Review TRENDS in Cognitive Sciences Vol.10 No.

3 March 2006

Neuroeconomics: cross-currents
in research on decision-making
Alan G. Sanfey1, George Loewenstein2, Samuel M. McClure3 and Jonathan D. Cohen3
1
Department of Psychology, University of Arizona, Tucson, AZ 85721, USA
2
Department of Social and Decision Sciences, Carnegie Mellon University, Pittsburgh, PA 15213, USA
3
Department of Psychology, and Center for the Study of Brain, Mind and Behavior, Princeton University, Princeton, NJ 08544, USA

Despite substantial advances, the question of how we Although we are perhaps not as optimistic as the most
make decisions and judgments continues to pose ardent believers in neuroeconomics when it comes to the
important challenges for scientific research. Historically, time-line of progress, we do believe that the field has real
different disciplines have approached this problem using potential for making important contributions to our
different techniques and assumptions, with few unifying understanding of decision-making, above and beyond
efforts made. However, the field of neuroeconomics has what has and will continue to be learned from work
recently emerged as an inter-disciplinary effort to bridge within each discipline independently. This is because
this gap. Research in neuroscience and psychology has neuroeconomics is able to draw upon the complementary
begun to investigate neural bases of decision predict- strengths of its contributing disciplines. In fact, the
ability and value, central parameters in the economic benefits of increasing contact between neuroscience,
theory of expected utility. Economics, in turn, is being psychology and economics are already apparent.
increasingly influenced by a multiple-systems approach The central argument of this article is that economics,
to decision-making, a perspective strongly rooted in psychology and neuroscience can each benefit from taking
psychology and neuroscience. The integration of these account of the insights that the other disciplines have to
disparate theoretical approaches and methodologies offer in understanding human decision-making. In the
offers exciting potential for the construction of more following, we first address how neuroscience can, and
accurate models of decision-making. already has, benefited from economics’ unitary perspec-
tive. We then discuss how economics can, and has begun
to, be enriched by taking account of cooperation and
competition between multiple specialized neural systems,
Introduction before closing with some thoughts on potentially fruitful
The question of how we make, and how we should make, future research directions.
judgments and decisions has occupied thinkers for many
centuries, with different disciplines approaching the
The view from economics: one unified theory
problem with characteristically different techniques. A
Economics contributes to the joint endeavor of neuroeco-
very recent approach, popularly known as neuroeco-
nomics by bringing its insights into the diverse outcomes
nomics, has sought to integrate ideas from the fields of
that can arise from the strategic and market interactions
psychology, neuroscience and economics in an effort to of multiple agents, and through a set of precise, formal,
specify more accurate models of choice and decision (for mathematical models to describe these interations and
reviews from the perspective of economics, see [1,2]). outcomes. However, the aspect of economics that may
How profitable the neuroeconomic approach will be is prove most useful to neuroscientists (and, indeed, that has
still unclear. Predictably, there are strong opinions on both already begun to bear fruit) is its embracing of a unified
sides. On the one hand, its strongest advocates (aided by theoretical framework for understanding human behavior
some exaggerated media reporting) have presented – namely the idea that behavior can be interpreted as
neuroeconomics as a new paradigm that will eventually choosing alternatives with the goal of maximizing utility.
replace the classical approaches. On the other, skeptics in The unitary perspective of economics can be seen in the
both communities have argued that economic models and assumptions that it makes about the two essential
neuroscientific techniques reflect disparate levels of dimensions of decision-making: choice (the evaluation of
analysis that have little to offer one another. Economists options and selection of actions), where economics
have, historically, been skeptical of the ability of ‘process assumes a consistent, stable set of preferences; and
measures’ to contribute to our understanding of economic judgment (information processing and probability esti-
and social behavior [3]; and neuroscientists commonly mation), with the assumption of a general reasoning
view economics as too abstract and removed from the system applicable to a wide range of problems. These
mechanisms of interest in the brain. assumptions have been criticized, as will be discussed
Corresponding author: Sanfey, A.G. ([email protected]). later, but the concept of decisions being made by
Available online 8 February 2006 comparing the utility signals for each of the decision
www.sciencedirect.com 1364-6613/$ - see front matter Q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tics.2006.01.009
 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006 109

