/13 3:43 PM

 Color Plate 4. fMRI studies delineating changes in brain function underlying cogni-
tive development. (A). Adolescents show decreased recruitment of dorsal anterior
cingulate cortex (dACC) during inhibitory errors reflecting limitations in performance
monitoring. From “Maturational Changes in Anterior Cingulate and Frontoparietal
Recruitment Support the Development of Error Processing and Inhibitory Control,”
by K. Velanova, M. E. Wheeler, and B. Luna, 2008, Cerebral Cortex, 18, p. 2516.
Copyright 2008 by Oxford University Press. Adapted with permission. (B). Adoles-
cents show decreased recruitment of the frontoposterior network supporting the
maintenance of cognitive control (Velanova et al., 2009). (C). Granger effective
connectivity indicates that prefrontal top-down connectivity becomes established
by adolescence and subsequently strengthens into adulthood (Hwang et al., 2010).
(D). Resting state connectivity shows that while the foundational hub architecture of
the brain is already established in childhood, prefrontal hub to non-hub connections
become established from childhood to adolescence subsequently showing small
refinements integrating cerebellar hubs. From “The Development of Hub Architec-
ture in the Human Functional Brain Network,” by K. Hwang, M. N. Hallquist, and
B. Luna, 2012, Cerebral Cortex. Advance online publication. Copyright 2012 by
Oxford University Press. Adapted with permission.

13490-11_Color Plates-3rdPgs.indd 3 12/9/13 12:00 PM

13490-11_Color Plates-3rdPgs.indd 4

Gain Anticipation Gain Outcome

Younger Adults + $5.00 Younger Adults
+ $0.50 Hit (+$)
+ $0.00 Miss (+0)

MPFC % signal ∆
0.3 0.3

VS % signal ∆
0 0

–0.3 –0.3
2s 2s

Older Adults + $5.00 Older Adults

+ $0.50 Hit (+$)
+ $0.00 Miss (+0)

MPFC % signal ∆
0.3 0.3
VS % signal ∆

0 0

–0.3 –0.3
2s 2s

Color Plate 5. Ventral striatal functional activity is modulated by magnitude during anticipation of monetary gains in both younger and
older adults (left panel). Medial prefrontal regions show similar differences in signal change between gain and nongain outcomes in both
younger and older adults (right panel). All statistical brain maps displayed are thresholded at p < .0001 uncorrected. All error bars on
timecourses are SEM (standard error of the mean).
11/15/13 3:43 PM
13490-11_Color Plates-3rdPgs.indd 5

Loss Anticipation Loss Outcome

Younger Adults –$5.00 Younger Adults
–$0.50 Hit (–0)
–$0.00 Miss (–$)

Insula % signal ∆
0.3 0.3

VS % signal ∆
0 0

–0.3 –0.3
2s 2s

Older Adults –$5.00 Older Adults

–$0.50 Hit (–0)
Insula % signal ∆ –$0.00 Miss (–$)
0.3 0.3

VS % signal ∆
0 0

–0.3 –0.3
2s 2s

Color Plate 6. Functional activity in the caudate and insula is modulated by magnitude during anticipation of monetary gains in younger
but not older adults (left panel). In contrast, medial prefrontal and ventral striatal regions show similar differences in signal change
between nonloss and loss outcomes in both younger and older adults (right panel). All statistical brain maps displayed are thresholded at
p < .0001 uncorrected. All error bars on timecourses are SEM (standard error of the mean).
11/15/13 3:43 PM
 .8 1 n.s.