alternatives has led to some real developments and has constructing theories about underlying neural function.
played an increasingly important role in guiding research Even as it is recognized that the brain (and consequent
investigating the underlying brain mechanisms. Addition- behavior) does not operate perfectly optimally, there are
ally, it is possible that well-established ideas from several reasons why these assumptions can nevertheless
economics will shed light on one of the least developed, be valuable. First, although complex forms of behavior
but most important, riddles for neuroscience: how the might not be optimal, simpler evolutionarily conserved
multiple, diverse and specialized neural systems of the mechanisms might prove to be closer to optimal, or at least
brain coordinate their activities to solve complex and often to have been so in the environment in which they evolved.
novel problems and give rise to coherent, goal-directed Second, an assumption of optimality can be a crucial step
behavior (see Box 1). in the development of formal theory, as it is often easiest to
define and precisely characterize the optimal behavior of a
system. Formal theory, in turn, enables the generation of
How economics can inform neuroscience: the benefits of precise, testable predictions about the system’s behavior.
a unitary perspective Finally, even when behavior (or neural function) turns out
In recent years, a growing number of neuroscientists have to be suboptimal, defining optimal performance can
turned to economic models as a framework both for provide a useful benchmark against which to compare
interpreting results and guiding new experimental work actual behavior. Identifying ways in which behavior
(e.g. [4,5]). This follows a small, but growing, tradition in systematically deviates from optimality can then generate
neuroscience in which optimal performance is defined for new insights into underlying mechanisms. The use of the
a given behavioral domain, and is then used for Expected Utility (EU) model [6] is one example of this
approach, and has been productively applied to research
Box 1. Brain systems and economics on the neural bases of reward and decision-making.
One potential area where economics can contribute is in under- According to the classical EU model, utility is computed
standing the dynamic processes by which the brain coordinates its as the product of the value and the probability of each
diverse systems to perform new, complex tasks. This problem has potential outcome (see Figure 1). Using this as a starting
received remarkably little attention in neuroscience research, point, research has sought to distinguish and identify the
although it is well-explored terrain in economics [76,77].
neural substrates for each of these components and to
There are striking parallels between the brain and a modern
corporation. Both can be viewed as complex systems transform- study their interaction. Although this is still a relatively
ing inputs into outputs. Both involve the interaction of multiple, new direction in neuroscience research, some useful
highly similar, agents (neurons are similar to one another, just as progress has been made.
are people), which, however, are specialized to perform particular
functions. Thus, in corporations, units often take the form of
departments that perform functions such as research, marketing,
Neural basis of value and reward
and so on. Similarly, the brain has systems specialized for Several decades of research have focused on the neural
different functions. As in a corporation, these functions may be bases of reward and punishment (the value function of the
more or less spatially segregated in the brain, depending upon EU model), identifying several systems that are consist-
the processing requirements of the specific functions and
ently responsive to reinforcement. However, we still have
their interactions.
Furthermore, there is hierarchical structure in both brains and a relatively weak understanding of the underlying neural
corporations. Both rely on ‘executive’ systems that make judgments computations and their influence on decision-making.
about the relative importance of tasks and decide how to mobilize Early studies of brain reward systems were concerned
specialized capabilities to perform those tasks. Several neuroima- primarily with establishing the generality of their
ging studies have demonstrated that brain areas involved
function and their neurochemical bases. These exper-
in executive function are actively engaged during the performance
of a novel and demanding task, and then show progressively less iments demonstrated that animals will withstand electric
activity as the task becomes more automatic [78,79]. These changes shock, exert significant physical effort, and even reduce
closely parallel the increase in activity in more specialized brain food intake to obtain electrical stimulation in appropriate
areas, as well as increases in performance speed. Similar improve- brain areas [7]. Furthermore, the reward value of
ments in speed and efficiency are observed in industry. Studies of
shipbuilding during World War II found that labor productivity rose
stimulation was shown to be reduced when dopamine
at an annual rate of 40 percent, and that speed of production receptor binding was inhibited, particularly in the ventral
increased even more rapidly. Although it often took as much as striatum [8].
300 days for a yard to deliver its first ship, by 1943 delivery times Inspired at least in part by an awareness of economic
were often less than one month and even as short as five days [80]. theory, more recent experiments have recognized the
Presumably such changes were accompanied by diminished
reliance on administrative and executive involvement, as is
importance of distinguishing between the magnitude and
observed in the brain. the probability of reward, and have tested directly for
It remains to be determined whether these similarities go deeper their separate influence on reward-related brain activity.
than mere analogy, and reflect fundamental principles of aggregate Single-cell recordings from dopamine neurons [9] and
behavior that apply across different levels of analysis. However,
neurons in the orbitofrontal cortex [10,11], striatum [12]
whether or not the principles of interaction are the same,
the theoretical methods that have been developed within and posterior cingulate cortex [13] have shown that neural
economics to analyze and construct models of interactions among responses scale reliably with reward magnitude. These
economic units, such as principal-agent theory, general equilibrium results have also been observed in human subjects:
theory, and the analysis of the firm, are likely to be useful in studying activity changes in many of these same brain areas scale
the interactions among brain systems that determine
directly with the magnitude of earned monetary reward
individual behavior.
[14–17], the appetitive value of food reward [18,19], as well
www.sciencedirect.com
110 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006

Rain
60
50%
Decision No rain
Yes 50%
30

Take Alternatives Beliefs Consequences

umbrella?

Rain
No 0
50%
50% No rain
100

TRENDS in Cognitive Sciences

Figure 1. Subjective Expected Utility theory posits two fundamental characteristics of an alternative that must be combined before reaching a decision, namely the value of
the alternative and the probability that this value will be attained. The value of each alternative is weighted by its attendant probability, and the option with the highest
probability-weighted values (or ‘utility’) is then chosen. An illustrative example is shown above. Imagine you are trying to decide whether or not to bring an umbrella as you
walk to work one morning. Utility theory posits that we must weigh the probability (in this case set at pZ0.5, perhaps gleaned from previous experience or from the morning
weather forecast), with the value of each of the four possible outcomes. These four outcomes are assigned a subjective rating (0Zworst outcome, 100Zbest outcome), which
of course varies depending on the individual decision-maker’s preferences, and a calculation is performed multiplying each outcome by its relevant probability, and then
combining across alternatives. Therefore, in this example, the utility for the option of taking the umbrella is [(0.5*60)C(0.5*30)]Z45, whereas the utility for not taking one’s
umbrella is [(0.5*100)C(0.5*0)]Z50. The utility maximizing decision-maker should therefore opt to leave the umbrella behind in these circumstances.

as with more abstract, social, rewards [20]. Furthermore, important finding from behavioral economics is that
brain stimulation experiments offer a precise mechanism people evaluate the outcome of decisions based on a
for investigating how reward magnitude might be flexible reference point. This predicts that the neural
encoded, suggesting that value is encoded as the integral systems responsible for utility assessment should be
of excitatory inputs to reward-related brain areas [21]. responsive to relative gains and losses rather than to
There is mounting evidence that the mesencephalic absolute levels of rewarding and punishing stimuli.
dopamine system plays a crucial role in value assessment Recent studies, using both fMRI [36] and scalp electrical
by signaling errors in reward prediction, which are used to recordings [37] have provided support for this prediction.
augment reward-producing behaviors both by generating
learning signals [22], and by adaptively updating goal Neural basis of probability estimation
states and attentional focus in working memory [23]. Probability has been less well studied than reward,
Similarly, it has been suggested that the norepinephrine although several recent neuroimaging studies have
(noradrenaline) system regulates the balance between begun to address this aspect. For example, one study has
exploitation (seeking to maximize utility from a given shown that activity in medial prefrontal cortex is inversely
source of reward) and exploration (seeking new sources) related to the probability of obtaining monetary reward
through its influence on mechanisms of learning and [38]. Other research has observed that activity in (human)
attention [24,25]. Although computational models have brain reward areas depends on uncertainty in the timing
been developed to describe functions of the dopamine and of reward delivery [39].
norepinephrine systems, their interaction remains rela-
tively unexplored. Neuroscientists could benefit from
Neural basis of the utility signal
interacting with economists who have been empirically Studies have also begun to examine the interaction
investigating similar issues, and are developing theo- between value and probability in the computation of
retical models that provide useful frameworks for inte- utility and the execution of decision-making behavior.
grating these different lines of neuroscientific research This work builds on the recent discovery of brain areas
(see [26] on learning from feedback and [27] on the trade- that appear to be directly related to decision-making,
off between exploitation and exploration). including the lateral intraparietal area (LIP) and regions
Although most neuroimaging work to date has focused of frontal cortex. Neurons in these areas closely track the
on positive rewards, recent findings also suggest that dynamics of decision-making in simple two alternative
there are complementary mechanisms for evaluating forced choice tasks: response-selective neurons exhibit a
negative utility. For example, there is evidence that the progressive increase in firing rate following stimulus
anterior cingulate cortex (ACC) responds to a variety of presentation, and the time at which they cross a threshold
signals that indicate negative utility (e.g. performance of activity predicts the timing of the behavioral response.
costs such as response conflict, errors, negative feedback This has been interpreted as evidence that these neurons
and pain [28–34]). Across these results, there is evidence integrate choice-relevant information and implement a
that increased activity in ACC correlates with the fundamental decision-making mechanism [40–43] that
magnitude of anticipated consequences, and current approximates the optimal algorithm [44–46]. These
efforts are beginning to test the extent to which these neurons are also directly sensitive to manipulations of
relationships conform to predictions of economic theory utility, exhibiting combined effects of value and prob-
(such as Prospect Theory [35]). For example, one ability in a manner predicted by the Expected Utility
www.sciencedirect.com
 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006 111

model [47]. Additionally, there have been human imaging challenge the standard economic model might be that
studies testing for activity that scales directly with these decisions do not emerge from a unitary process, but
expected utility, manipulating both reward magnitude rather from interactions between distinguishable sets
and probability [36,38]. This research has found that of processes.
activation in striatal reward areas correlates with
expected utility, demonstrating that utility-like measures Multiple processes in decision-making
are found outside motor preparation areas such as LIP. The most general distinction, and the most important for
The investigations described above reflect the influence neuroeconomics, is one that psychologists make between
that economic theory is already having on neuroscience automatic and controlled processes [49,50]. Automatic
studies. An independent metric for assessing reward value processes are fast and efficient, can often be carried out in
is crucial for a system that must often choose between parallel, but are highly specialized for domain-specific
rewards delivered in different modalities (e.g. do we take operations and therefore relatively inflexible. They are
extra vacation time or an additional paycheck?), and the thought to reflect the operation of highly over-trained
neural systems outlined above may well provide the basis (and, in some cases, possibly ‘hardwired’) mechanisms.
for such a signal. However, humans also have a capability for controlled
processing underlying our higher cognitive faculties.
The view from neuroscience: multiple systems Controlled processes are highly flexible, and thus able to
Psychology and neuroscience also bring much to the support a wide variety of goals, but are relatively slow to
neuroeconomics table, contributing a rich tradition of engage and rely on limited capacity mechanisms – that is,
empirical research and increasingly precise methods for they can support only a small number of pursuits at a
studying behavior and the neural mechanisms by which it time. Furthermore, the operations involved in controlled
is governed. Of particular relevance to economics is the processes (such as reasoning) are often accessible to
growing insight into mechanisms that are responsible for introspective, explicit description, whereas those involved
the assessment of utility and execution of decision-making in automatic processes (such as recognizing a face) are
behavior, as outlined in the previous section. However, usually much less so. An example commonly given of the
perhaps the single most important perspective that distinction between these processes is driving a stick-shift
neuroscience brings is to challenge the core assumption car: the novice is thought to rely on controlled processing,
in economics that behavior can be understood in terms of requiring focused concentration on a sequence of oper-
unitary evaluative and decision-making systems. ations that can be articulated but that are effortful and
easily disrupted by distraction. By contrast, the well-
How neuroscience can inform economics: the benefits of practiced driver, relying on automatic processes, can carry
a multiple-system approach out the same task efficiently while engaged in other
As noted in the first section of this article, economic theory activities (e.g. conversing), but might no longer be able to
assumes that people choose between alternative courses of articulate clearly the individual operations involved.
action based on a rational evaluation of the consequences, The distinction between controlled and automatic
and economists have developed detailed theoretical processing is probably best thought of as a continuum,
models for dealing with many decision situations: for rather than a qualitative dichotomy [51,52]. Nevertheless,
example, the Expected Utility model (described above) it has proven extremely useful in characterizing the
for decisions under risk, and the Discounted Utility model dynamics of behavior involving competing processes.
for decisions with consequences spread over time. These This distinction has appeared prominently in the field of
models have the virtue that they are formally explicit, decision research, where investigators refer to two types of
analytically tractable, and can be used to make quanti- evaluative systems: System 1 and System 2 [53,54].
tatively precise predictions about decision-making in a System 1 is automatic and heuristic-based; quickly
wide variety of circumstances. As such, they have proposing intuitive answers to problems as they arise.
provided a strong and unifying foundation for the System 2, which corresponds closely with controlled
development of theory about decision-making, with an processes, monitors the quality of the answer provided
assumption that decisions reflect the operation of a by System 1 and sometimes corrects or overrides
unitary all-purpose information processor. these judgments.
However, psychological research on judgment and
decision-making has produced a wealth of evidence Behavioral evidence for multiple systems
demonstrating that, in practice, these models do not The description of System 2 bears a close resemblance to
provide a satisfactory description of human behavior [48]. the rational, general-purpose processor presupposed by
There is a long legacy of research within psychology, standard economic theory. However, a bounty of exper-
strongly supported by findings from neuroscience, to imental findings suggest that controlled processing
suggest that human behavior is not the product of a single accounts for only part of our overall behavioral repertoire,
process, but rather reflects the interaction of different and in some circumstances can face stiff competition from
specialized subsystems. Although most of the time these domain-specific automatic processes.
systems interact synergistically to determine behavior, at There is now an extensive literature describing the
times they compete, producing different dispositions many ways in which human decision-making violates the
towards the same information. A major cause of these principles of rationality as defined by the Expected Utility
observed idiosyncrasies of behavior that have been used to (EU) [55] and Discounted Utility (DU) [56] models. Early
www.sciencedirect.com
112 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006