Proportion of optimal choices

Risk seeking (stock)
n.s.
Risk aversion (bond)
Proportion of choices

.6 .75

.4 .50

.2 .25

0 0
0 20 40 60 80 100 YA OA
Age BL BL DV EV

Color Plate 7. In a risky financial investment task (N = 110), older adults make more
mistakes when choosing assets that are probabilistically associated with rewards
(selection of stocks) but do not differ from younger adults in risk aversion (selection of
bonds). These behavioral age differences were associated with increased neural vari-
ability with age in the striatum (N = 53). Statistical maps are thresholded at p < .0001,
uncorrected. In a follow-up study, we found that decision aids which provided a visual
depiction of the running history of prior outcomes (discrete value, DV) or a sum-
mary of prior performance and prediction of future performance (integrated value,
IV) increased performance in older adults to the performance level of younger adults
at baseline (BL). From “Variability in Nucleus Accumbens Activity Mediates Age-
Related Suboptimal Financial Risk Taking,” by G. R. Samanez-Larkin, C. M. Kuhnen,
D. J. Yoo, and B. Knutson, 2010, The Journal of Neuroscience, 30, pp. 1429–1430.
Copyright 2010 by the Society for Neuroscience. Adapted with permission. And from
“Expected Value Information Improves Financial Risk Taking Across the Adult Life
Span,” by G. R. Samanez-Larkin, A. D. Wagner, and B. Knutson, 2011, Social
Cognitive and Affective Neuroscience, 6, p. 214. Copyright 2010 by Oxford University
Press. Adapted with permission.

0.550 1.0
WM integrity

Learning

0.425 0.5

0.300 0
20 55 90 0.30 0.425 0.55
Age WM integrity

Color Plate 8. White matter integrity from the thalamus to the medial prefrontal
cortex (left, blue) and the medial prefrontal cortex to the ventral striatum (right,
green) is associated with both age and reward learning performance. From
“Frontostriatal White Matter Integrity Mediates Adult Age Differences in Proba-
bilistic Reward Learning,” by G. R. Samanez-Larkin, S. M. Levens, L. M. Perry,
R. F. Dougherty, and B. Knutson, 2012, The Journal of Neuroscience, 32, p. 5335.
Copyright 2012 by the Society for Neuroscience. Adapted with permission.

13490-11_Color Plates-3rdPgs.indd 6 12/9/13 12:00 PM

 Color Plate 9. Activation maps and percent change in magnetic resonance (MR) sig-
nal during the “hot” go/no-go task as a function of go and no-go trials. (A). The right
inferior frontal cortex was associated with correct inhibition of a response (no-go)
relative to making a correct response (go). (Left) Activation map depicting right infe-
rior frontal gyrus activation, thresholded at p < .05, whole-brain corrected displayed
on a representative high-resolution T1 weighted axial image. (Right) Percent change
in MR signal for go and no-go trials in the inferior frontal gyrus. (B). The left primary
motor cortex was associated with making a correct response (go) relative to correctly
withholding a response. (Left) Activation map depicting activation in left precentral
gyrus, thresholded at p < .05, whole-brain corrected displayed on a representative
high-resolution T1 weighted axial image. (Right) Percent change in MR signal for go
and no-go trials in the left precentral gyrus. (C). The cerebellum was associated with
making a correct response (go) relative to correctly withholding a response. (Left)
Activation map depicting cerebellum activation, thresholded at p < .05, whole-brain
corrected displayed on a representative high resolution T1 weighted axial image.
(Right) Percent change in MR signal for go and no-go trials in the cerebellum. Error
bars represent 1 standard error above and below the mean. From “Behavioral and
Neural Correlates of Delay of Gratification 40 Years Later,” by B. J. Casey et al., 2011.
Proceedings of the National Academy of Sciences of the United States of America,
108, p. 15000. Copyright 2011 by the National Academy of Sciences. Adapted with
permission.

13490-11_Color Plates-3rdPgs.indd 7 11/15/13 3:43 PM

 Color Plate 10. Low delay ability in early childhood predicts greater recruitment
of ventral striatum to happy no-go trials (i.e., when inhibiting responses to positive
social cues) 40 years later. (Left) Activation map for the three-way interaction of task,
emotion, and delay group depicting ventral striatum activity thresholded at p <.05,
small volume corrected displayed on a representative high-resolution T1 weighted
axial image. (Right) Percent change in magnetic resonance (MR) signal to happy
no-go trials in high and low delayers. Error bars represent 1 standard error above
and below the mean. From “Behavioral and Neural Correlates of Delay of Gratifica-
tion 40 Years Later,” by B. J. Casey et al., 2011. Proceedings of the National Acad-
emy of Sciences of the United States of America, 108, p. 15001. Copyright 2011 by
the National Academy of Sciences. Adapted with permission.

13490-11_Color Plates-3rdPgs.indd 8 11/15/13 3:43 PM