work in economics revealed situations (e.g. Ellsberg and Neural evidence for multiple systems
Allais paradoxes) whereby behavior violated key axioms of With the advent of neuroeconomics, discoveries about the
the EU model. More recently, the ‘Heuristics and Biases’ neural mechanisms involved in perception, attention,
approach in psychology has documented many instances learning and action selection have begun to drive the
of deviations from economic rationality [57]. For example, development of new, mechanistically explicit models of
most people are reluctant to play a gamble with 50% decision-making. In several instances, where these models
chance of winning $25 and 50% chance of losing $20, have been expressed formally, they have also begun to
despite the gamble’s overall positive expected value. This make contact with economic theory [62]. In some cases,
illustrates the phenomena of ‘loss aversion’, whereby this has provided validation of some of the basic principles
people often place disproportionate weight on losses of economic theory (see previous section), whereas in
relative to gains of similar absolute value [58]. others it has begun to provide insight into how and why
When it comes to choice over time, there is also ample human behavior deviates from optimality as defined by
evidence of violations of the DU model. Perhaps most economic models. Psychological research, buttressed by
importantly, there is strong evidence that discounting is recent neuroscientific findings, has begun to identify
separable systems that contribute to decision-making
much steeper for short time delays than for longer delays, a
and behavior, systems that for the most part work
phenomenon known as ‘hyperbolic time discounting’ [59].
cooperatively, but sometimes compete. Recent neuroscien-
For example, offered a choice between $10 today and $11 in a
tific research has begun to characterize the engagement
week, many people are likely to choose the immediate $10.
and disposition of these neural systems under a variety of
However, offered the choice between $10 in a year and $11 in
conditions in which behavior seems to deviate from the
a year and a week, most people would chose the $11, now
expectations of economic theory.
considering the extra week of wait inconsequential. From the
economist’s perspective, however, this implies a reversal of
preference (i.e. whether an extra dollar is worth a week’s wait Systems for emotion and deliberation. Perhaps the
or not), and therefore does not conform to the rational distinction with the greatest immediate ramifications for
model [56]. economic theories is between systems supporting emotion
Responding to the documented limitations of EU and DU, and those supporting deliberation, which closely parallels
a variety of alternative models of decision-making under the distinction between automatic and controlled processes.
uncertainty and intertemporal choice have been constructed The nature of emotions has been the subject of intense
(including hyperbolic time discounting and Prospect The- inquiry and debate for entire fields of science, a full consi-
ory). These models have made notable progress in describing deration of which is well beyond the scope of this article. For
real-world decision-making [60], and even in designing present purposes, we will use ‘emotion’ to refer to low-level
policy interventions to change behavior [61]. psychological processes engaged by events that elicit strong
Additionally, economists have begun to construct valenced and stereotyped behavioral responses (e.g. fear is a
models that play on the automatic vs. controlled distinc- response to threatening stimuli that leads to freezing or
tion. For example, Benhabib and Bisin (J. Benhabib and withdrawal). Accordingly, emotions are rapid, highly auto-
A. Bisin, unpublished) propose a model of decision-making matic responses to specific stimuli or events, well adapted to
in which automatic processes are initially allowed to some circumstances but not to others. These contrast with
determine behavior, but controlled processes are activated the capacity for controlled processing discussed above – the
whenever the costs from letting the automatic processes ability to respond flexibly to circumstances, to deliberate
carry on become too large. They apply this framework to a rationally about the long-term consequences of our behavior,
dynamic saving-consumption model, and describe how its and to plan behavior accordingly.
predictions differ from traditional economic theory. The Although emotional processes, like other automatic
key insight here is that, rather than thinking about processes, share common neural substrates with controlled
human behavior as being governed by a unitary, general processes, it is becoming increasingly clear from neuroima-
purpose mechanism, it can often be better described in ging studies that these different types of processes do involve
terms of the interaction – and sometimes competition – some distinguishable neural components. There is a general
between different subsystems that might favor different consensus that high-level, deliberative processes, such as
problem-solving and planning, consistently engage anterior
alternatives for a given decision.
and dorsolateral regions of prefrontal cortex as well as areas
However, although these new models of decision-
of posterior parietal cortex [63–66]. By contrast, automatic
making have made great strides towards providing
processes appear to rely heavily on more posterior cortical
descriptively realistic models of behavior, as yet they
structures, as well as subcortical systems. Emotional
have not provided insight into the actual mechanisms
processes, in particular, seem reliably to engage a set of
responsible for the deviations from the normative models.
structures classically referred to as the limbic system, which
Recently, neuroimaging studies have begun to identify the includes brainstem reward-processing structures (such as
contribution that different systems make to decision- the ventral tegmental area), areas of the midbrain and cortex
making, in particular when this involves competition to which they project (such as the nucleus accumbens and
between controlled processing and emotional responses – ventromedial frontal, orbitofrontal, and anterior cingulate
a special case of automatic processes that we consider in cortex), and several other areas such as the amygdala and
the next section. insular cortex [67].
www.sciencedirect.com
 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006 113

Strategic interactions. The neural mechanisms respon- ventral striatum related to exhibitions of trust in an
sible for deliberation and emotion are clearly closely inter- investment game [71].
related (e.g. [23,24,68]). Nevertheless, distinguishing
between these could help shed light on many of the most
basic patterns uncovered by behavioral economics, such as Decisions over time. Hyperbolic time discounting pro-
nonlinear probability weighting and loss aversion, as well vides another illustration of the benefits of investigating
as several other behaviors that challenge the standard emotion–deliberation interactions. One hypothesis that
economic model, including many that involve market and has been advanced to explain this phenomenon is that it
non-market interactions between individuals. An example reflects the operation of two fundamentally different
of the latter is behavior in a well-studied decision task mechanisms, one affective, which heavily values the
known as the Ultimatum Game. In this game, participants present and steeply discounts all future opportunities,
must decide either to accept or reject an offer of money and the other deliberative, which discounts options more
made by a partner, whose task is to divide a sum of money consistently across time [72,73]. However, it has been
between the two. The twist is that if the offer is rejected, difficult, from behavioral data alone, to provide evidence
then neither player receives anything. Whereas standard for separate mechanisms, or to motivate them from purely
economic models would predict that all non-zero offers theoretical considerations. Recently, neuroimaging has
should be accepted, on the premise that some money is produced evidence for this distinction [74], showing that
better than none, in fact about half of the unfair offers are choices involving the option of an immediate reward
typically rejected. This is the case even when the game actively engage the ventral striatum, as well as medial
involves single (one-shot) confidential interactions, in and orbitofrontal areas – areas rich in dopaminergic
which bargaining or the establishment of a reputation innervation and, as noted above, consistently associated
cannot explain the findings. A neuroimaging study with the evaluation of reward. These are areas of the brain
examining the Ultimatum Game [69] found two brain that appear to have been highly conserved across
regions that were particularly active when the participant evolution, and their steep form of discounting might
was confronted with an unfair offer, the anterior insula reflect adaptations to the computation of utility that
and the dorsolateral prefrontal cortex (dlPFC) (Figure 2). occurred in the absence of neural structures capable of
Activation in these areas has been shown to correlate with sophisticated planning for the future. By contrast, areas of
emotional and deliberative processing, respectively, and it frontal and parietal cortex commonly associated with
was found that if the insular activation was greater than more abstract forms of reasoning and planning were
the dlPFC activation, participants tended to reject the consistently involved, independently of when the reward
offer, whereas if the dlPFC activation was greater, they became available (Figure 3). Furthermore, for choices
tended to accept the offer. This offers neural evidence for a pitting an immediate reward against a greater but later
two-system account of decision-making in this task. reward, the relative activity in these different areas was
The conclusion that strategic interactions between closely correlated with the choice made, with greater
individuals involves an interplay between emotion and fronto-parietal activity associated with selection of the
deliberation is underlined by research that has observed deferred but more valuable option, suggesting that this
activation in the caudate nucleus related to satisfaction system appears to value rewards in the future in a more
derived from punishing norm violations [70], and in the even-handed fashion.

(a) (b) 1.2

Anterior insula ACC
Accept
1.0
Reject
dlPFC
Contrast values

0.8

0.6

0.4

0.2

0
R. anterior insula R. dlPFC

TRENDS in Cognitive Sciences

Figure 2. Deciding whether to accept or reject an unfair offer in the Ultimatum Game leads to conflict between emotional, ‘reject’, and cognitive, ‘accept’, systems. (a)
Activation related to the presentation of an unfair offer from another human in the Ultimatum Game, showing activation of bilateral anterior insula and anterior cingulate
cortex (ACC, right), and activation of right dorsolateral prefrontal cortex (dlPFC, left). Areas in orange showed greater activation following unfair as compared with fair offers
(P!0.001). (b) Right anterior insula and right dlPFC activation for all unfair offer trials, categorized by subsequent acceptance or rejection. Trials in which the offer was
subsequently rejected had significantly higher anterior insula activation than trials where the offer was subsequently accepted. Adapted from [69].

www.sciencedirect.com
114 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006

(a) (c)
PCC mPFC V.Str.

0.05

Normalized signal change

δ areas

x = 4 mm y = 8 mm 0.0

β areas
(b)
10
R.Par. dlPFC
–0.05

T13
Choose Choose
early late

lOFC 0
x = 44 mm
TRENDS in Cognitive Sciences

Figure 3. Beta and delta systems in intertemporal choice. When people choose between two rewards separated in time, two different brain systems are involved. (a) The first,
termed the b system, responds preferentially to the presence of reward available in the immediate future. The b system comprises brain areas associated with the midbrain
dopamine system and relatively automatic reward processing, including the ventral striatum (V.Str.) and medial prefrontal cortex (mPFC). (b) The second system, the d
system, comprises brain areas implicated in cognition, including the dorsolateral prefrontal cortex (dlPFC) and right posterior parietal cortex (R. Par.). (c) When intertemporal
choices involve deciding between an immediate and delayed reward, choice tends to reflect the relative activity in the b and d systems. When people select the later, larger
reward, activity tends to predominate in the d system. Adapted from [74].

Multiple systems; multiple interactions endeavor would be to arrive at a common language, and
The above findings provide an initial toehold for measur- perhaps a common set of ‘decision tasks’, to ensure that the
ing physical mechanisms responsible for utility evaluation hitherto heartening level of collaboration across these
and decision-making in the brain. Such studies offer the diverse fields continues in a productive fashion. With
promise that we will be able to identify and precisely increasing collaboration among researchers from across
characterize these mechanisms, and the factors that the fields of interest, it also seems certain that other
influence their engagement and interaction. Even at this interesting avenues of research will open up in the near
early stage, however, results highlight the fact that future (see Box 2 for some potential questions of interest).
decision-making involves the interaction among multiple
subsystems governed by different parameters and possibly
even different principles. As is the case for the distinction Box 2. Questions for future research
between automatic and controlled processes, economists
† How (and where) are value and probability combined in the brain to
have already begun to propose dual-process models that provide a utility signal? What are the dynamics of this computation?
incorporate a distinction similar to that between emotion † What neural systems track classically defined forms of expected and
and deliberation ([72,75]; Fudenberg, D. and Levine, discounted utility? Under what conditions do these computations
D. unpublished). These models have mainly been applied break down?
† To what extent does tracking utility computations generalize to
to intertemporal choice, but in one case (Loewenstein, G.
more complex decision tasks?
and O’Donoghue, T. see http://papers.ssrn.com/sol3/ † How is the utility metric different from neural signals of attention
papers.cfm?abstract_idZ539843) have also been applied and motor preparation?
to decision-making under uncertainty and to interperso- † How is negative utility signaled? Is there a negative utility
nal interactions of the type examined in the prediction signal comparable to the one for positive utility?
† How are rewards of different types mapped onto a common neural
Ultimatum Game.
currency like utility?
† How do systems that seem to be focused on immediate decisions
and actions interact with systems involved in longer-term planning
The future (e.g. making a career decision)? For example, does an unmet need
This article has reviewed two general ways in which the generate a tonic and progressively increasing signal (i.e. a mounting
neuroeconomic endeavor can make important contributions ‘drive’), or does it manifest as a recurring episodic/phasic signal with
to research on decision-making – firstly, the incorporation increasing amplitude?
† Under what circumstances do these various systems cooperate or
into neuroscience and psychology of the formal, rigorous
compete? When there is competition, how and where is it adjudicated?
economic modeling approach, and secondly, the awareness † Psychologists, neuroscientists and behavioral economists all
within the economic community of the evidence for multiple- seem to agree that various automatic forms of behavior (including
systems involved in decision-making. One current challenge emotional responses) reflect the operation of a multiplicity of
is to ensure that researchers are communicating produc- mechanisms. However, do higher-level deliberative processes rely
similarly on multiple mechanisms, or a single, more tightly
tively; often, terms such as ‘choice’, ‘judgment’ and ‘decision’ integrated (unitary) set of mechanisms?
are used in different ways by different fields. A useful
www.sciencedirect.com
 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006 115

Neuroeconomics is still far from providing a full set of 26 Camerer, C.F. and Ho, T. (1999) Experience-Weighted Attraction
answers to these questions, but it seems reasonable to (EWA) learning in normal-form games. Econometrica 67, 827–874
27 Merlo, A. and Schotter, A. (1999) A surprise-quiz view of learning in
expect that, as it converges on the mechanisms of utility economic experiments. Games Econ. Behav. 28, 25–54
evaluation and of decision-making, and characterizes these 28 Falkenstein, M. et al. (1995) Event related potential correlates of
in terms of formal models that can inform economic theory, errors in reaction tasks. In Perspectives of Event-Related Potentials
all of the fields involved stand to gain. Research (Karmos, G. et al., eds), pp. 287–296, Elsevier
29 Carter, C.S. et al. (1998) Anterior cingulate cortex, error detection, and
the online monitoring of performance. Science 280, 747–749
References 30 Holroyd, C.B. and Coles, M.G. (2002) The neural basis of human error
1 Glimcher, P.W. and Rustichini, A. (2004) Neuroeconomics: the processing: reinforcement learning, dopamine, and the error-related
consilience of brain and decision. Science 306, 447–452 negativity. Psychol. Rev. 109, 679–709
2 Camerer, C. et al. (2005) Neuroeconomics: how neuroscience can 31 Miltner, W.H. et al. (2003) Implementation of error-processing in the
inform economics. J. Econ. Lit. 43, 9–64 human anterior cingulate cortex: a source analysis of the magnetic
3 Bruni, L. and Sugden, R. The road not taken: two debates about the equivalent of the error-related negativity. Biol. Psychol. 64, 157–166
role of psychology in economics. Econ. J. (in press) 32 Peyron, R. et al. (2000) Functional imaging of brain responses to pain.
4 Shizgal, P. (1997) Neural basis of utility estimation. Curr. Opin. A review and meta-analysis. Clin. Neurophysiol. 30, 263–288
Neurobiol. 7, 198–208 33 Gehring, W.J. and Willoughby, A.R. (2002) The medial frontal cortex
5 Glimcher, P.W. (2003) Decisions, Uncertainty, and the Brain: The
and the rapid processing of monetary gains and losses. Science 295,
Science of Neuroeconomics, MIT Press
2279–2282
6 von Neumann, J. and Morgenstern, O. (1947) Theory of Games and
34 Yeung, N. and Sanfey, A.G. (2004) Independent coding of reward
Economic Behavior, Princeton University Press
magnitude and valence in the human brain. J. Neurosci. 24,
7 Olds, J. (1977) Drives and Reinforcements: Behavioral Studies of
6258–6264
Hypothalamic Function, Raven Press
35 Kahneman, D. and Tversky, A. (1979) Prospect theory: an analysis of
8 Colle, L.M. and Wise, R.A. (1988) Effects of nucleus accumbens
decisions under risk. Econometrica 47, 262–291
amphetamine on lateral hypothalamic brain stimulation reward.
36 Breiter, H.C. et al. (2001) Functional imaging of neural responses to
Brain Res. 459, 361–368
expectancy and experience of monetary gains and losses. Neuron 30,
9 Tobler, P.N. et al. (2005) Adaptive coding of reward value by dopamine
619–639
neurons. Science 307, 1642–1645
37 Holroyd, C.B. et al. (2004) Context dependence of the event-related
10 Tremblay, L. and Schultz, W. (1999) Relative reward preference in
brain potential associated with reward and punishment. Psychophy-
primate orbitofrontal cortex. Nature 398, 704–708
siology 41, 245–253
11 Roesch, M.R. and Olson, C.R. (2004) Neuronal activity related to
38 Knutson, B. et al. (2005) Distributed neural representation of expected
reward value and motivation in primate frontal cortex. Science 304,
value. J. Neurosci. 25, 4806–4812
307–310
39 Berns, G.S. et al. (2001) Predictability modulates human response to
12 Cromwell, H.C. and Schultz, W. (2003) Effects of expectations for
reward. J. Neurosci. 21, 2793–2798
different reward magnitudes on neural activity in primate striatum.
40 Schall, J.D. (2001) Neural basis of deciding, choosing and acting. Nat.
J. Neurophysiol. 89, 2823–2838
Rev. Neurosci. 2, 33–42
13 McCoy, A.N. et al. (2003) Saccade reward signals in posterior cingulate
41 Shadlen, M.N. and Newsome, W.T. (2001) Neural basis of a perceptual
cortex. Neuron 40, 1031–1040
14 Knutson, B. et al. (2001) Dissociation of reward anticipation and decision in the parietal cortex (area LIP) of the rhesus monkey.
outcome with event related fMRI. Neuroreport 12, 3683–3687 J. Neurophysiol. 86, 1916–1936
15 O’Doherty, J. et al. (2001) Abstract reward and punishment 42 Roitman, J.D. and Shadlen, M.N. (2002) Response of neurons in the
representations in the human orbitofrontal cortex. Nat. Neurosci. 4, lateral intraparietal area during a combined visual discrimination
95–102 reaction time task. J. Neurosci. 22, 9475–9489
16 Delgado, M.R. et al. (2003) Dorsal striatum responses to reward and 43 Sugrue, L.P. et al. (2004) Matching behavior and the representation of
punishment: effects of valence and magnitude manipulations. Cogn. value in the parietal cortex. Science 304, 1782–1787
Affect. Behav. Neurosci. 3, 27–38 44 Usher, M. and McClelland, J.L. (2001) The time course of perceptual
17 Elliott, R. et al. (2003) Differential response patterns in the striatum choice: the leaky, competing accumulator model. Psychol. Rev. 108,
and orbitofrontal cortex to financial reward in humans: a parametric 550–592
functional magnetic resonance imaging study. J. Neurosci. 23, 45 Gold, J.I. and Shadlen, M.N. (2001) Neural computations that
303–307 underlie decisions about sensory stimuli. Trends Cogn. Sci. 5, 10–16
18 McClure, S.M. et al. (2004) Neural correlates of behavioral preference 46 Brown, E.T. et al. (2005) Simple neural networks that optimize
for culturally familiar drinks. Neuron 44, 379–387 decisions. Int. J. Bifurcat. Chaos 15, 803–826
19 Kringelbach, M.L. et al. (2003) Activation of human orbitofrontal 47 Platt, M.L. and Glimcher, P.W. (1999) Neural correlates of decision
cortex to a liquid food stimulus is correlated with its subjective variables in parietal cortex. Nature 400, 233–238
pleasantness. Cereb. Cortex 13, 1064–1071 48 Tversky, A. and Kahneman, D. (1974) Judgment under uncertainty:
20 Rilling, J.K. et al. (2004) Opposing BOLD responses to reciprocated heuristics and biases. Science 185, 1124–1131
and unreciprocated altruism in putative reward pathways. Neurore- 49 Posner, M. and Snyder, C. (1975) Facilitation and inhibition in the
port 15, 2539–2543 processing of signals. In Attention and Performance V (Rabbitt, P.M.A.
21 Mark, T.A. and Gallistel, C.R. (1993) Subjective reward magnitude of and Dornic, S., eds), pp. 669–682, Academic Press
medial forebrain stimulation as a function of train duration and pulse 50 Schneider, W. and Shiffrin, R.M. (1977) Controlled and automatic
frequency. Behav. Neursci. 107, 389–401 human information processing: I. Detection, search, and attention.
22 Montague, P.R. et al. (1996) A framework for mesencephalic dopamine Psych Rev 84, 1–66
systems based on predictive Hebbian learning. J. Neurosci. 16, 51 Kahneman, D. and Treisman, A. (1984) Changing views of attention
1936–1947 and automaticity. In Varieties of Attention (Parasuraman, R. and
23 Braver, T.S. and Cohen, J.D. (2000) On the control of control: the role Davies, D.R., eds), pp. 29–61, Academic Press
of dopamine in regulating prefrontal function and working memory. In 52 Cohen, J.D. et al. (1990) On the control of automatic processes: a
Attention and Performance (Monsell, S. and Driver, J., eds), pp. parallel distributed processing account of the Stroop effect. Psychol.
713–737, Academic Press Rev. 97, 332–361
24 Aston-Jones, G. and Cohen, J.D. (2005) An integrative theory of locus 53 Sloman, S.A. (2002) Two systems of reasoning. In Heuristics and
coeruleus-norepinephrine function: adaptive gain and optimal per- Biases: The Psychology of Intuitive Judgment (Gilovich, T. and
formance. Annu. Rev. Neurosci. 28, 403–450 Griffin, D., eds), pp. 379–396, Cambridge University Press
25 Yu, A.J. and Dayan, P. (2005) Uncertainty, neuromodulation, and 54 Kahneman, D. (2003) A perspective on judgment and choice: mapping
attention. Neuron 46, 681–692 bounded rationality. Am. Psychol. 58, 697–720
www.sciencedirect.com
116 Review TRENDS in Cognitive Sciences Vol.10 No.3 March 2006

55 Starmer, C. (2000) Developments in non-expected utility theory: the 68 Ochsner, K.N. and Gross, J.J. (2005) The cognitive control of emotion.
hunt for a descriptive theory of choice under risk. J. Econ. Lit. 38, Trends Cogn. Sci. 9, 242–249
332–382 69 Sanfey, A.G. et al. (2003) The neural basis of economic decision-
56 Frederick, S. et al. (2002) Time discounting and time preference: a making in the Ultimatum Game. Science 300, 1755–1758
critical review. J. Econ. Lit. 40, 351–401 70 de Quervain, D.J.F. et al. (2004) The neural basis of altruistic
57 Kahneman, D. et al. (1982) Judgment under Uncertainty: Heuristics punishment. Science 305, 1254–1258
and Biases, Cambridge University Press 71 King-Casas, B. et al. (2005) Getting to know you: reputation and trust
58 Kahneman, D. et al. (1991) The endowment effect, loss aversion, and in a two-person economic exchange. Science 308, 78–83
the status quo bias: anomalies. J Econ Perspec 5, 193–206 72 Shefrin, H.M. and Thaler, R.H. (1988) The behavioral life-cycle
59 Ainslie, G. and Haslam, N. (1992) Hyperbolic discounting. In Choice over hypothesis. Econ. Inq. 26, 609–643
Time (Loewenstein, G. and Elster, J., eds), pp. 57–92, Russell Sage 73 Loewenstein, G. (1996) Out of control: visceral influences on behavior.
60 Camerer, C.F. (2000) Prospect theory in the wild: evidence from the
Org Behav Hum Dec Proc 65, 272–292
field. In Choices, Values, and Frames (Kahneman, D. and Tversky, A.,
74 McClure, S.M. et al. (2004) Separate neural systems value immediate
eds), pp. 288–300, Cambridge University Press
and delayed monetary rewards. Science 306, 503–507
61 Benartzi, S. and Thaler, R.H. (2004) Save more tomorrow: using
75 Bernheim, D. and Rangel, A. (2004) Addiction and cue-triggered
behavioral economics to increase employee saving. J. Polit. Econ. 112,
decision processes. Am. Econ. Rev. 94, 1558–1590
S164–S187
76 Williamson, O. (1967) Hierarchical control and optimum firm size.
62 Montague, P.R. and Berns, G.S. (2002) Neural economics and the
J. Polit. Econ. 76, 123–138
biological substrates of valuation. Neuron 36, 265–284
63 Duncan, J. et al. (1996) Intelligence and the frontal lobe: the 77 Williamson, O. (1979) Transaction cost economics: the governance of
organization of goal-directed behavior. Cognit. Psychol. 30, 257–303 contractual relations. J. Law Econ. 22, 233–261
64 Smith, E.E. and Jonides, J. (1999) Storage and executive processes in 78 Haier, R.J. et al. (1992) Regional glucose metabolic changes after
the frontal lobes. Science 283, 1657–1661 learning a complex visuospatial/motor task: a PET study. Brain Res.
65 Miller & Cohen. (2001) An integrative theory of prefrontal cortex 570, 134–143
function. Annu. Rev. Neurosci. 24, 167–202 79 Karni, A. et al. (1998) The acquisition of skilled motor performance:
66 Stuss, D.T. and Knight, R.T. (2002) Principles of Frontal Lobe fast and slow experience-driven changes in primary motor cortex.
Function, Oxford University Press Proc. Natl. Acad. Sci. U. S. A. 95, 861–868
67 Dalgleish, T. (2004) The emotional brain. Nat. Rev. Neurosci. 5, 80 Thompson, P. (2001) How much did the Liberty Shipbuilders learn?
583–589 New evidence for an old case study. J. Polit. Econ. 109, 103–137

Symposium "Of Minds and Language: An encounter with Noam Chomsky"

June 2006

On the occasion of the 25th Anniversary of the Summer Courses of the Universidad del Paı́s Vasco (Euskal herriko Unibertsitatea)
http://www.sc.ehu.es/scrwwwsu/index.htm, Massimo Piattelli-Palmarini (University of Arizona), Pello Salaburu (University of the
Basque Country) and Juan Uriagereka (University of Maryland and University of the Basque Country) are organizing the
symposium "Of Minds and Language: An encounter with Noam Chomsky", to be held in the Palacio de Miramar, San Sebastian,
Spain, 19–23 June 2006.

Confirmed participants: Noam Chomsky (MIT), Thomas G. Bever (University of Arizona), Cedric Boeckx (Harvard University),
Christopher Cherniak (University of Maryland), Janet D. Fodor (CUNY), Jerry A. Fodor (Rutgers University), Angela D. Friederici
(Max Planck Institute, Leipzig), Randy C. Gallistel (Rutgers University), Rochel Gelman (Rutgers University), Lila Gleitman (UPenn),
Morris Halle (MIT), Marc D. Hauser (Harvard University), James T. Higginbotham (University of Southern California), Wolfram
Hinzen (University of Amsterdam), Itziar Laka (University of the Basque Country), Jacques Mehler (Institute of Advanced Studies,
Trieste), Luigi Rizzi (University of Siena), Nuria Sebastian Galles (University of Barcelona), Donata Vercelli (University of Arizona)

www.sciencedirect.com